Protein micro patterned lattices to probe a fundamental lengthscale involved in cell adhesion.

Protein micro patterned lattices to probe a fundamental lengthscale involved in cell adhesion.

Hervé Guillou Institut Néel, CNRS UPR2940 et Université Joseph Fourier, BP166, 38042 Grenoble, France    Benoit Vianay Institut Néel, CNRS UPR2940 et Université Joseph Fourier, BP166, 38042 Grenoble, France    Jacques Chaussy Institut Néel, CNRS UPR2940 et Université Joseph Fourier, BP166, 38042 Grenoble, France    Marc R. Block INSERM U823, Equipe DySAD, Institut Albert Bonniot, Site Santé, BP170, 38047 LaTronche, France

Cell adhesion, a fundamental process of cell biology is involved in the embryo development and in numerous pathologies especially those related to cancers. We constrained cells to adhere on extracellular matrix proteins patterned in a micro lattices. The actin cytoskeleton is particularly sensitive to this constraint and reproducibly self organizes in simple geometrical patterns. Such highly organized cells are functional and proliferate. We performed statistical analysis of spread cells morphologies and discuss the existence of a fundamental lengthscale associated with active processes required for spreading.

thanks: author’s fax: 33+456 38 70 87

In vivo, cells proliferate on the 3D structure of extracellular proteins that forms a highly organized scaffold: the extracellular matrix (ECM) is an arrangement of different proteinsalberts (). Cell adhesion onto the ECM is a fundamental process involved in the embryo development as well as in numerous pathologies especially those related to cancers thiery2003 (). In many cell types, the integrin family of transmembrane proteins is responsible for cell interactions with ECM proteins. This specific interaction among other micro environmental physicochemical parameters such as support compliance discher2005 () or shape huang1999 (); Chen1997 (); thery2006 () has been shown to be relevant for the regulation of cell proliferation and to the accomplishment of their functions.

Following huang1999 (); Chen1997 (); thery2006 () we use adequately tailored micro lattice of proteins to obtained very reproducible and simple actin cytoskeleton organization. The microfabrication of the protein array and the simplified actin cytoskeleton morphology that we systematically observe are described next. The result of the statistical analysis of cell shapes are presented in a second part. In particular, we discuss results indicating the existence of a characteristic length. This length is associated with the actin based, active protrusive processes needed to explore the environment and to establish new adhesive contacts required for spreading and migration.

Figure 1: Kinetic of fibronection adsorption on hydrophobic glass substrates. Lower inset: adsorption isotherm on hydrophobic surface. Upper inset: durability of protein patterns with cells cultured on fibronectin square islands.

According to amphiphilic properties of globular proteins they tend to accumulate and adsorb at interfaces israelachvili1991 (). Hydrophobic surfaces yield protein films of larger density than hydrophilic surfaces sorribas2002 (). The principle of fabrication of the protein lattice is based on such non specific protein immobilization on hydrophobic surfaces through a patterned photo-resist mask followed by a lift-off. The photo-resist is insolated using optical UV lithography. Patterns with a critical size of are thus achievable. In this study, although several proteins where successfully tested (fibronectin, fragment of fibronectin, antibodies, vitronectin, gelatin) best results in terms of spatial resolution were obtained with a fragment of fibronectin made of domains of type III 7 to 10Leahy1996 (). Fig. 1 shows the adsorption kinetic of the fragment at . The adsorption isotherm on hydrophobic surface is shown in the inset of fig. 1. Both graphs indicate that the adsorption is limited as expected guemouri2000 (). After adsorption, the photo-resist mask is lifted-off. The yet photo-resist protected surface must be blocked to prevent further adsorption of protein from the culture medium that would otherwise rapidly blur out the protein pattern. This major problem in protein patterning and has been solved zhang1998 (); liu2002 () by using a commercially available polymer pluronic. () that spontaneously adsorb on hydrophobic surface in the same manner as proteins. We were able to maintain consistent cell patterns and to keep the repellent properties of the surface for more than 4 days as shown in the images of figure 1: the square adhesive islands kept their geometry despite confluent cell crowding. The lift-off of proteins on hydrophobic surfaces followed by the immobilization of blocking polymer is a very efficient way to produce high resolution adhesive patterns of controlled geometry. Nevertherless it will never reach the resolution achieved in Cavalcanti-Adam2007 () where patterning at the molecular level is realised. However, in contrast with microstamping it produces patterns of very high quality and allows easy alignement with existing structures.

Figure 2: Typical single cell shapes observed for various protein lattice. Red is F-actin cytosqueleton, green is vinculin and blue is fibronectin. Scale : white bar is . Non patterned substrate (a), lattice pitch (b), (c), (d), (e).

We tested the biocompatibility of the protein lattice by studying cell spreadings. Cells were harvested and plated onto the patterns and spread for 4 hours before fixation in paraformaldehyde. The actin cytoskeleton was revealed using rhodamine labelled phalloidin and focal contacts by primary antibody directed toward vinculin. Cover slips were mounted in mowiol for fluorescence microscopy. More details on the methods are given in Guillou2008 (). For non patterned substrates (fig. 2a) no particular actin cytoskeleton organization can be observed. For small distances between adhesive dots (fig. 2b), the cytoskeleton organisation is similar to the one observed on non patterned substrates. However, as the distance between adhesive spots is increased, the shape of cells is very quickly simplified into elementary geometries (fig. 2c-d). It is remarquable how the complex cytoskeleton organization of fig.s 2a-b is simplified: stress fibers are exclusively located on the edges of the cell body and linked to well identified focal contacts at their ends.

