On-Chip Quantitative Measurement of Mechanical Stresses During Cell Migration with Emulsion Droplets

The ability of immune cells to migrate within narrow and crowded spaces is a critical feature involved in various physiological processes from immune response to metastasis. Several in-vitro techniques have been developed so far to study the behaviour of migrating cells, the most recent being based on the fabrication of microchannels within which cells move. To address the question of the mechanical stress a cell is able to produce during the encounter of an obstacle while migrating, we developed a hybrid microchip made of parallel PDMS channels in which oil droplets are sparsely distributed and serve as deformable obstacles. We thus show that cells strongly deform droplets while passing them. Then, we show that the microdevice can be used to study the influence of drugs on migration at the population level. Finally, we describe a quantitative analysis method of the droplet deformation that allows measuring in real-time the mechanical stress exerted by a single cell. The method presented herein thus constitutes a powerful analytical tool for cell migration studies under confinement.

Force spectroscopy of a single artificial biomolecule bond: the Kramers' high-barrier limit holds close to the critical force

We use a minimal system with a single micron-size bead trapped with optical tweezers to investigate the kinetics of escape under force. Surprisingly, the exponential decay of the off rate with the barrier energy is still valid close to the critical force. Hence, the high viscosity approximation derived by Kramers in the case of a high energy barrier holds even for an energy barrier close to the thermal energy. Several recent models describe a single biomolecule bond by a smooth single-barrier energy profile. When this approach is accurate enough, our result justifies the use of Kramers' approximation in the high-force regime, close to the critical force of the system, as done in recent single biomolecule bond studies.

Indirect evidence of submicroscopic pores in giant unilamellar [correction of unilamelar] vesicles

Formation of pore-like structures in cell membranes could participate in exchange of matter between cell compartments and modify the lipid distribution between the leaflets of a bilayer. We present experiments on two model systems in which major lipid redistribution is attributed to few submicroscopic transient pores. The first kind of experiments consists in destabilizing the membrane of a giant unilamellar vesicle by inserting conic-shaped fluorescent lipids from the outer medium. The inserted lipids (10% of the vesicle lipids) should lead to membrane rupture if segregated on the outer leaflet. We show that a 5-nm diameter pore is sufficient to ease the stress on the membrane by redistributing the lipids. The second kind of experiments consists in forcing giant vesicles containing functionalized lipids to adhere. This adhesion leads to hemifusion (merging of the outer leaflets). In certain cases, the formation of pores in one of the vesicles is attested by contrast loss on this vesicle and redistribution of fluorescent labels between the leaflets. The kinetics of these phenomena is compatible with transient submicroscopic pores and long-lived membrane defects.

Hemifusion and fusion of giant vesicles induced by reduction of inter-membrane distance

Proteins involved in membrane fusion, such as SNARE or influenza virus hemagglutinin, share the common function of pulling together opposing membranes in closer contact. The reduction of inter-membrane distance can be sufficient to induce a lipid transition phase and thus fusion. We have used functionalized lipids bearing DNA bases as head groups incorporated into giant unilamellar vesicles in order to reproduce the reduction of distance between membranes and to trigger fusion in a model system. In our experiments, two vesicles were isolated and brought into adhesion by the mean of micromanipulation; their evolution was monitored by fluorescence microscopy. Actual fusion only occurred in about 5% of the experiments. In most cases, a state of "hemifusion" is observed and quantified. In this state, the outer leaflets of both vesicles' bilayers merged whereas the inner leaflets and the aqueous inner contents remained independent. The kinetics of the lipid probes redistribution is in good agreement with a diffusion model in which lipids freely diffuse at the circumference of the contact zone between the two vesicles. The minimal density of bridging structures, such as stalks, necessary to explain this redistribution kinetics can be estimated.

How hard is a colloidal "hard-sphere" interaction?

Poly-12-hydroxystearic acid (PHSA) is widely used as a coating on colloidal spheres to provide a "hard-sphere-type" interaction. These hard spheres have been widely used in fundamental studies of nucleation, crystallization, and glass formation. Most authors describe the interaction as "nearly" hard sphere. In this paper we directly measure this interaction, using layers of PHSA adsorbed onto mica sheets in a surfaces force apparatus. We find that the layers, in appropriate solvents, have no long-range interaction. When the solvent is decahydronaphthalene (decalin), the repulsion rises from zero to the maximum measurable over a distance range of 15-20 nm. The data is converted to equivalent forces between spheres of different diameters, and modeled using a hard core potential. Using zeroth-order perturbation theory and computer simulation, we demonstrate that the equation of state does not deviate from that of a perfect hard-sphere system under any relevant experimental conditions.

