DSP applications of the Helia Analyzer

Here we describe a few examples of how downstream processing (DSP) can be improved to produce more protein product per processing cycle. The methods exploit the composition fluctuations of incoming batches to increase the protein production of your processing steps.

The manufacturing of high-value proteins requires expensive downstream processing (DSP) steps, such as extractions and purifications. Examples are the extraction of lactoferrin from bovine milk, the extraction of specific proteins from blood plasma, and the purification of proteins from cell-based production methods.

A fundamental challenge in DSP is to balance productivity and quality, which means maximizing yield and reducing processing times while strictly adhering to regulatory guidelines and purity standards.

Fluctuations are central to the DSP challenge, as the concentrations of proteins and impurities can differ strongly between incoming batches, typically by several tens of percents.

This raises an interesting question: can the fluctuating levels of proteins and impurities, be exploited to increase the yield and quality of your DSP?

The proposed methods are called “Protein-level selective processing” and “Impurity-level selective processing”. The basic concept is to separate incoming batches and process intermediates into two categories: low-protein versus high-protein, or low-impurity versus high-impurity, and to process these accordingly. The concepts are illustrated in the examples below.

Protein-level selective processing

Let’s assume a specific example to illustrate the concept of protein-level selective processing:

A manufacturing plant extracts Lactoferrin (LF) from milk using ion-exchange columns
The incoming milk has an average LF level of 120 mg/L, with a normal distribution of LF levels
The LF levels in incoming milk trucks vary in the range 120±40 mg/L (mean±SD)
The LF levels in milk collection tanks vary in the range 120±10 mg/L (mean±SD)

Three cases are shown below: Tank-level selective processing (50% milk use), Truck-level selective processing (50% milk use) and Truck-level selective processing (100% milk use).

Tank-level selective processing (50% milk use)

The figure shows a case where 50% of the milk is processed for Lactoferrin (LF) extraction and the other 50% is not used for LF extraction. In this example, the selection of milk is based on measurements in milk collection tanks in the factory. Due to the segregation, the average level of LF is higher in the selected milk compared to the incoming milk. The LF-level selective processing gives (128-120)/120 = 7% more LF in the selected milk, compared to the traditional method of LF-extraction without selecting the milk.

Truck-level selective processing (50% milk use)

The figure shows a case where the LF-level selections are done based on measurements of incoming milk trucks. The LF-level selective processing gives (152-120)/120 = 27% more LF in the selected milk, compared to the traditional method of LF-extraction without selecting the milk.

Truck-level selective processing (100% milk use)

The figure shows a case where all milk (100%) is processed for LF extraction. The LF-level selections are done based on measurements of milk trucks. After the separation, the average LF level is 152 mg/L in the high-LF milk, and 88 mg/L in the low-LF milk. Importantly, the respective LF levels have very low spreads (8 mg/L). The low spreads enable the production of more LF per column cycle. Calculations show that the segregation of the milk gives 6.6% more LF per extraction cycle, compared to processing without milk selection.

Impurity-level selective processing

Impurity-level selective processing involves the segregation of incoming batches according to their impurity levels. Assume a manufacturing plant where a specific protein is extracted and purified from a complex biological matrix such as blood plasma, or where a biological medicine is purified after a cell-based production process:

A unit operation uses column separations to remove an impurity from a process intermediate
Assume that incoming batches have an average impurity level of 100 mg/L, with a normal distribution with a standard deviation (SD) of 20 mg/L
The batches are segregated in two groups, based on measurements of their impurity levels: low-impurity batches (<100 mg/L) versus high-impurity batches (>100 mg/L)
Two columns are used for the purification: one column operates with settings optimized for low-impurity plasma; the other column is optimized for high-impurity plasma

The figure illustrates how impurity-level selective processing can help to increase the production of purified protein. After segregation of low-impurity and high-impurity batches, the respective impurity-level distributions have lower spreads: SD of 12 mg/L instead of the original 20 mg/L. The low spreads enable better purifications and allow the production of more protein per column cycle, compared to traditional processing without impurity-level selections.

Protein-level selective processing and impurity-level selective processing exploit concentration fluctuations to increase the protein production of your DSP processing steps. In this way, the Helia Analyzer can significantly increase the protein production of your manufacturing process.

We will be happy to speak with you about protein-level selective processing and impurity-level selective processing. The Helia Analyzer is at your service!

P.S. Please let us know if any information on this page is incorrect or unclear.