Investigation and prediction of protein precipitation by polyethylene glycol using quantitative structure–activity relationship models

Hämmerling, Frank; Effio, Christopher Ladd; Andris, Sebastian; Kittelmann, Jörg; Hubbuch, Jürgen

doi:10.1016/j.jbiotec.2016.11.014

Cited by 12 publications

(7 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This indicates that the molecular weight, or the overall surface charge of proteins is not the only factor influencing PEG-induced precipitation. This is consistent with the investigation of Haemmerling et al [ 36 ]. They showed, with the help of quantitative structure activity relationship models (QSAR), that surface characteristics, such as charge distribution, amount of hydrophobic patches, and their distribution on the surface, might play an important role for the precipitation behavior of proteins as well.…”

Section: Discussionsupporting

confidence: 94%

Precipitation of complex antibody solutions: influence of contaminant composition and cell culture medium on the precipitation behavior

Großhans

Suhm

Hubbuch

2019

Bioprocess Biosyst Eng

Self Cite

View full text Add to dashboard Cite

Preparative protein precipitation is known as a cost-efficient and easy-to-use alternative to chromatographic purification steps. This said, at the moment, there is no process for monoclonal antibodies (mAb) on the market, although especially polyethylene glycol-induced precipitation has shown great potential. One reason might be the highly complex behavior of each component of a crude feedstock during the precipitation process. For different investigated mAbs, significant variations in the host cell protein (HCP) reduction are observed. In contrast to the precipitation behavior of single components, the interactions and interplay in a complex feedstock are not fully understood yet. This work discusses the influence of contaminants on the precipitation behavior of two different mAbs, an IgG1, and an IgG2. By spiking the mAbs with mock solution, a complex feedstock could successfully be mimicked. Spiking contaminants influenced the yield and purity of the mAbs after the precipitation step, compared to the precipitation behavior of the single components. The mixture showed a decrease in the contaminant and mAb solubility. By re-buffering the mock solution prior to spiking, special salts, small molecules like amino acids, vitamins, or sugars could be depleted while larger ones like HCP or DNA were still present. Therefore, it was possible to distinguish the influence of small molecules and larger ones. Hence, mAb–macromolecular interaction could be identified as a possible reason for the observed higher precipitation propensity, while small molecules of the cell culture medium were identified as solubilisation factors during the precipitation process.

show abstract

Section: Discussionsupporting

confidence: 94%

Precipitation of complex antibody solutions: influence of contaminant composition and cell culture medium on the precipitation behavior

Großhans

Suhm

Hubbuch

2019

Bioprocess Biosyst Eng

Self Cite

View full text Add to dashboard Cite

show abstract

“…The similarity of the values determined for the mAb at pH 7.5 and pH 8.5 supports this assumption. Furthermore, Hämmerling et al confirmed this assumption, but pointed out the influence of additional factors, such as protein shape and other surface characteristics [27]. …”

Section: Resultsmentioning

confidence: 99%

“…Sim et al generalized the model on the basis of hydrodynamic radii [26]. Quantitative structure–activity relationship (QSAR) modeling was used to estimate the parameters of the Cohn equation [27]. Anyhow, all these models have difficulty dealing with small molecular weight proteins and are not capable to cover variations in the initial protein concentration.…”

Section: Introductionmentioning

confidence: 99%

Water on hydrophobic surfaces: mechanistic modeling of polyethylene glycol-induced protein precipitation

