Filippo Caschera

Programmed Vesicle Fusion Triggers Gene Expression

The membrane properties of phospholipid vesicles can be manipulated to both regulate and initiate encapsulated biochemical reactions and networks. We present evidence for the inhibition and activation of reactions encapsulated in vesicles by the exogenous addition of charged amphiphiles. While the incorporation of cationic amphiphile exerts an inhibitory effect, complementation of additional anionic amphiphiles revitalize the reaction. We demonstrated both the simple hydrolysis reaction of β-glucuronidase and the in vitro gene expression of this enzyme from a DNA template. Furthermore, we show that two vesicle populations decorated separately with positive and negative amphiphiles can fuse selectively to supply feeding components to initiate encapsulated reactions. This mechanism could be one of the rudimentary but effective means to regulate and maintain metabolism in dynamic artificial cell models.

Detection of association and fusion of giant vesicles using a fluorescence-activated cell sorter.

We have developed a method to evaluate the fusion process of giant vesicles using a fluorescence-activated cell sorter (FACS). Three fluorescent markers and FACS technology were used to evaluate the extent of association and fusion of giant vesicles. Two fluorescent markers encapsulated in different vesicle populations were used as association markers; when these vesicles associate, the two independent markers should be observed simultaneously in a single detection event. The quenched fluorescent marker and the dequencher, which were encapsulated in separate vesicle populations, were used as the fusion marker. When the internal aqueous solutions mix, the quenched marker is liberated by the dequencher and emits the third fluorescent signal. Although populations of pure POPC vesicles showed no detectable association or fusion, the same populations, oppositely charged by the exogenous addition of charged amphiphiles, showed up to 50% association and 30% fusion upon population analysis of 100,000 giant vesicles. Although a substantial fraction of the vesicles associated in response to a small amount of the charged amphiphiles (5% mole fraction compared to POPC alone), a larger amount of the charged amphiphiles (25%) was needed to induce vesicle fusion. The present methodology also revealed that the association and fusion of giant vesicles was dependent on size, with larger giant vesicles associating and fusing more frequently.

Stable Vesicles Composed of Monocarboxylic or Dicarboxylic Fatty Acids and Trimethylammonium Amphiphiles

The self-assembly of cationic and anionic amphiphile mixtures into vesicles in aqueous media was studied using two different systems: (i) decanoic acid and trimethyldecylammonium bromide and (ii) hexadecanedioic acid (a simple bola-amphiphile) and trimethyldecylammonium bromide. The resulting vesicles with varying amphiphile ratios were characterized using parameters such as the critical vesicle concentration, pH sensitivity, and encapsulation efficiency. We also produced and observed giant vesicles from these mixtures using the electroformation method and confocal microscopy. The mixed catanionic vesicles were shown to be more stable than those formed by pure fatty acids. Those containing bola-amphiphile even showed the encapsulation of a small hydrophilic solute (8-hydroxypyrene-1,3,6-trisulfonic-acid), suggesting a denser packing of the amphiphiles. Compression and kinetics analysis of monolayers composed of these amphiphiles mixtures at the air/water interface suggests that the stabilization of the structures can be attributed to two main interactions between headgroups, predominantly the formation of hydrogen bonds between protonated and deprotonated acids and the additional electrostatic interactions between ammonium and acid headgroups.

Automated Discovery of Novel Drug Formulations Using Predictive Iterated High Throughput Experimentation

Background We consider the problem of optimizing a liposomal drug formulation: a complex chemical system with many components (e.g., elements of a lipid library) that interact nonlinearly and synergistically in ways that cannot be predicted from first principles. Methodology/Principal Findings The optimization criterion in our experiments was the percent encapsulation of a target drug, Amphotericin B, detected experimentally via spectrophotometric assay. Optimization of such a complex system requires strategies that efficiently discover solutions in extremely large volumes of potential experimental space. We have designed and implemented a new strategy of evolutionary design of experiments (Evo-DoE), that efficiently explores high-dimensional spaces by coupling the power of computer and statistical modeling with experimentally measured responses in an iterative loop. Conclusions We demonstrate how iterative looping of modeling and experimentation can quickly produce new discoveries with significantly better experimental response, and how such looping can discover the chemical landscape underlying complex chemical systems.