Each actin cytoskeleton organization was classified according to its area and compacity compacity (). Fig. 3 shows the population of cells in each class for a lattice pitch of 10 µm.

Figure 3: Distribution of shape obtained for a dot spacing of . The colorcode represent the number of cells observed in a particular configuration. The maxima of occurence are obtained for the 4 dots square and 6 dots rectangle shown in the inset.

Each peak corresponds to a simple geometrical compact shape that can be drawn between adhesive islands. It shows that the most probable organization is the 4 dots square and that cell shapes can be classified.

Figure 4: Shape selection through protrusion length. Left axis, open squares: average mean area of single spread cells on proteins arrays with different pitches. Right axis, open circles: probability to find a filopodia of a given length (averaged on 20 different cells and 400 filopodia.

On a non patterned substrate the ECM layer is dense and almost every membrane protrusion leads to engagement of integrins. Very quickly cells adopt a migrating phenotype in which the cytoskeleton is continuously rearranging. Cell shapes cannot be classified. By contrast, on a protein lattice, a finite distance between neighbouring adhesive spots is introduced and the membrane protrusion are less efficient in finding new ligands. Fig. 4 supports this idea and compares the mean area of spread cells as a function of the lattice pitch with the distribution of the length of filopodia. The mean area decreases very rapidly once the distance is increased. As shown on fig. 4 membrane protrusions are able to explore only the close environment of cells. The probability that a protrusion exceeds a distance of is low. Moreover, because of the geometry of the lattice, all protrusions are not successful in finding new ligand, reducing further the probability to engage membrane receptors. The distance between adhesive spots introduce thus a length that can be compared with the average length of protrusion. By increasing the distance between adhesive spots the probability to change the set of adhesive constraint is gradually reduced. The actin cytoskeleton has thus time to relax and to self organize into very simple shapes. The protein lattice probe thus a fundamental length related to the protrusion length.

In this letter we studied single cell spreading onto micro fabricated patterns of controlled geometries. Protrusions are the limiting process resulting in a greatly reduced spread cell area as soon as the pitch is greater than their typical length. As a result stress fibers are exclusively located at the edges of the cell body and attached to two focal adhesion at their ends.


  • [1] B. Alberts, D. Bray, J. Lewis, M. Raff, K. Roberts, and J.D. Watson. The Molecular Biology of the Cell. Garland Publishing, New-York, 1994.
  • [2] J.P. Thiéry. Cell adhesion in cancer. C.R. Phys., 4:289, 2003.
  • [3] D.E. Discher, P. Janmey, and Y.L. Wang. Tissue cells feel and respond to the stifness of their substrate. Science, 310:p1139, 2005.
  • [4] S. Huang and D.E. Ingber. The structural and maechanical complexity of cell-growth control. Nat. Cell Bio., 1:pE131–8, 1999.
  • [5] C.S. Chen, M. Mrksich, S. Huang, G.M. Whiteside, and D.E. Ingber. Geometric control of cell life and death. Science, 276:1425, 1997.
  • [6] M. Théry, V. Racine, M. Piel, A. Pépin, A. Dimitrov, Y. Chen, J.B. Sibarita, and M. Bornens. Anisotropy of cell adhesive microenvironment governs cell internal organization and orientation of polarity. Proc. Natl. Acad. Sci. USA, 103:p19771, 2006.
  • [7] J.N. Israelachvili. Intermolecular and surface forces. Academic Press, 1991.
  • [8] H. Sorribas, C. Padeste, and Tiefenauer L. Photolithographic generation of protein micropatterns for neuron culture applications. Biomaterials, 23:p893, 2002.
  • [9] D.J. Leahy, I. Aukhil, and H.P. Erickson. 2.0 a crystal structure of a four-domain segment of human fibronectin encompassing the rgd loop and synergy region. Cell, 84:pp155–164, 1996.
  • [10] L. Guemouri, J. Ogier, Z. Zekhini, and J.J. Ramsden. The architecture of fibronectin at surfaces. J. Chem. Phys., 113:p8183–8186, 2000.
  • [11] M. Zhang, T. Desai, and M. Ferrari. Proteins and cells on peg immobilized silicon surfaces. Biomaterials, 19:p953–960, 1998.
  • [12] V.A. Liu, W.E. Jastromb, and S. N. Bhatia. Engineering protein and cell adhesivity using peo-terminated triblock polymers. J. Biomed. Mater. Res., 60:p126, 2002.
  • [13] Pluronic is a BASF copolymer. BASF, Germany.
  • [14] E.A. Cavalcanti-Adam, T. Volberq, A. Micoulet, H. H. Kessler, B. Geiger, and J.P. Spatz. Cell spreading and focal adhesion dynamics are regulated by spacing of integrin ligands. Biophys. J., 92:2964, 2007.
  • [15] H. Guillou, A. Depraz-Depland, E. Planus, B. Vianay, J. Chaussy, A. Grichine, C. Albigès-Rizo, and M.R. Block. Lamellipodia nucleation by filopodia depends on integrin occupancy and downstream rac1 signaling. Exp. Cell Res., 314:478, 2008.
  • [16] compacity is defined as in the IMAQ library of the Labview programming software from National Instrument, Austin, Texas, USA.
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
Add comment
Loading ...
This is a comment super asjknd jkasnjk adsnkj
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test description