(25b) Multiscale Computer Modeling to Investigate Polymeric Micelle Based Nanocarriers for Anticancer Drug Delivery

Authors: 
Duran, T., University of Connecticut
Costa, A., UConn
Xu, X., Office of Testing and Research, U.S. Food and Drug Administration
Burgess, D., UConn
Chaudhuri, B., University of Connecticut
Purpose:

Typical micelle-based drug delivery carriers are formed from amphiphilic block copolymers which self-assemble into core-shell nanoarchitectures with a hydrophobic core being able to encapsulate drug molecules and a biocompatible hydrophilic shell. An innovative turbulent co-flow platform of water and ethanol (with drug and polymers) was previously developed to continuously process drug loaded micelle nanocarriers. Polymer aggregation is the result of the intermolecular forces among molecules and along with the jet flow characteristics impact to the formation of micelles. However, the underlying mechanism and the detailed effects from material attributes are only partially understood. In order to investigate the underlying mechanism and quantitative estimation of the effect of material attributes and process parameters on the quality of the polymeric micelle formulation, we implemented a multiscale computational approach to study micelle formation using a coaxial turbulent jet flow.

Methods:

All-atom (AA), coarse-grained molecular dynamics (CG-MD), as well as computational fluid dynamics (CFD) simulations have been conducted to not only reveal the effects to material attributes and processing parameters during the micelle formation, but also as parametric case studies for this process. The initial conditions applied into the simulations were based on polymeric micelle experiments in which PEG-PLA (2kD-1.7kD) was used as model drug carrier. The CFD simulation was implemented using Large Eddy Simulation (LES) model in COMSOL Multiphysics incorporating energy equation and high-resolution mesh in mixing area. The MD simulation trajectories and their analysis were carried out using GROMACS package. We applied both CHARMM and MARTINI force-fields for all-atom as well as coarse-grained simulations, respectively. The initial coordinates of models and their force field parameters were generated by CHARMM software and MARTINI forward mapping approach. The steepest decent algorithm was used in energy minimization with a 20 fs time step followed by 10 ns equilibration step in isothermal−isochoric NVT ensemble at temperature of 300 K, then the production runs were performed beyond 1 μs in the NPT ensemble using the Nosé−Hoover thermostat and the Parrinello−Rahman barostat at pressure of 1 atm with constant temperatures at every 20 fs time steps. The simulations were carried out using periodic boundary conditions. Computations were performed in High Performance Center Supercomputer Cluster at the University of Connecticut.

Results:

The self-assembled process of polymeric micelle was studied in the system with and without drug molecules, and in each case, block copolymer was randomly packed in the simulation box with explicit water and ethanol for all simulations. With optimized CG-MD forward mapping strategy and force field parameters incorporating MARTINI standard beads and recent developed S-beads, we observed that polymeric micelles formed successfully and fell into the experimental size range. The simulation of micelles formation and the size distribution the micelles indicate that CHARMM, MARTINI force fields (FFs) can effectively capture the experiment behavior and results . By adjusting Smagorinsky coefficient (Cs) of LES model companied with energy and mass transfer equations, CFD simulations successfully modeled flow patterns and formation temperatures of co-axial turbulent jet flow from experiments, as well as successful comparison and verification. Our CG-MD and CFD simulations are in good agreement with experimental results.

Conclusions:

In this present work, the model predictions provide an understanding of the impact of drug-polymer interactions, drug-polymer ratio, initial organic phase polymer concentration, as well as maintaining the agreement between the computational predictions and experimental findings highlight the power of the multiscale approach to be able to cover the long and short length scale predictions. Moreover, the details of the formation process was revealed in micro, macro, and meso scales by connecting discrete and continuum computational modeling work which are found to be complementary to one another. The multiscale approach used in this work can be effectively utilized as a powerful tool in the process of discovery, development, and optimization of new drug delivery systems in the co-flow continuous processing.

Acknowledgements: FDA Grant# 1U01FD005773-01.

Disclaimer: This article reflects the views of the authors and should not be construed to represent FDA’s views or policies.