Clemens M. Franz

Two‐Component Polymer Scaffolds for Controlled Three‐Dimensional Cell Culture

O N Cell behavior is governed by interactions with the cellular environment. [ 1–3 ] These interactions include cell–cell as well as cell–extracellular matrix (ECM) contacts and act, in addition to soluble growth factors, as key regulators of cell survival, proliferation, and differentiation. However, not only the molecular composition of the contact sites, but also their spatial distributions impact cell behavior. [ 4 , 5 ] Realizing cell-culture scaffolds that mirror the complex in vivo arrangement of ECM components is an active area of biomaterials engineering. Corresponding 2D lithographically defi ned micropatterned structures are widely used and have even become commercially available. [ 6–9 ] In addition to patterned ligand distributions, mechanical interactions between cells and their environment also play an important role in regulating cellular functions. [ 10 ] Consequently, corresponding fl exible substrates were developed that allow measurements of cellular forces in 2D on planar substrates or pillar arrays. [ 11 , 12 ]

Elastic Fully Three‐dimensional Microstructure Scaffolds for Cell Force Measurements

2010 WILEY-VCH Verlag Gm Mechanical interactions between cells and their environment play an important role in regulating many cellular functions, such as migration, differentiation, and proliferation. They are also involved in complex processes like tissue formation during development of multicellular organisms. During the last two decades, several methods have been developed to visualize and measure traction forces produced by single cells. Initial experiments were performed by growing fibroblasts on a thin silicon rubber substrate that was deformed by forces exerted by migrating cells. Subsequently, elastic polyacrylamide substrates embedded with fluorescent latex beads were used to get an insight into the force vectors exerted by the cells. These methods were further improved by producing regular arrays of traceable markers in the elastic substrates allowing a quantification of traction forces exerted by adhering cells. A different approach has used arrays of elastic posts positioned on a flat 2D substrate, coated with extracellular-matrix (ECM)molecules. Because each post effectively acts like a harmonic spring, local traction forces at multiple cell adhesion sites could then be measured by simply monitoring post deflections. A modification of this assay was recently applied to measure contraction forces of single cardiomyocytes. Although these different approaches have provided valuable insight into our understanding of cellular forces, they have the disadvantage that the cells are forced into a 2D growth pattern, which significantly differs from the in vivo situation. Therefore, the development of in vitro model systems that capture more of the complexity present in 3D tissue scaffolds are highly desirable. Technologies for fabricating macroscopic tissue architectures in the millimeter and micrometer range have become available only during recent years. In contrast, methods to reliably produce flexible 3D cellular environments in the micrometer to nanometer range are still missing. Ideally, these scaffolds should have a controllable distribution of cell-substrate contact sites and an adjustable stiffness. In this Communication, we show that such biocompatible scaffolds can be fabricated by means of direct laser writing (DLW) and subsequent surface functionalization. We furthermore demonstrate that these scaffolds can be rhythmically deformed by single beating cardiomyocytes. In addition, we quantitatively evaluate the involved contraction forces by calibrating these structures using atomic force microscopy (AFM). In particular, with the method presented here, beamlike structures with defined geometries and physiologically relevant stiffness values can be produced; thus, these scaffolds structurally and mechanically mimic large macromolecules surrounding cells in 3D tissues, such as collagen fibrils. In essence, DLW (see the Experimental section) can be thought of as a pencil of light in three dimensions. Any connected structure envisioned on a computer can easily be implemented by scanning a photoresist relative to the focus of a femtosecond-laser beam (Fig. 1a). After development of a negative resist, sufficiently exposed regions correspond to the remaining polymer (Fig. 1b). Lithography of 3D polymeric templates by DLW with lateral feature sizes down to the 100 nm range has become routine in the field of nanophotonics. Along these lines, fully 3D (Fig. 1c) scaffolds can be fabricated. Sample footprints of mm up to cm dimensions can be realized by combining 300mm 300mm areas that are accessible within the piezo scanning range. The height of the structures is limited only by the working distance of the microscope lens (80mm in our case). Typical writing times for the structures shown in this Communication are less than 10min. Here, we employ the commercially available resist system Ormocomp (obtained from Micro Resist Technology GmbH). Ormocomp is a member of Ormocer, a class of inorganic (Si O Si)–organic hybrid polymer systems, previously developed at the Fraunhofer Institute for Silica Research (Wurzburg, Germany). Its detailed chemical composition is proprietary. With Ormocomp as a photoresist, complex fully 3D structures (Fig. 1a–c) can be easily manufactured and, after functionalization with fibronectin, directly used for cell culture (Fig. 1d). To demonstrate that single cells can lead to noticeable deformation of the 3D Ormocomp scaffolds, primary cardiomyocytes are isolated from chicken embryos (see the

