An update on the minimal cell project: From the physics of solute encapsulation to the experimental modeling of cell communities

The minimal cell (MC) project aims at understanding the emergence of cellular life by constructing experimental models of cells, according to a synthetic (constructive) biology approach. Our strategy – also known as the semi-synthetic one – is based on the encapsulation of the minimal number of biomolecular components inside lipid vesicles (liposomes). Being interested in studying the key step for constructing semisynthetic cells, namely the physical entrapment of the solutes, we have recently reported that the mechanism of vesicle formation can lead to a spontaneous local increase in concentration of proteins inside vesicles (Luisi et al., ChemBioChem 2010, 11, 1989-1992). In particular, it was shown that the protein ferritin can reach intravesicle concentration of at least one order of magnitude higher when compared to the bulk (external) concentration. This selforganization phenomenon might give a rational account for the formation of functional cells from diluted solutions, and therefore help to understand the origin of metabolism. The effective encapsulation of solutes, however, is only one of the ways for achieving functional cells. The second route is fusion of vesicles or the exchange of solutes among vesicles (Caschera et al., J. Coll. Inter. Sci. 2010, 345, 561-565). Both processes allow the combination of different solutes to give compartments that can exhibit improved reactivity. Aiming at developing a realistic model for cooperative interactions among vesicles, we have recently developed a cell colony model. This is based on the formation of lipid vesicles clusters adherent to a solid substrate, representing a minimal model of cell communities. Here we summarize the most significant aspects of our recent activities. The physics of solute encapsulation Looking at the physico-chemical mechanisms that have lead to the origin of cellular life, a still open question is whether functional cells have been originated from the encapsulation of an already developed metabolism (metabolismor replicator-first scenarios), or whether the cell metabolism was entirely (or almost entirely) developed inside compartments (compartment-first scenario). In both cases, there are some aspects that need clarification, as the low probability of coentrapping all required molecules in the same compartment in the first hypothesis, or the lack of permeability control in the second hypothesis (Luisi et al., 2010). In particular, although the encapsulation of solutes into liposomes is a well-established field, especially due to the large amount of work done in the field of drug delivery, we still miss a complete view of the physics underlying this important mechanism. In fact, with a few exceptions (Sun and Chiu, 2005; Dominak and Keating, 2007; Lohse et al., 2008), all experimental studies deal with the average entrapment yield, and no attention has been given to the entrapment behavior at the level of single vesicles, also due to technical difficulties. We have recently started a systematic study on the encapsulation of biopolymers into lipid vesicles. This study was inspired by our report on the protein expression inside 200 nm (diameter) vesicles, that suggested the possible deviation from the expected intravesicle solute distribution (Souza et al., 2009). As a model system, we have used the protein ferritin, an ironstorage protein, consisting of a nucleus of electron dense ferrihydrite-like iron salts surrounded by 24 protein subunits. Ferritin can be directly visualized as single molecule by electronmicroscopy, so that it becomes possible to directly count the number of ferritin molecules inside vesicles imaged via cryo-transmission electron microscopy. After analyzing about 7,700 submicrometric vesicles (Fig. 1a), prepared by varying the concentration of ferritin, the preparation method and the membrane lipid composition, we have concluded that the encapsulation of this solute inside lipid vesicles does not follow the expected behavior. In our experimental conditions, this is given by the Poisson distribution of N solutes inside vesicles that are expected to entrap, on average, μ solutes: