June 15, 2009 (Vol. 29, No. 12)

Åsa Åsa Hagner-McWhirter
Maria Winkvist
Lennart Björkesten örkesten
Gunnar Malmquist

New Methodology Can Be Used to Facilitate Host-Cell Protein Determination

There are strong demands from the FDA and EMEA to assure the safety of therapeutic products. Thus, it is becoming increasingly important for biopharmaceutical manufacturers to obtain a deeper understanding of how process conditions affect the quality of the final product with regard to impurities.

Impurities can be related to the process, product, or host cell and may include host-cell proteins (HCP), DNA, viruses, or IgG aggregates. For biopharmaceuticals manufactured using recombinant DNA technology, the majority of impurities encountered are host-cell proteins.

Because HCP can be immunogenic and possibly cause anaphylactic reactions in patients, routine HCP determination is compulsory for all recombinant therapeutics including enzymes/proteins and hormones. ELISA is currently the standard approach for HCP determination, although it is usually performed in combination with other methods such as 1-D SDS-PAGE or IEF-PAGE to give more information about the proteins present.

In this study we used a model system to illustrate the possibilities of using 2-D DIGE (2-D Difference Fluorescence Gel Electrophoresis) to improve the understanding of process conditions in the production of biopharmaceuticals. The 2-D DIGE method is proven to have high sensitivity and specificity, and can be used as a stand-alone method for characterization of HCP profiles without the need for antibodies.

HCP Profiling

Although ELISA, Coomassie-stained SDS-PAGE, and IEF-PAGE are the methods currently used as standard for HCP determination, they each present problems with regard to HCP analysis. While immunoassays such as ELISA have the advantage of specifically detecting immunoreactive proteins, they do not detect weakly reactive proteins or non-immunoreactive proteins. Hence, the FDA is expressing concerns that these assays might not be sufficient to give a complete picture of contamination with HCP.

Immunologic methods are not always easy to work with, as they are arduous to develop, and specific antibodies must be acquired. False negatives may be detected by antibody-based assays due to sample denaturation or steric hindrance causing epitopes to be missed, and false positives may also be detected due to cross-reactivity. In addition, although immunoassays detect the presence of immunoreactive HCP and quantify the total amount, complimentary methods are needed in order to elucidate HCP patterns.

The HCP pattern can be observed and quantitated with 1-D SDS-PAGE or IEF-PAGE, but the quantitation level varies depending on the equipment (e.g., software) available, both resolution and sensitivity are limited, and IEF-PAGE data is difficult to correlate to SDS-PAGE data. Also, these methods lack reliable internal markers for accurate quantitation.

2-D electrophoresis is an established method, commonly used in proteomic research for analyzing complex protein mixtures. 2-D DIGE was developed with the aim of simplifying the process and improving quantitation. 2-D DIGE is an electrophoretic method that allows accurate quantitation of small differences in protein abundance between different samples with high statistical confidence.

The Ettan™ DIGE technology from GE Healthcare allows for the simultaneous co-separation of two different samples, plus an internal standard, on an individual gel. The samples and internal standard are each fluorescently labeled with a different CyDye™ DIGE Fluor dye. Because two samples are multiplexed using the same internal standard on each gel, and each protein spot has its own corresponding internal standard spot used for normalization, all gels can be directly compared, reducing the number of gels required for each experiment.

In contrast to immunoassays, 2-D DIGE detects all proteins, independent of their immunogenic response. Using DeCyder™ 2-D software, 2-D DIGE can detect the smallest possible real differences in protein expression, quantitatively and with statistical certainty. These advantages address the need for quantitative and sensitive HCP profiling.

Reduction of Impurities

Identification and quantification of impurities early in the drug development process is crucial, and an improper investigation of this can lead to serious consequences. Therefore, it is important to investigate manufacturability at an early stage in the process by varying upstream conditions and analyzing the effects on downstream processing (i.e., purification). As shown in Figure 1 there are several upstream conditions (e.g., cell cloning and growth conditions) that may have quantitative and qualitative effects on HCP content in the final product.

In controlled experiments, upstream conditions can be modified, and the downstream effects on critical impurities can be measured. By analyzing the proteome, upstream conditions can be altered to optimize bioprocessing and predict the yield and purity of the target protein. This type of analysis could also be performed on the transcriptome or peptidome to gain an even deeper understanding about the effects of a process change. Using a systematic approach, the acquired knowledge can be used to control bioprocessing on an entirely new level.


Figure 1. Improved bioprocess understanding leads to better process control.

Selecting Optimal Conditions

We cultured E. coli cells expressing histidine-tagged green fluorescent protein (GFP) at 37°C. Some of these samples were cooled to 20°C before IPTG induction and then cultured at 20°C to test different upstream conditions. After induction, six biological replicate samples were taken at each time point and temperature (Figure 2).

Analysis using 2-D DIGE was performed on the samples, and the resulting 2-D protein spot maps of different gels were compared to the internal standard with DeCyder 2-D software. A t-test between the two temperatures showed 438 differentially expressed proteins (0.0001 level of significance). Of these, 130 proteins of interest were picked and identified.


Figure 2. A model system using E. coli cells expressing His-GFP was used. IPTG induction was performed on the cultures at two different temperatures, and the differences were analyzed.

The majority of proteins differentially expressed in E. coli were downregulated at 20°C, possibly due to stress induced by reducing the temperature after culturing. At 37°C, most of the proteins were upregulated over time. The samples taken at 20°C and time point t4 showed similarities to those taken at the early time point and 37°C (Figure 3). A possible explanation for this is that the cultures had adapted to the low temperature and started to produce similar proteins to the 37°C cultures.

Different downstream processing conditions were analyzed by fractionating the collected samples and using either a Capto™ Q anion exchange column or a HisTrap™ IMAC HP column. Different fractions were collected from each column and analyzed by 2-D DIGE. Chromatograms from the two purification methods are shown in Figure 4A. Two examples of protein spot maps of samples cultured at 37°C and fractionated using either ion exchange or affinity chromatography are shown in Figure 4B.

Because all gels in the experiment used the same internal standard, it was possible to link the spot maps of the fractionated samples back to the spot maps of the start material (cultured at 37°C or 20°C). The spot maps showed that several host-cell proteins were still present after IMAC purification. These were likely to be histidine- or tryptophan-containing proteins, which can also bind to the resin. Many variants/isoforms of the histidine-tagged target protein could also be separated and identified. Using information obtained from 2-D DIGE it is possible to reduce certain impurities in the eluate by selecting optimal upstream and downstream conditions.


Figure 3. DeCyder 2-D software analysis shows the principal component analysis plot where data from 37°C cultures (circled in red) is compared to induction at 20°C (circled in blue).


Figure 4. Downstream processing is analyzed by 2-D DIGE.

Summary

Gaining control over purity and yield in biopharmaceutical production is becoming essential for regulatory approval of drugs. Organizations such as the FDA are calling for better process understanding. Systematic analysis is crucial for making improvements in upstream and downstream processes.

The 2-D DIGE method is a promising tool that enables the analysis of optimal growth conditions and purification methods. The use of an internal standard enables the linking of analyses of upstream and downstream changes. As 2-D DIGE runs an in-gel standard with each sample, comparisons among samples are precise, thereby providing reliable detection and quantitation of changes in impurities resulting from process manipulations. 2-D DIGE is an easily validated method for HCP determination that delivers high-quality results. 

Asa Hagner-McWhirter, Ph.D. ([email protected]),
and Maria Winkvist, Ph.D., are scientists, and Lennart Björkesten, Ph.D., and Gunnar Malmquist, Ph.D., are staff scientists at GE Healthcare. Web: www.gelifesciences.com/2DE. Capto, CyDye, DeCyder, Ettan,
and HisTrap are trademarks of GE Healthcare.

Previous articleCustomized Cancer Vaccines Finally (Maybe) Arrive
Next articleLinkMed to Distribute Qiagen’s Genomic HLA Testing Products in Transplantation Market