Analyzing focal adhesion structure by atomic force microscopy.

Atomic force microscopy (AFM) can produce high-resolution topographic images of biological samples in physiologically relevant environments and is therefore well suited for the imaging of cellular surfaces. In this work we have investigated focal adhesion complexes by combined fluorescence microscopy and AFM. To generate high-resolution AFM topographs of focal adhesions, REF52 (rat embryo fibroblast) cells expressing YFP-paxillin as a marker for focal adhesions were de-roofed and paxillin-positive focal adhesions subsequently imaged by AFM. The improved resolution of the AFM topographs complemented the optical images and offered ultrastructural insight into the architecture of focal adhesions. Focal adhesions had a corrugated dorsal surface formed by microfilament bundles spaced 127±50 nm (mean±s.d.) apart and protruding 118±26 nm over the substratum. Within focal adhesions microfilaments were sometimes branched and arranged in horizontal layers separated by 10 to 20 nm. From the AFM topographs focal adhesion volumes could be estimated and were found to range from 0.05 to 0.50 μm3. Furthermore, the AFM topographs show that focal adhesion height increases towards the stress-fiber-associated end at an angle of about 3°. Finally, by correlating AFM height information with fluorescence intensities of YFP-paxillin and F-actin staining, we show that the localization of paxillin is restricted to the ventral half of focal adhesions, whereas F-actin-containing microfilaments reside predominantly in the membrane-distal half.

Studying integrin-mediated cell adhesion at the single-molecule level using AFM force spectroscopy.

The establishment of cell adhesion involves specific recognition events between individual cell-surface receptors and molecules of the cellular environment. However, characterizing single-molecule adhesion events in the context of a living cell presents an experimental challenge. The atomic force microscope (AFM) operated in force spectroscopy mode provides an ultrasensitive method to investigate cell adhesion forces at the level of single receptor-ligand bonds. With a living cell attached to the AFM cantilever, the number of cell-substrate interactions can be controlled and limited to the formation of single receptor-ligand bonds. From force-distance (F-D) curves recorded during cell detachment, the strength of single receptor-ligand bonds can be determined. Furthermore, by varying the rate of force application during bond rupture, a dynamic force spectrum (DFS) can be generated from which additional parameters that describe the energy landscape of the interaction, such as dissociation rate and energy barrier width, can be obtained. Using the example of α2β1 integrin–mediated adhesion to type I collagen, we provide a detailed description of how dynamic AFM single-cell force spectroscopy (SCFS) adhesion measurements can be performed with single-molecule sensitivity, and how specific energy landscape parameters of the integrin-collagen bond can be extracted from the DFS.

Lumican inhibits cell migration through α2β1 integrin.

