Modeling results summary
Modeling was key to optimize both the structure and the function of our synthetic liposomes, and ultimately enhance our targeted cancer treatment. We used modeling to improve our understanding of the functioning of our liposome, to identify some bottlenecks, and ultimately to design a liposome 2.0 with improved specificity and efficiency. We summarize in this section the model-driven modifications for such liposomes.
Structural optimization of cancer cell anchoring and sensing
First, we built a structure of the split T7 RNAP and used it as a template to determine the optimal length of the linker to promote the assembly of the split T7 RNAP after binding of the two antibodies on HER2. In collaboration with the LAAS laboratory, we applied a mathematical algorithm which samples conformations including disordered regions. Based on these results, we evaluated the optimal length between each predetermined structure (antibodies, transmembrane linker ZipA and the split T7 RNAP) to allow enough mobility and flexibility.
We focused on one of the most used disordered linkers ((GGGS)n), and we determined that a 13 repeats sequence (containing in total 52 amino acids) between the transmembrane part ZipA and the antibodies is optimal, and that a 8 repeats sequence (32 amino acids) is optimal between ZipA and the split T7 RNAP. However, expression of such repetitive sequences could be hard to achieve in a cell-free system. For this reason, we performed the same analysis using a sequence with the same number of amino acids but less repetitions. The conformations obtained validated the disordered characteristic of this sequence, and showed that it was long enough to allow movement without being too flexible and provoke interferences.
Therefore, we have successfully modeled the split T7 RNAP and designed new linkers to activate the production of the anticancer agent by recognizing a tumor marker, HER2. For our liposome 2.0, we will therefore work with the following linkers:
Optimal sequence of the transmembrane linker for T7Nterm-SL-Pertuzumab:
- T7Nterm-SGGGASGGGASGEGGSGPGGSGGGESAAAGSGRLILIIVGAIAIIALLVHGFGASGEGGSGPGGSGGGESAAAGSGGGASGGGASGEGGSGPGGS GGGESAAAG-Pertuzumab
Optimal sequence of the transmembrane linker for Trastuzumab-SL-T7Cterm:
- Trastuzumab-GAAASEGGGSGGPGSGGEGSAGGGSAGGGSGAFGHVLLAIIAIAGVIILILRGSGAAASEGGGSGGPGSGGEGSAGGGSAGGGS-T7Cterm
Figure 1: Possible conformations of the Linker_V2. Red: HER2. Gray from top to bottom: Pertuzumab linked to the Nterm subunit of the T7 RNAP. Green from top to bottom: Trastuzumab linked to the Cterm subunit of the T7 RNAP.
Functional optimization of anticancer drug production
Our second challenge was the optimization of the production of 5-FU by the liposome. In that aim, we built a model to simulate the dynamics of drug production based on a set of ordinary differential equations.
We used this model to design experiments that we performed. Comparison of experimental data to model predictions confirmed inhibition of drug production by DhdR and revealed that the model was more sensitive to 2-HG than observed experimentally. Therefore, we calibrated the biosensing module using the experimental data to improve the predictive capabilities of the model. Using this refined model, we performed a sensitivity analysis to identify the species that control 5-FU production. Based on these results, we optimized the initial concentration of several species (including amino acids, DNA, DhdR, and Tegafur) to boost 5-FU production. With these improvements, we expect the concentration of 5FU to reach the IC50 value reporter in literature within a lifespan of 12 hours, which is compatible with liposome stability in the body.
Therefore, we have successfully modeled the kinetics of our liposomes and found optimal concentrations of each component to improve anti-cancer drug production. To prepare our liposomes 2.0, we will therefore prepare liposomes with the concentrations listed in Table 1.
Table 1: Initial concentrations of each species used to create our liposomes 2.0.
