Molekulare Modellierung und Multiskalenmodellierung der Auflösung pharmazeutischer Wirkstoffe

• Antragsteller: Professor Dr.-Ing. Heiko Briesen; Professor Dr. Karsten Reuter
• Fachliche Zuordnung: Theoretische Chemie: Elektronenstruktur, Dynamik, Simulation ; Chemische und Thermische Verfahrenstechnik
• Förderung: Förderung von 2011 bis 2022
• Projektkennung: Deutsche Forschungsgemeinschaft (DFG) - Projektnummer 210064482

• Erstellungsjahr: 2022

Zusammenfassung der Projektergebnisse

The bioavailability of the active pharmaceutical ingredients (APIs) is governed by the dissolution properties of the drug. Predicting the dissolution properties of a new drug candidate within the human body long before entering the clinical phase is an important step in lowering the cost and shortening the time of drug development. Therefore, this project aims at a protocol for the prediction of dissolution rates of APIs that supports decisions already during in-silico drug development. The predictions of the multiscale approach are based on molecular structure and on crystal packaging. In case the latter is not available, in silico predictions about the crystal packing of single-component, as well as co-crystals can be made by means of computationally cheap, yet accurate atomistic models obtained from the developed ML approach. The dissolution is first investigated in a MD framework. The dissolution of a three-dimensional crystal with real habit is simulated on MD time scale. Sticky dummy atom are introduced into the simulation box to guarantee constant undersaturation. Then, rate constants for the kMC-simulations are obtained from the MD trajectories. For this purpose, molecules belonging to the crystal are classified into different edge or face categories and rate constants are calculated for each face/edge type by counting the corresponding transition events as well as the possibilities for these transitions to occur. Length and time scales of kMC simulations are significantly extended compared to the MD simulations, thus crystal face displacement velocities can be calculated. In a next step, continuum simulations predict the dissolution rates of the API on a macroscopic level. The face displacement rates of the kMC simulations are used as face-specific boundary conditions and the equations of mass transport are solved numerically. However, a sensitivity analysis for all parameters describing the dissolution shows, that e.g. for aspirin the dissolution in stagnant water is strongly diffusion limited and therefore almost insensitive to the dissolution rates. If the dissolution occurs in a flow, such as the drug-delivery in the human body, the diffusion limitation is overcome. With a Mach-Zehnder interferometer setup with an integrated flow cell, it was shown that the extent of the diffusion zone at the surface directed to the flow decreases with increasing inflow velocity and the dissolution rate increases. Thus, in this setup, the diffusion limitation can be overcome, allowing the validation of the kMC displacement rates. The elaborated procedure was successfully tested on the dissolution of an aspirin crystal in stagnant water, where the simulation result are in good agreement with experimental data. MD simulations of the dissolution of the 1:1 co-crystal paracetamol : oxalic acid and of the influence of salts on the growth of a-glycine, respectively, show that the method is also applicable to multicomponent like APIs with hydrates, salts and co-crystals, which are intended to improve solubility.

Projektbezogene Publikationen (Auswahl)

• The suitability of classical force fields for the molecular modeling of diffusion-limited crystal dissolution processes. International Workshop Molecular Modeling and Simulation: Natural Science Meets Engineering, 2013
  Greiner, M., Elts, E., Schneider, J., Reuter, K. & Briesen, H.
  
• Data filtering for effective analysis of crystal–solution interface molecular dynamics simulations. Journal of Chemical Theory and Computation, 2014; 10(4): 1686-1697
  Elts, E., Greiner, M. & Briesen, H.
  
• Dissolution study of active pharmaceutical ingredients using molecular dynamics simulations with classical force fields. J. Cryst. Growth, 2014;M 405: 122-130
  Greiner, M., Elts, E., Schneider, J., Reuter, K. & Briesen, H.
  
• Efficient calculation of microscopic dissolution rate constants: The aspirin-water interface. J. Phys. Chem. Lett., 2014; 5: 3859-3862
  Schneider, J. & Reuter, K.
  
• Insights into Pharmaceutical Nanocrystal Dissolution: A Molecular Dynamics Simulation Study on Aspirin. Mol. Pharmaceutics, 2014; 11: 3009-3016
  Greiner, M., Elts, E. & Briesen, H.
  
• Thermodynamics of surface defects at the aspirin/water interface. J. Chem. Phys., 2014; 141: 124702
  Schneider, J., Zheng, C. & Reuter, K.
  
• In silico prediction of dissolution rates of pharmaceutical ingredients. Chem. Phys. Lett., 2016; 662: 52-55
  Dogan, B., Schneider, J. & Reuter, K.
  
• Multiscale modeling of aspirin dissolution: From molecular resolution to experimental scales of time and size. CrystEngComm, 2016; 18(28): 5302-5312
  Greiner, M., Choscz C., Eder C., Elts, E. & Briesen, H.
  
• Predicting dissolution kinetics for active pharmaceutical ingredients based on their molecular structure. Cryst. Growth Des., 2016; 16(7): 4154-4164
  Elts, E., Greiner, M. & Briesen, H.
  
• Bridging from nano- to macroscale for crystal growth and dissolution prediction. Proc. 20th International Symposium on Industrial Crystallization, 2017
  Elts, E., Greiner, M. & Briesen, H.
  
• In silico prediction of growth and dissolution rates for organic molecular crystals: a multiscale approach. Crystals, 2017; 7(10): 288
  Elts, E., Greiner, M., & Briesen, H.
  
• Mapping Materials and Molecules. Acc. Chem. Res., 2020; 53: 1981-1991
  Cheng, B., Griffiths, R., Wengert, S., Kunkel, C., Stenczel, T., Zhu, B., Deringer, V. L., Bernstein, N., Margraf, J. T., Reuter, K. & Csányi, G.
  
• Application of image analysis methods to enhance crystal shape identification and face indexing. In Proc. 21st International Symposium on Industrial Crystallization, 2021
  Luxenburger, F., Elts, E., & Briesen, H.
  
• Data-efficient machine learning for molecular crystal structure prediction. Chem. Sci., 2021; 12: 4536-4546
  Wengert, S., Csányi, G., Reuter, K. & Margraf, J. T.
  
• Influence of Monovalent Salts on α-Glycine Crystal Growth from Aqueous Solution: Molecular Dynamics Simulations at Constant Supersaturation Conditions. The Journal of Physical Chemistry B, 2021; 125(42): 11732-11741
  Elts, E., Luxenburger, F., & Briesen, H.
  
• The suitability of classical force fields for the molecular modeling of co-crystal dissolution processes. In Jahrestreffen der ProcessNet Fachgruppe Kristallisation, 2021
  Luxenburger, F., Lüttich, R., Elts, E., & Briesen, H.
  
• A Hybrid Machine Learning Approach for Structure Stability Prediction in Molecular Co-crystal Screenings. J. Chem. Theory Comput., 2022; 18: 4586-4593
  Wengert, S., Csányi, G., Reuter, K. & Margraf, J. T.