A new step towards the use of syncell technology in healthcare?
How do you adapt a technology invented in the laboratory to real-world applications? In a recent article published in Advanced Biology, Professor Avi Schroeder’s research team tackled several bottlenecks in the manufacture of protein-producing cell mimics to meet the criteria for therapeutic use. His team optimised an existing manual production method and proposed a solution, using robotics and machine learning, to automate the production process, thereby reducing processing time, increasing the production and consistency of these cell mimics and improving their quality control. A step towards preclinical and clinical applications?
Also known as bottom-up synthetic biology, synthetic cell (syncell) research aims to understand how cells work by trying to mimic from scratch the complex processes, structures and functions of living cells.
Several challenges in manufacturing cell mimics for therapeutic purposes
For Noga Sharf-Pauker and Shanny Ackerman, the co-authors of this publication, syncell research holds great technological potential, particularly for therapeutic purposes. The controllability and programmability of these man-made microparticles inspired by living cells make them possible candidates for producing insulin in vivo, eliminating breast cancer cells, supporting tissue regeneration processes and more. But before reaching the preclinical stage, the production of these microparticles must meet a number of criteria.
“When we explored in vivo applications, we realised that we couldn’t find a production method suitable for therapeutic use: capable of manufacturing protein-producing synthetic cells quickly and in large quantities, while offering low batch-to-batch variability. So we took a close look at an existing production process and optimised it,” explained Shanny Ackerman.
New solutions to optimise and automate the production process
In their paper, they present a method for optimising conventional pipetting and vortexing steps to prepare giant unilamellar vesicles (GUVs). Their strategy reduces processing time and manual labour, and allows higher GUV activity and quantity to be obtained. In addition, they demonstrate that with their method, the solutions needed to assemble them can be prepared in advance in large volumes and stored for at least one month.
In addition, they propose solutions for replacing the main manual steps with automated ones, thereby ensuring a reproducible method. “The existing methods involve a lot of manual work, so the results vary from one laboratory to another, as they depend on the person who prepared the solution, their level of skill and the calibration of the instruments,” added Shanny.
- Solution preparation: GUV assembly requires a large number of stock solutions that need to be mixed in the correct ratio. The team tested the effectiveness of an automated liquid handler system, LiHa. Despite its high price, the robotic arm proved effective in preparing solutions, reducing time and improving consistency.
- Emulsification: “This step is the main bottleneck in scaling up GUV production,” stressed Noga. The authors demonstrated great innovative thinking skills by adapting a device that was not designed for this purpose: the tissue dissociator. Present in many laboratories working on in vivo experiments, this device extracts living cells from organs. “Our target application was far from the original use of the instrument, but we thought that if it could extract living single cells from organs, perhaps it could also help us generate synthetic cells,” continued Noga. Properly calibrated, the tissue dissociator ensured uniform emulsification, enabling a 30-fold increase in production scale while maintaining consistency and functionality.
- Characterisation and quality control: Although not linked to production, a robust and reliable characterisation tool is essential for refining and optimising the process and guaranteeing the reproducibility and robustness of the production method for therapeutic purposes. Noga and Shanny designed a machine learning-based characterisation method that proved more reliable than traditional software.
What next?
“We believe that the AI-powered automation strategy we are proposing can be of real benefit to other laboratories. It also lays the ground for therapeutic applications,” emphasised Noga.
Noga and Shanny hope to see syncell technologies reach the clinical stage. Currently, the two researchers plan to explore the in vivo aspects and to prolong the activity of protein-producing synthetic cells, while their laboratory plans to use their method for wider applications, in food, for example.
Orginal publication
Scaling Up Synthetic Cell Production Using Robotics and Machine Learning Toward Therapeutic Applications
Noga Sharf-Pauker, Ido Galil, Omer Kfir, Gal Chen, Rotem Menachem, Jeny Shklover, Avi Schroeder, and Shanny Ackerman
Advanced Biology, 2025