We present a protocol to functionalize glass with nanometric protein patches surrounded by a fluid lipid bilayer. These substrates are compatible with advanced optical microscopy and are expected to serve as platform for cell adhesion and migration studies.
Currently there is considerable interest in creating ordered arrays of adhesive protein islands in a sea of passivated surface for cell biological studies. In the past years, it has become increasingly clear that living cells respond, not only to the biochemical nature of the molecules presented to them but also to the way these molecules are presented. Creating protein micro-patterns is therefore now standard in many biology laboratories; nano-patterns are also more accessible. However, in the context of cell-cell interactions, there is a need to pattern not only proteins but also lipid bilayers. Such dual proteo-lipidic patterning has so far not been easily accessible. We offer a facile technique to create protein nano-dots supported on glass and propose a method to backfill the inter-dot space with a supported lipid bilayer (SLB). From photo-bleaching of tracer fluorescent lipids included in the SLB, we demonstrate that the bilayer exhibits considerable in-plane fluidity. Functionalizing the protein dots with fluorescent groups allows us to image them and to show that they are ordered in a regular hexagonal lattice. The typical dot size is about 800 nm and the spacing demonstrated here is 2 microns. These substrates are expected to serve as useful platforms for cell adhesion, migration and mechano-sensing studies.
Cell adhesion takes place through specialized cell adhesion molecules (CAMs), proteins present on the cell membrane that are capable of binding to their counterpart on the extra cellular matrix or on another cell. On adhered cells, most adhesion molecules including the ubiquitous integrin and cadherin, are found in the form of clusters 1. The interaction of T lymphocytes (T cells) with antigen presenting cells (APCs) provides a particularly striking illustration of the importance of receptor clusters formed at the interface between the two cells — often called an immunological synapse. Upon forming the first contact with the APC, T cell receptors (TCRs) on the surface of the T cell form micron scale clusters that serve as signaling platforms 2,3,4, and are eventually centralized to form a larger central supramolecular cluster (cSMAC) 5,6,7. Recently, it was shown that on the APC side, the ligands of the TCR are also clustered 8.
In the context of T cell-APC interaction, the deployment of hybrid systems, where the APC is mimicked by an artificial surface functionalized with relevant proteins, has greatly advanced our understanding of the synaptic interface 2,3,4,5,6,7. In this context, it is highly relevant to design APC mimetic surfaces that capture one or more aspects of the target cell. For example, if ligands are grafted on supported lipid bilayers, they can diffuse in the plane of the bilayer, mimic the situation on the APC surface and at the same time allow the formation of the cSMAC 6,7. Similarly, the clusters on the APC have been mimicked by creating islands of ligands in a sea of polymers 9,10,11,12,13,14. However, these two features have so far not been combined.
Here we describe a novel technique to create nano-dots of anti-CD3 (an antibody that targets the TCR complex) surrounded by a lipid bilayer with diffusing lipids. The bilayer is deposited using Langmuir-Blodgett/Langmuir-Schaefer technique 7,15,16 and if desired, could be functionalized with a specific protein — for example, the ligand of the T cell integrin (called ICAM1). In addition, the anti-CD3 protein dots could be replaced with another antibody or CAM. While we have chosen the proteins for future use as platform for T cell adhesion studies, the strategy detailed here can be adapted for any protein and even DNA.
1. Cleaning Glass Cover-slides and Observation Chambers
2. Fabrication of Protein Nano-dots
3. Deposition of Supported Lipid Bilayer (SLB)
4. Functionalization with Ligands
5. Cell Deposition (see reference 7 for details)
6. Observation
The fluorescence images were analyzed to measure the spacing and size of the dots. Typical spacing was found to be 1,900 ± 80 nm and typical dot-size was 600 ± 100 nm (Figure 1g). The spacing is set by the size of the beads used for the mask. The dot-size is set by the bead-size as well as deposition conditions. The SLB is deposited uniquely around the protein dots and not on them (Figure 2), with perfect complementarity between the holes seen in SLB imaging channel and the dots seen in the NAV imaging channel. Analysis of continuous photo bleaching data shows that the lipids in the patterned bilayer remain fluid and have a typical diffusion constant of 5 µm²/s (Figure 3).