Coping with complexity: Machine learning optimization of cell‐free protein synthesis

Biological systems contain complex metabolic pathways with many nonlinearities and synergies that make them difficult to predict from first principles. Protein synthesis is a canonical example of such a pathway. Here we show how cell‐free protein synthesis may be improved through a series of iterated high‐throughput experiments guided by a machine‐learning algorithm implementing a form of evolutionary design of experiments (Evo‐DoE). The algorithm predicts fruitful experiments from statistical models of the previous experimental results, combined with stochastic exploration of the experimental space. The desired experimental response, or evolutionary fitness, was defined as the yield of the target product, and new experimental conditions were discovered to have ∼350% greater yield than the standard. An analysis of the best experimental conditions discovered indicates that there are two distinct classes of kinetics, thus showing how our evolutionary design of experiments is capable of significant innovation, as well as gradual improvement. Biotechnol. Bioeng. 2011;108:2218–2228.

Machine learning for drug design, molecular machines and evolvable artificial cells

An artificial cell is a complex chemical system with many components fabricated and assembled in the laboratory. The molecular components can be designed to interlock in a variety of different way to achieve the emergence of minimal life [1][2]. One experimental design is composed of three modules or sub-systems: lipid vesicles, a metabolic system and a cell free expression system. Due to the high number of molecular species and their non-trivial interactions in an artificial cell any prediction of the emerging properties in this high dimensional space of compositions is extremely difficult. Previously we have developed and used a machine learning process Evo-DoE (Evolutionary Design of Experiments) coupled with a robotic workstation for liquid handling to optimize a liposomal drug formulation [3] as well as a cell free expression system for the synthesis of the GFP (green fluorescent protein in vitro) [4]. In addition we have results of vesicle fusion providing a protocol to design a life-cycle for evolvable artificial cells. Now we propose how our technologies could be used to optimize artificial cells.

Machine Learning Optimization of Evolvable Artificial Cells

Abstract An evolvable artificial cell is a chemical or biological complex system assembled in laboratory. The system is rationally designed to show life-like properties. In order to achieve an optimal design for the emergence of minimal life, a high dimensional space of possible experimental combinations can be explored. A machine learning approach (Evo-DoE) could be applied to explore this experimental space and define optimal interactions according to a specific fitness function. Herein an implementation of an evolutionary design of experiments to optimize chemical and biochemical systems based on a machine learning process is presented. The optimization proceeds over generations of experiments in iterative loop until optimal compositions are discovered. The fitness function is experimentally measured every time the loop is closed. Two examples of complex systems, namely a liposomal drug formulation and an in vitro cell-free expression system are presented as examples of optimization of molecular interactions in high dimensional space of compositions [1] , [4] . These represent, for instance, the modules or subsystems that could be optimized by “mixing the protocols” to achieve the high level of sophistication that artificial cells requires. In addition a replication cycle of oil in water emulsions is presented. They represent the container for the artificial cells.

Models of Minimal Physical Intelligence

Abstract The oil droplet is used as a simple model system to study the physical basis of intelligence. The oil droplet is a structure produced by self-assembly. By coupling a chemical reaction in the oil droplet with a physical instability, the droplet displays life-like properties such as autonomous motility, sensory-motor coupling, gradient following and agent-to-agent signaling. By developing this system towards ICT capabilities that include information processing, transmission and transformation, data storage and production, we explore the minimal physicochemical basis of intelligence.

Ribosome synthesis and construction of a minimal cell using a cell-free expression platform

The creation of wet artificial life in the laboratory is a nontrivial challenge for biologists, chemists, and computer scientists (1-4). Such a challenge revolves around the modular integration of complex reactions networks to obtain functional biochemical units able of self-replication, self-reproduction, spatial-temporal control and ultimately open-ended evolution, e.g. minimal and artificial cells (1,5-8). As a step towards building minimal cells, we have developed a cell-free expression system for bacterial ribosome synthesis named iSAT: integrated Synthesis, Assembly and Translation for in vitro construction of Escherichia coli ribosomes (9). The ribosomal RNA, transcribed from its natural operon, selfassembles with ribosomal proteins added to the reaction mixture. Afterwards, in vitro built synthetic ribosomes translate a reporter gene (10,11). Such system is important to design ribosome with new functions, and for the bottom-up construction of a minimal cell.