Protein-excipient Interactions and Protein-Protein Interactions in formulation

Project Information

PIPPI

Grant agreement ID: 675074

DOI

10.3030/675074

Project closed

EC signature date 31 July 2015

Start date 1 January 2016

End date 31 December 2019

Funded under

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Total cost

€ 4 067 745,48

EU contribution

€ 4 067 745,48

Coordinated by

DANMARKS TEKNISKE UNIVERSITET
Denmark

Periodic Reporting for period 2 - PIPPI (Protein-excipient Interactions and Protein-Protein Interactions in formulation)

Reporting period: 2018-01-01 to 2019-12-31

Summary of the context and overall objectives of the project

The increasing number of protein therapeutics on the market and in industrial pipelines drives the need for a better understanding of their formulation to meet emerging trends such as ultra-high protein concentrations, aggregate formation by novel formats, and train specialists in the growing biopharmaceuticals sector. This led to the formation of the ITN network PIPPI (Protein-excipient Interactions and Protein-Protein Interactions in formulation), where 15 Ph.D. students were trained within protein formulation, with an emphasis in biophysical and structural characterization of protein therapeutics.

The consortium were located in Denmark, Sweden, Germany and United Kingdom. The students worked in an international and interdisciplinary network learning to combine systematic investigations of physicochemical behavior of a number of proteins with an in-depth understanding of the molecular interactions behind the macroscopic behavior.

We have published a comprehensive database for diverse protein architectures, which can be interrogated for the prediction of their properties and stability when complemented with a minimum number of experiments. The database contains characterization results from both high throughput methods, structural characterization and in-silico methods. From the 32 articles published so far the synergy between experimental and computational methods and between high-throughput and low-throughput methods are clear.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

The protein molecules for the representative library were selected upon public availability and library diversity. The selected molecules cover a large size range from approximately 4 kDalton all the way up to 150 kDalton. Their isoelectric points range from 4.7 to around 9, with intermediates also represented. Most importantly, the PIPPI protein library covers over 80% of the higher order protein folds found in biologics approved by the FDA.

A systematic study of the overall behaviour of the candidate proteins as a function of pH and ionic strength is almost complete, and will form the basis for further investigations. The protein molecular behaviour has been investigated by structural analyses and the critical formulation attributes have been identified and are being measured over a large space of formulation conditions. Furthermore, the database is online.

The consortium has organised six internal workshops covering formulation development, light scattering, X-ray scattering, nuclear magnetic resonance, molecular dynamics simulations, lyophilization and protein aggregation and data analysis consisting of lectures, exercises and case studies. We organized an international conference on protein stability and aggregation in Heidelberg 2019 https://www.proteininteractions.eu/home/. The ESR students have presented their concepts and results on a large number of international conferences; 32 articles have been published or accepted for publication and at least 30 articles are in preparation.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The large amount of data collected on the PIPPI proteins has enabled us to test if protein sequence and structural data could be used to predict protein stability parameters, such as melting temperature and aggregation temperature using Artificial Neural Networks (ANNs). Consequently, an ANN model was successfully fitted to the full screening data set of monoclonal antibodies and used to predict important protein stability indicators. ANNs compared to classical statistical methods lead to better fit and prediction. This work has been published. We also showed an ANN model could be used to predict storage stability from high throughput measurements of protein biophysical properties and accelerated stability tests.

The massive collaborative efforts in collecting comprehensive data on the same proteins, in the same conditions – stability, modeling, nuclear magnetic resonance, small-angle X-ray scattering, light scattering, rheology to mention some, opens the possibility for unique studies of the interplay between the microscopic atomic structure and the macroscopic overall behavior and stability of the proteins in the library. The data is being deposited in a publicly available database https://pippi-data.kemi.dtu.dk/ which will make it possible for other researchers in the community to benefit from these data in the future. We expect this data will enable us to introduce new tools into the world of biologics for accelerating protein drug development.

PIPPI ESRs, supervisors and advisory board at DTU workshop September 2017