Lumican, an extracellular matrix protein of the small leucine-rich proteoglycan family, has been shown to impede melanoma progression by inhibiting cell migration. In the present study, we show that lumican targets α2β1 integrin thereby inhibiting cell migration. A375 melanoma cells were transfected with siRNA directed against the α2 integrin subunit. Compared to A375 control cells, the anti-migratory effect of lumican was abrogated on transfected A375 cells. Moreover, lumican inhibited the chemotactic migration of Chinese hamster ovary (CHO) cells stably transfected with α2 integrin subunit (CHO-A2) but not that of wild-type CHO cells (CHO-WT) lacking this subunit. In contrast to CHO-WT cells, we observed in time-lapse microscopy a decrease of CHO-A2 cell migration speed in presence of lumican. Focal adhesion kinase phosphorylated at tyrosine-397 (pFAK) and total FAK were analysed in CHO-WT and CHO-A2 cells. A significant decrease of the ratio pFAK/FAK was shown in presence of recombinant human lumican. Using solid phase assays, a direct binding between lumican and the α2β1 integrin was demonstrated. This interaction did not involve the glycan moiety of lumican and was cation independent. Lumican was also able to bind the activated I domain of the α2 integrin subunit with a K(d)≥200nM. In conclusion, we demonstrated for the first time that the inhibition of cell migration by lumican depends on a direct binding between the core protein of lumican and the α2β1 integrin.

Directing nuclear deformation on micropillared surfaces by substrate geometry and cytoskeleton organization

We have recently demonstrated strong nuclear deformation of SaOs-2 osteosarcoma cells on poly-L-lactic acid (PLLA) micropillar substrates. In the present study, we first demonstrated that chemical and mechanical properties of the micropillar substrates have no dominant effect on deformation. However, SaOs-2 nucleus deformation could be strongly modulated by varying the pillar size and spacing, highlighting the importance of geometric constraints for shaping the nucleus. Furthermore, comparing the capacity for nuclear deformation in three different osteosarcoma cell lines (SaOs-2, MG-63 and OHS-4) revealed strong cell-type specific differences. Surprisingly, the highly-deformable SaOs-2 cell line displayed the highest cell stiffness as assessed by AFM-based colloidal force spectroscopy and featured a more prominent array of actin fibres above the nucleus, suggesting a link between actin-mediated cell stiffness and cell nucleus deformation. In contrast, in MG-63 and OHS-4 cells dense microtubule and vimentin networks seem to facilitate some nuclear deformation even in the absence of a prominent actin cytoskeleton. Together these results suggest that an interaction of all three cytoskeletal elements is needed for efficient nuclear deformation. In conclusion, the dominant parameters influencing nuclear deformation on micropillar substrates are not their material properties but the substrate geometry together with cell phenotype and cytoskeleton organization.

Multifunctional polymer scaffolds with adjustable pore size and chemoattractant gradients for studying cell matrix invasion

Transmigrating cells often need to deform cell body and nucleus to pass through micrometer-sized pores in extracellular matrix scaffolds. Furthermore, chemoattractive signals typically guide transmigration, but the precise interplay between mechanical constraints and signaling mechanisms during 3D matrix invasion is incompletely understood and may differ between cell types. Here, we used Direct Laser Writing to fabricate 3D cell culture scaffolds with adjustable pore sizes (2-10 μm) on a microporous carrier membrane for applying diffusible chemical gradients. Mouse embryonic fibroblasts invade 10 μm pore scaffolds even in absence of chemoattractant, but invasion is significantly enhanced by knockout of lamin A/C, a known regulator of cell nucleus stiffness. Nuclear stiffness thus constitutes a major obstacle to matrix invasion for fibroblasts, but chemotaxis signals are not essential. In contrast, epithelial A549 cells do not enter 10 μm pores even when lamin A/C levels are reduced, but readily enter scaffolds with pores down to 7 μm in presence of chemoattractant (serum). Nuclear stiffness is therefore not a prime regulator of matrix invasion in epithelial cells, which instead require chemoattractive signals. Microstructured scaffolds with adjustable pore size and diffusible chemical gradients are thus a valuable tool to dissect cell-type specific mechanical and signaling aspects during matrix invasion.

