My HCP antibodies show a reaction towards a drug substance band in my Western blot.
Is it possible that it is the drug substance homolog purified from the CHO cell line?
Can you identify if there are any HCPs present in this band, and can you relatively quantify the amount of this HCP?
Yes, it is possible to both identify and quantify the HCPs in a drug substance band. You do this by mass spectrometric peptide mapping, by including heavy peptides.
I recommend the following procedure for identification of host cell protein:
- Separate the drug substance sample by SDS-PAGE. NOTE! It is very important that you use the same or similar conditions as used for the SDS-PAGE for the Western blot. You should use a stain optimized for compatibility with mass spec analysis.
- When you have your SDS-PAGE, you should excise the protein band of interest from the gel. Digest it using trypsin and analyze it by nano-LC MS/MS peptide analysis for protein identification.
- You can now search the data against a database. The database should contain both the drug substance sequence and the sequences of the potential host cell homologs. Alternatively, search for the entire host cell proteome.
This analysis identifies the proteins above a certain threshold in the drug substance band.
I usually see HCPs around 100 ppm with great results. Some HCPs I can even detect at as low as 2.5 ppm.
Things to try if the test does not identify any HCPs
To increase the sensitivity it can be very beneficial if you separate as much material as possible in the SDS-PAGE run as well as in nano-LC-MS/MS.
If you do not identify any host cell proteins, try estimating the detection limit. You do this by spiking in one or more protein standards in different known amounts. Make sure that the molecular weight of the standards is the same as the drug substance band. In addition, the spike-in must occur before separation by SDS-PAGE.
Procedure for accurate quantification of identified HCPs
When you have identified one or more host cell proteins, you can quantify them by spiking in heavy peptides after enzymatic digestion. Their sequences should be unique to the drug substance and the host cell protein, respectively.
I find that the quantification becomes more accurate if the amount of heavy peptides in the drug substance and the host cell protein is close to 1:1. It can also be relevant to test different parameters such as linearity and sensitivity.
Challenges you may encounter when the amount of host cell protein is low compared to the drug substance
When you have low amounts of a HCP in high amounts of drug substance, the following factors may influence the results:
- The dynamic range – the maximum difference in signal between the drug substance and the HCP detectable by MS.
- The homology between proteins. In many cases, this is less critical because homology at the sequence level does not necessarily mean that the tryptic peptides used for analysis are identical.
Here is an example: Human and bovine albumin have the same molecular weight and 76% of the amino acid sequences are the same. When you compare the tryptic peptides generated from these sequences, 59% (49/83 and 48/82 peptides in human and bovine albumin, respectively) are unique to either species. Therefore, there is a high chance that you may identify unique peptides from both species in the MS-analysis (if they are present).