Figure 1: Schematic representation of fabrication steps. (a) Primary bead mask. Inset shows a scanning electron microscopy (SEM) image of the mask, made with beads of 2 µm diameter and covered with a thin layer of aluminum. Parameters used in this image are: nominal aluminum thickness 200 nm deposed at 400 W RF, initial pressure of 9 x 10-7 Torr, Argon flux 10 sccm, process pressure 6.2 mTorr, imaged at acceleration voltage of 5 kV. The images confirm observation made with optical microscopy (image not included) before aluminum deposition that the beads are arranged in a monolayer centered hexagonal lattice on the glass-slide. (b) Secondary mask of aluminum created by the sputter deposition after removal of the primary bead-mask. (c) Deposition of organosilane and BSA-biotin though the secondary mask. (d) Removal of aluminum revealing nano-dots of BSA-biotin. (e) Deposition of SLB. (f) Binding NaV to BSA-biotin. (g) Binding anti-CD3 to NaV. Inset shows epi-fluorescence image of the nano-dots arrays. Here the NaV is fluorescently labelled. Typical dot-size is 600±100 nm and typical spacing is 1,900 ± 80 nm. Please click here to view a larger version of this figure.
Figure 2: Complementarity of the nano-dot and SLB pattern. (a,b) Epi-fluorescence images of fluorescent NaV nano-dots and of SLB with fluorescent tracer lipids. (c) Composite image of the NaV dots (red) in the sea of SLB (green) shows perfect complementarity of the NaV and SLB. Fast Fourier Transform (FFT) image in the inset indicates long-range order. Scale bar: 4 µm. Please click here to view a larger version of this figure.
Figure 3: Quantification of lipid diffusion in the SLB. (a) Epi-fluorescence image of a SLB before bleaching. The protein dots show up as dark holes in a bright sea of lipids. The field-diaphragm limits the illuminated area. (b) Epi-fluorescence of SLB after bleaching continuously for 50 s. The halo visible inside the region delimited by the field-diaphragm indicates that the lipids are mobile. Scale bar: 10 µm. (c) Average intensity profile along the edge of the field-diaphragm (top) and the decay of intensity over time during the beaching process (bottom). This data is analyzed to extract the diffusion constant, which is typically 5 µm²/s.
The critical steps within the protocol described above are related to the formation of the protein nano-dots or the back-filling of the space around the dots by a supported lipid bilayer. The first critical step with respect to protein nano-dots is the preparation of the bead-mask. The cleaning of the cover-slide is critical. The slides need to be either cleaned with a detergent solution that is recommended for cleaning quartz cuvettes, or with oxygen plasma. Other cleaning techniques like immersion in ethanol or iso-propanol treatment do not render the glass sufficiently hydrophilic and therefore do not support the formation of a large-coverage bead monolayer. At the same time, the surface of the slide needs to be compatible with the subsequent step of formation of the SLB by Langmuir-Blodgett technique 7,15,16. The next critical step is the deposition and removal of aluminum. If the deposition is via sputtering technique, as done here, the target should to be doped with silicon (at 1%). Otherwise, if the target is made of pure aluminum, the deposited layer is hard to remove with the alkali hydroxide solution as described above, probably due to the formation of aluminum oxide and interpenetration between deposited aluminum layer and glass slide substrate. The duration of deposition determines the size of the protein dots 9, however, here we have worked with a single deposition time and therefore single dot size. The sample can be stored under ambient conditions for several months after aluminum deposition and for about a week after the deposition of BSA-biotin.
The second crucial step concerns the deposition of the SLB. Cleaning of the glass cover-slide is again a crucially important point. As is the case for any Langmuir-Blodgett deposition work, all the material used should be made of glass or Polytetrafluoroethylene (PTFE) and should be scrupulously clean. After deposition of the first lipid monolayer, the cover-slides can be stored for a couple of days but after the deposition of the second monolayer, they need to be used immediately.
While we have demonstrated the protocol for creating anti-CD3 nano-dots for use in T lymphocyte adhesion studies 9,13, the procedure is highly flexible and can be adapted for any biotinylated protein. The composition of the lipid bilayer can be easily changed and it can be further functionalized if desired. One important point to consider is the possible unspecific absorption of proteins, especially on the lipid bilayer surrounding the dots.