Großhans

Wang

Hubbuch

2018

Bioprocess Biosyst Eng

Self Cite

View full text Add to dashboard Cite

For the purification of biopharmaceutical proteins, liquid chromatography is still the gold standard. Especially with increasing product titers, drawbacks like slow volumetric throughput and high resin costs lead to an intensifying need for alternative technologies. Selective preparative protein precipitation is one promising alternative technique. Although the capability has been proven, there has been no precipitation process realized for large-scale monoclonal antibody (mAb) production yet. One reason might be that the mechanism behind protein phase behavior is not completely understood and the precipitation process development is still empirical. Mechanistic modeling can be a means for faster, material-saving process development and a better process understanding at the same time. In preparative chromatography, mechanistic modeling was successfully shown for a variety of applications. Lately, a new isotherm for hydrophobic interaction chromatography (HIC) under consideration of water molecules as participants was proposed, enabling an accurate description of HIC. In this work, based on similarities between protein precipitation and HIC, a new precipitation model was derived. In the proposed model, the formation of protein–protein interfaces is thought to be driven by hydrophobic effects, involving a reorganization of the well-ordered water structure on the hydrophobic surfaces of the protein–protein complex. To demonstrate model capability, high-throughput precipitation experiments with pure or prior to the experiments purified proteins lysozyme, myoglobin, bovine serum albumin, and one mAb were conducted at various pH values. Polyethylene glycol (PEG) 6000 was used as precipitant. The precipitant concentration as well as the initial protein concentration was varied systematically. For all investigated proteins, the initial protein concentrations were varied between 1.5 mg/mL and 12 mg/mL. The calibrated models were successfully validated with experimental data. This mechanistic description of protein precipitation process offers mathematical explanation of the precipitation behavior of proteins at PEG concentration, protein concentration, protein size, and pH.

show abstract

“…Nevertheless, the following model approaches have been made for precipitation. To begin with, correlations describing equilibrium and solubility of proteins have been found [40,58] and extended for the change in solubility in presence of PEG molecules [32,34,36,59]. Further, empirical models like quantitative structure-activity relationship (QSAR) are applied to predict the discontinuity point of proteins in presence of PEG molecules [35,36].…”

Section: Modeling Of Precipitationmentioning

confidence: 99%

“…To begin with, correlations describing equilibrium and solubility of proteins have been found [40,58] and extended for the change in solubility in presence of PEG molecules [32,34,36,59]. Further, empirical models like quantitative structure-activity relationship (QSAR) are applied to predict the discontinuity point of proteins in presence of PEG molecules [35,36]. Moreover, model approaches derived from a mechanistic model approach for crystallization are described also for precipitation [60][61][62][63][64][65].…”

Section: Modeling Of Precipitationmentioning

confidence: 99%

Accelerating Biologics Manufacturing by Modeling: Process Integration of Precipitation in mAb Downstream Processing

Lohmann

Strube

2020

Processes

View full text Add to dashboard Cite

The demand on biologics has been constantly rising over the past decades and has become crucial in modern medicine. Promising approaches to cope with widespread diseases like cancer and diabetes are gene therapy, plasmid DNA, virus-like particles, and exosomes. Due to progress that has been made in upstream processing (USP), difficulties arise in downstream processing and demand for innovative solutions. This work focuses on the integration of precipitation using a quality by design (QbD) approach for process development. Selective precipitation is achieved with PEG 4000 resulting in an HCP depletion of ≥80% respectively to IgG. Dissolution was executed with a sodium phosphate buffer (pH = 5/50 mM) reaching an IgG recovery of ≥95%. However, the central challenge in process development is still an optimal process design, which is transferable for a broad molecular variety of new products. This is where rigorous modeling becomes vital in order to generate digital twins to support early-stage process development and reduce the experimental overhead. Therefore, a model development and validation concept for construction of a process model for precipitation is also presented.

show abstract

Investigation and prediction of protein precipitation by polyethylene glycol using quantitative structure–activity relationship models

Cited by 12 publications

References 48 publications

Precipitation of complex antibody solutions: influence of contaminant composition and cell culture medium on the precipitation behavior

Precipitation of complex antibody solutions: influence of contaminant composition and cell culture medium on the precipitation behavior

Water on hydrophobic surfaces: mechanistic modeling of polyethylene glycol-induced protein precipitation

Accelerating Biologics Manufacturing by Modeling: Process Integration of Precipitation in mAb Downstream Processing

Contact Info

Product

Resources

About