From protein structure to an optimized chromatographic capture step using multiscale modeling | JoVE Visualize

Area of Science:

Biopharmaceutical Process Development
Computational Chemistry
Chromatography

Background:

Optimizing biopharmaceutical purification processes is challenging due to limited process understanding, necessitating extensive experimentation.
In silico techniques like mechanistic or data-driven modeling can enhance process understanding, leading to more cost-effective and time-efficient optimization.

Purpose of the Study:

To develop and validate a multiscale modeling strategy integrating Quantitative Structure-Property Relationship (QSPR) and Mechanistic Models (MM) for optimizing cation exchange (CEX) chromatography.
To reduce experimental effort in biopharmaceutical process development by predicting chromatographic behavior from protein structure.

Main Methods:

Developed QSPR models using protein structural characteristics to predict physicochemical behavior and retention volumes.
Integrated QSPR-derived parameters into Mechanistic Models (MM) to predict chromatograms.
Experimentally determined retention profiles for six proteins across various pH conditions to train and validate models.
Optimized the CEX capture step using the integrated modeling approach.

Main Results:

QSPR models accurately predicted retention volumes and characteristic charge for training proteins.
For unseen proteins, the model predicted retention peak differences within 0.2% relative to the gradient length.
The integrated QSPR-MM approach successfully optimized the CEX capture step, yielding results consistent with experimental methods.
Model parameter confidence analysis identified viable process conditions, with one matching experimental optimization outcomes.

Conclusions:

The multiscale modeling approach effectively reduces experimental requirements for biopharmaceutical process optimization.
Integrating QSPR and MM provides a powerful tool for predicting and optimizing chromatographic purification steps.
This strategy enables the identification of initial process conditions, streamlining downstream optimization efforts.