Short-range specific forces are able to induce hemifusion

Working with pure lipidic systems (giant unilamellar vesicles, 10-150 microm in diameter) as models for biological membranes, we have considered possible structures of the contact area of two adherent membranes by investigating the diffusion of fluorescent lipid analogues from one vesicle to another. Two bilayers in close contact can almost be seen as a lamellar structure in equilibrium. This is the usual configuration of two adherent vesicles, in which the interbilayer distance is estimated to be 3 nm. We have increased the attraction between the membranes by either adding depletion forces or by using a trick, inspired from the interaction between nucleic bases in nucleosides (herein adenosine and thymidine). The nucleosides were attached to the polar head of amphiphilic molecules that behave like phospholipids and were incorporated in the model membrane. The extra attraction between two membranes, resulting from base pairing, strongly decreased the interbilayer distance down to about 1 nm. This change of the water content induced lipid rearrangements, which could also be viewed in terms of a phase transition at low water content. These rearrangements were not observed in the case of depletion forces. We conclude that the introduction of an additional attractive force in the system modifies the equilibrium state, leading to a drastic change in the membrane behavior, which will tentatively be related to hemifusion.

From macroscopic adhesion energy to molecular bonds: a test of the theory

We present a statistical mechanical treatment relating the macroscopic adhesion energy of two surfaces, which can be obtained by micropipette aspiration studies, to the microscopic adhesion energy between individual bonds. The treatment deals with the case of weak reversible bonds, so that the equilibrium partition function has significance. This description is coherent with previous theories. Experiment and theory are compared to probe the nature of weak bonds in membranes, where local equilibria can be obtained. The case of a bead and a vesicle decorated by nucleosides was considered.

Ultraweak sugar-sugar interactions for transient cell adhesion

Carbohydrate-carbohydrate interactions are rarely considered in biologically relevant situations such as cell recognition and adhesion. One Ca(2+)-mediated homotypic interaction between two Lewis(x) determinants (Le(x)) has been proposed to drive cell adhesion in murine embryogenesis. Here, we confirm the existence of this specific interaction by reporting the first direct quantitative measurements in an environment akin to that provided by membranes. The adhesion between giant vesicles functionalized with Le(x) was obtained by micropipette aspiration and contact angle measurements. This interaction is below the thermal energy, and cell-cell adhesion will require a large number of molecules, as illustrated by the Le(x) concentration peak observed at the cell membranes during the morula stage of the embryo. This adhesion is ultralow and therefore difficult to measure. Such small interactions explain why the concept of specific interactions between carbohydrates is often neglected.

Do trehalose and dimethyl sulfoxide affect intermembrane forces?

The sugar trehalose is produced in some organisms that survive dehydration and desiccation, and it preserves the integrity of membranes in model systems exposed to dehydration and freezing. Dimethyl sulfoxide, a solute which permeates membranes, is added to cell suspensions in many protocols for cryopreservation. Using a surface forces apparatus, we measured the very large, short-range repulsion between phosphatidylcholine bilayers in water and in solutions of trehalose, sorbitol, and dimethyl-sulfoxide. To the resolution of the technique, the force-distance curves between bilayers are unchanged by the addition of trehalose or sorbitol in concentrations exceeding 1 kmol.m-3. A relatively small increase in adhesion in the presence of trehalose and sorbitol solutions may be explained by their osmotic effects. The partitioning of trehalose between aqueous solutions and lamellar phases of dioleylphosphatidylcholine was measured gravimetrically. The amount of trehalose that preferentially adsorbs near membrane surfaces is at most small. The presence of dimethyl sulfoxide in water (1:2 by volume) makes very little difference to the short-range interaction between deposited bilayers, but it sometimes perturbs them in ways that vary among experiments: free bilayers and/or fusion of the deposited bilayers were each observed in about one-third of the experiments.