Benzylguanine thiol self-assembled monolayers for the immobilization of SNAP-tag proteins on microcontact-printed surface structures.

The site-selective, oriented, covalent immobilization of proteins on surfaces is an important issue in the establishment of microarrays, biosensors, biocatalysts, and cell assays. Here we describe the preparation of self-assembled monolayers consisting of benzylguanine thiols (BGT) to which SNAP-tag fusion proteins can be covalently linked. The SNAP-tag, a modified O(6)-alkylguanine-DNA alkyltransferase (AGT), reacts with the headgroup of BGT and becomes covalently bound upon the release of guanine. Bacterially produced recombinant His-tag-SNAP-tag-GFP was used to demonstrate the site-specific immobilization on BGT surface patterns created by microcontact printing (microCP). With this versatile method, any SNAP-tag protein can be coupled to a surface.

Quantitative methods for analyzing cell–cell adhesion in development

During development cell-cell adhesion is not only crucial to maintain tissue morphogenesis and homeostasis, it also activates signalling pathways important for the regulation of different cellular processes including cell survival, gene expression, collective cell migration and differentiation. Importantly, gene mutations of adhesion receptors can cause developmental disorders and different diseases. Quantitative methods to measure cell adhesion are therefore necessary to understand how cells regulate cell-cell adhesion during development and how aberrations in cell-cell adhesion contribute to disease. Different in vitro adhesion assays have been developed in the past, but not all of them are suitable to study developmentally-related cell-cell adhesion processes, which usually requires working with low numbers of primary cells. In this review, we provide an overview of different in vitro techniques to study cell-cell adhesion during development, including a semi-quantitative cell flipping assay, and quantitative single-cell methods based on atomic force microscopy (AFM)-based single-cell force spectroscopy (SCFS) or dual micropipette aspiration (DPA). Furthermore, we review applications of Förster resonance energy transfer (FRET)-based molecular tension sensors to visualize intracellular mechanical forces acting on cell adhesion sites. Finally, we describe a recently introduced method to quantitate cell-generated forces directly in living tissues based on the deformation of oil microdroplets functionalized with adhesion receptor ligands. Together, these techniques provide a comprehensive toolbox to characterize different cell-cell adhesion phenomena during development.

Quantitative analysis of type I collagen fibril regulation by lumican and decorin using AFM

Lumican and decorin, two members of the small leucine-rich repeat proteoglycan (SLRP) family, have been implicated as regulators of collagen I fibril structure in different tissues. Both proteoglycans consist of a core protein and a glycosaminoglycan (GAG) chain, but quantitative information regarding the precise role of the protein and GAG moieties in regulating collagen structure is still limited. In this study, we used AFM imaging and a model system of aligned collagen I nanofibrils to investigate the role of lumican and decorin on collagen I fibril structure with high resolution. When co-assembled with collagen I, recombinant lumican or decorin proteins lacking the GAG chains decreased collagen fibril width to values below <100nm and increased interfibrillar spacing in a dose-dependent manner. At lower concentrations, lumican appeared to have a stabilizing effect on newly-formed collagen fibrils, while at higher concentrations both lumican and decorin inhibited collagen fibrillogenesis. GAG-containing decorin also increased interfibrillar spacing, decreased fibril width and ultimately inhibited fibrillogenesis, but these effects required lower concentrations compared to recombinant decorin, indicating that the decorin core protein alone cannot compensate for the full regulatory and structural contribution of the GAG chain during collagen I fibrillogenesis. Using a 2D autocorrelation approach, we furthermore analyzed and compared the effects of recombinant and glycosylated decorin on collagen ultrastructure, providing a quantitative measure for the observed structural differences. AFM analysis of ordered fibrillar collagen arrays in combination with quantitative autocorrelation image analysis thus provides a useful tool for investigating SLRP-dependent nanoscale effects on collagen fibril structure.