The main limitation of the technique arises from the use of colloidal-bead self-assembly for the primary mask. Being a bottom-up technique, it shares some of the problems of all such approaches for example, lack of flexibility and full control over the pattern shape. The pattern lattice necessarily reflects the symmetry of the bead mask and is therefore always hexagonal. The shape of the pattern motif is a circle and is determined by a combination of bead-size and the duration of aluminum deposition 9,13. Alternative techniques for controlling the dot size have also been suggested 14,17,18.
Substrates nano-patterns with proteins, protein-fragments or peptides have been extensively used in the past to probe cell-surface interactions, especially adhesion 19 and migration 20. Pioneering work has shown that tissue-forming cells fail to spread on patterns with a pitch greater than a given threshold 21, and further investigation showed that the length-scale of this phenomenon is set by the size of talins, which are instrumental in linking integrin receptors to the actin cytoskeleton 22. However, in all these studies, the proteins were linked to gold nano-particles, which were themselves immobilized on glass.
In the context of T cells, proteo-lipidic membranes, typically mimicking antigen presenting cells, have been extensively used to understand fundamental aspects of T cell function 6. Ingenious nano-patterning techniques have been used to create corrals with metal barriers separating protein functionalized SLB patches, which have provided insight into the structure and connectivity of the T cell/APC interface 23. This kind of nano-patterning is however very different from the protein nano-dots proposed here. Recently, nano- clusters were created using chemical linkage of the protein functionalized lipids 3, which shed light on the consequence of receptor clustering. The advantage was that, unlike the present case, the clusters could in principle be themselves mobile. However, such spontaneously formed clusters are necessarily less well controlled in terms of size and density than pre-formed protein nano-dots described here.
We envisage that the protein nano-dot decorated SLBs presented here can be used to investigate different aspects of cell-cell adhesion. One obvious question that arises is whether, as described above for tissue forming cells 19, lymphocytes too have an intrinsic length-scale associated with adhesion. Preliminary results seem to indicate that at least when the adhesion is mediated by the TCR complex, the density of ligands rather than spacing is the defining parameter for spreading and activation 10,11,12,13. Whether inclusion of mobile ligands in the surrounding SLB impacts this observation and how the mobile and immobile fractions work together is a possible question, which was partly addressed using self-assembled SLB bound clusters 24. Another interesting application will be in the context of mechano-sensing where cell adhesion/activation on mobile and immobilized ligands were shown to be different not only for T cells 7,25 but also for cells that habitually adhere to the extra cellular matrix 26.
The two main advantages of these proteo-lipidic patterns are the compatibility of the substrates with advanced optical microscopy and the ease of preparation which makes them compatible with use-and-throw applications. Subsequent to aluminum deposition, all the preparation steps can be carried out on a standard wet-lab bench. In the future, it can be envisaged that the aluminum-coated and glass-supported bead-masks produced in a specialized facility are transferred and stored in biology laboratories for use as and when required. With this in view, we believe that these substrates have the potential to become the platform of choice for studying the interaction of cells with controlled nano-patterned proteo-lipidic membranes.
The authors have nothing to disclose.
We thank Laurent Limozin, Pierre Dillard and Astrid Wahl for continuing fruitful discussions about cellular applications. We also thank Frederic Bedu from PLANETE cleanroom facility for his help with SEM observations. This work was partially funded by the European Research Council via grant No. 307104 FP/2007-2013/ERC.
Glass coverslips | Assistent, Karl Hecht KG | |
Observation chamber | Home made | |
Alkaline surfactant concentrate (Hellmanex) | Hella Analytics | 9-307-011-4-507 |
Ultra-sonicator | ThermoFisher | |
Desiccator | Labbox | |
Crystallizer | Shott | |
Neutravidine | Thermo Fischer Scientifique | 84607 |
PBS | Sigma-aldrich | P3813 |
Water MQ | ELGA, Veolia France | |
Silica beads | Corpuscular Inc | 147114-10 |
APTES | Sigma-aldrich | A3648 |
BSA-Biotin | Sigma-aldrich | A8549 |
DOPC | Avanti Polar Lipids | 850375C |
Dansyl-PE | Avanti Polar Lipids | 810330C |
Chloroform | Sigma-aldrich | 650471 |
Gastight syringe | Dominique Dustcher , France | 74453 |
Film balance | NIMA | Medium |
Microscope | Zeiss, Germany | TIRF-III system |
Aluminium Target | Kurt J. Lesker Compagny, USA | |
Radio Frequency Magnetron sputtering Système | modified SMC 600 tool by ALCATEL , France |