Precise and Robust Cell-free Synthesis of Artificial Extracellular Vesicles

Introduction:: There has been a boom in extracellular vesicle (EV) research since their abilities in wound healing and neuroprotection were discovered. These vesicles, which package material like microRNA and proteins, are decorated with membrane proteins from the surface of the parent cell. These proteins allow the extracellular vesicles to target specific cells and in conjunction with the cargo enable them as powerful therapeutics. The origin of the EV determines the native abilities they possess. Since native EVs are harvested from living cells, they have high heterogeneity and low harvesting rates. High heterogeneity means one batch to another may be less/more effective, making dosing extremely difficult and approval from the FDA nearly impossible. The low harvest rates mean the cost and time required to harvest enough for a therapeutic effect is a bottleneck that prevents commercialization. To solve these issues, I will synthesize artificial EVs that mimic native EVs using a cell-free synthetic biology approach¹. Cell-free synthesis works by utilizing the intracellular machinery of E. Coli, however, lysing all of the bacteria in the process¹. This approach will generate the protein of interest from a DNA template, and when mixed with liposomes will anchor the proteins as they are being generated¹. Therefore, my system will utilize the easier generation methods of liposomes and reduce heterogeneity by including only the proteins of interest. With these two factors, my work will be translatable to the therapeutic space for a variety of different diseases.


Materials and Methods:: Sonication was used for generating ~75 nm liposomes. Nanotracking Analysis was used to determine the size, standard deviation, and concentration of liposomes. Inserted proteins were generated by first developing a plasmid through golden gate synthesis, which incorporated a promoter, protein of interest, and terminator. Next, soluble/extracellular domains of various membrane proteins were synthesized using cell-free protein synthesis¹. When the liposomes, cell-free systems, and plasmids were incubated overnight, the proteins were synthesized and inserted into the liposome membrane¹. Protein insertion was validated using flow cytometry, western blotting, ELISA, and super-resolution microscopy. We opted to use a universal detection method by using the HIS tag since only extracellular domains are being generated. Single EV flow cytometry was done using CY5-labeled liposomes and a 525 HIS antibody for protein labeling. Western blot analysis is done both before and after SEC filtration to show insertion. ELISAs were performed by using a HIS capture antibody and capturing fluorescent liposomes that displayed the HIS protein. Success was quantified by looking for fluorescence. Finally, the ONI Super-resolution microscope is used in conjugation with fluorescent liposomes and anti-HIS antibodies. Artificial EV’s function was characterized by a cell migration assay and cell uptake studies. Cell migration is performed by scratching a monolayer of endothelial progenitor cells and adding treatment. The effectiveness of treatment is quantified by the rate of migration. Cell uptake is a measure of the cell-free generated proteins’ ability to be recognized. Fluorescent liposomes are added to the culture and analyzed using confocal microscopy and flow cytometry.

Results, Conclusions, and Discussions:: I have tested and compared three molecular scaffolds to attach proteins to liposomes. Two of the molecular scaffolds, Aquaporin-Z, and transferrin, are self-inserting membrane proteins. One molecular scaffold is Strep-tag conjugation which uses Streptavidin for protein conjugation. I first synthesize various EV proteins, including common tetraspanins and growth factors, using my custom cell-free protein synthesis system^1,2. Next, I quantify the effectiveness and robustness of CD9 attachment to liposomes using the three molecular scaffolds through flow cytometry, western blot analysis, ELISA, and super-resolution microscopy. I find that the AquaporinZ-molecular scaffold showed robust signals of all tetraspanins in western blots. Furthermore, it generates 50-60% of protein-attached liposomes, as shown in single-liposome flow cytometry. The most robust molecular scaffold is then used to attach a broad class of proteins found on native MSC EV to liposomes². These extracellular domains of EV proteins are identified from our previous proteomics results of native MSC EV². The proteins include tetraspanins, growth factors, and galectin. The artificial EVs are then investigated using various assays, including a endothelial progenitor cells (EPC) migration assay (simulates wound-healing), a fibroblast and epithelial cells co-culture test (for targeting specificity), and a RAW 264.7 and THP1 macrophage uptake assay (for immune evasion)³.



My work demonstrates that cell-free protein synthesis can generate human tetraspanins proteins and attach them to the liposome surface, creating artificial EVs. The cell-free approach allows one-step co-synthesis-co-attachment of proteins to liposomes without using living cells¹. As a result, it significantly accelerates the workflow of prototyping artificial EVs. In addition, my approach is highly robust in attaching proteins to liposomes due to the use of the self-inserting transmembrane domains. The artificial EVs will have a broad impact on the translation of EVs as therapeutics.


Acknowledgements (Optional): : I would like to thank the NDSEG for the funding source. Additionally I would like to thank the members of the Cheemeng Tan lab, Aijun Wang lab, and the Randy Carney lab.


References (Optional): : 1. Contreras-Llano LE, Meyer C, Liu Y, et al. Holistic engineering of cell-free systems through proteome-reprogramming synthetic circuits. Nat Commun. 2020;11(1):3138. doi:10.1038/s41467-020-16900-7

2. Clark K, Zhang S, Barthe S, et al. Placental Mesenchymal Stem Cell-Derived Extracellular Vesicles Promote Myelin Regeneration in an Animal Model of Multiple Sclerosis. Cells. 2019;8(12):1497. doi:10.3390/cells8121497

3. Gao K, Kumar P, Cortez-Toledo E, et al. Potential long-term treatment of hemophilia A by neonatal co-transplantation of cord blood-derived endothelial colony-forming cells and placental mesenchymal stromal cells. Stem Cell Research & Therapy. 2019;10(1):34. doi:10.1186/s13287-019-1138-8