A protocol is presented for enriching host cell proteins (HCPs) from drug products (DP) and detecting peptides using proteome enrichment beads. The method is demonstrated using an in-house manufactured monoclonal antibody (mAb) drug substance (DS), which is a well-characterized reference material for evaluating and comparing different methods in terms of performance.
Host cell proteins (HCPs) are impurities that can adversely affect therapeutic proteins, even in small quantities. To evaluate the potential risks associated with drug products, methods have been developed to identify low-abundance HCPs. A crucial approach for developing a sensitive HCP detection method involves enriching HCPs while simultaneously removing monoclonal antibodies (mAbs) before analysis, utilizing liquid chromatography-mass spectrometry (LC-MS).
This protocol offers detailed instructions for enriching host cell proteins using commercially available proteome enrichment beads. These beads contain a diverse library of hexapeptide ligands with specific affinities for different proteins. The protocol also incorporates limited digestion and subsequent peptide detection using nano LC-MS/MS. By employing these techniques, HCPs with low abundance can be enriched over 7000-fold, resulting in an impressive detection limit as low as 0.002 ppm. Significantly, this protocol enables the detection of 850 HCPs with a high level of confidence using a NIST mAb. Moreover, it is designed to be user-friendly and includes a video demonstration to assist with its implementation. By following these steps, researchers can effectively enrich and detect HCPs, enhancing the sensitivity and accuracy of risk assessment for drug products.
Host cell proteins (HCPs) are impurities that are released from the cell culture of the host organism and co-purified with monoclonal antibody (mAb)1,2,3,4. Trace levels of HCPs can negatively impact the quality of the drug product5,6,7,8,9,10,11,12,13,14,15, and therefore, a sensitive HCP analysis method is desired to detect HCPs in sub-ppm to ppm levels.
Orthogonal methods can be applied to detect HCPs in low abundance. Enzyme-linked immunosorbent assay (ELISA) is generally used to quantitate overall HCPs, and it can also detect and quantitate individual HCPs if the corresponding antibodies are available16. However, the production of HCP-specific antibodies is time-consuming and labor-intensive. In contrast, liquid chromatography coupled with mass spectrometry (LC-MS) can provide comprehensive information about individual HCPs in mAb drug products and is widely applied for HCP identification4,7,9,10,12,13,14,15,17,18,19,20,21,22,23,24,25,26,27.
Several methods have been developed to detect HCPs with LC-MS/MS, including limited digestion20, filtration17, Protein A deletion21, immunoprecipitation (IP), and ProteoMiner enrichment (PM)18. Most methods aim to reduce the amount of mAb and enrich HCPs prior to LC-MS/MS analysis, thereby decreasing the dynamic range between mAb peptides and HCP peptides. This protocol presents a proteomic sample enrichment method that combines ProteoMiner technology and limited digestion (PMLD)28. The ProteoMiner enrichment principle involves using commercially available proteome enrichment beads containing a diverse library of combinatorial peptide ligands. These ligands specifically bind to proteins on antibody-drug products, allowing for the removal of excess molecules while concentrating low-abundance host cell proteins (HCPs) on their respective affinity ligands. On the other hand, the principle of limited digestion involves using a low concentration of trypsin. This concentration is sufficient to digest low-abundance HCPs but not enough to digest all antibody drug products. This approach enables the recovery and enrichment of digested HCP peptides from the solution.
Compared to filtration methods, the PMLD technique is not limited by the size of the detected HCPs17. Protein A deletion methods are specific to detecting HCPs associated with antibodies21, while immunoprecipitation is restricted to predefined HCPs from a particular cell line (such as the Chinese Hamster Ovary (CHO) cell line), where an anti-HCP antibody was generated4. In contrast, PMLD can be applied to detect HCPs from any drug modules and host cell proteins co-purified with drug products from various cell lines. Additionally, PMLD exhibits better sensitivity compared to the mentioned methods17,18,20,21,24.
This approach can enrich the HCP concentration by 7000-fold and lower the detection limit to 0.002 ppm28. The experimental setup is illustrated in Figure 1.
Abbreviations used in the protocol are listed in Supplementary Table 1.
1. Preparation of solutions and buffers
NOTE: The commercial details of all the reagents are listed in the Table of Materials.
2. Preparation of monoclonal antibody (mAb) solutions
3. Preparation of proteome enrichment beads
4. Protein enrichment
5. Nano LC-MS/MS analysis
6. Data analysis
This protocol presented a sample preparation workflow, termed protein enrichment coupled with limited digestion (PMLD), for the analysis of host cell proteins (HCPs) in a monoclonal antibody (mAb) sample. Figure 1 illustrates the step-by-step procedure of PMLD. The researchers compared the results of HCP analysis using direct digestion (shown in the top panel of Figure 2) and PMLD (shown in the bottom panel of Figure 2). The Total Ion Chromatogram (TIC) profiles indicated that PMLD significantly reduced or eliminated major mAb peptides while allowing the observation of some HCP peptides comparable to mAb peptides. Detailed parameters for nano LC (Table 1) and MS/MS (Table 2) analyses are provided. Additionally, Table 3 displays a summary of identified HCPs associated with the NIST mAb, including information such as accession number, HCP name, species, coverage percentage, PSMs, unique peptides, molecular weight, expected isoelectric point (pI), Mascot score, Sequest HT score, and the number of peptides identified by Mascot and Sequest HT.
Figure 1: Sample preparation workflow of protein enrichment coupled with limited digestion (PMLD). Please click here to view a larger version of this figure.
Figure 2: Total Ion Chromatogram (TIC) profile for HCP analysis of NIST mAb. Top panel: TIC profile for HCP analysis of NIST mAb using direct digestion. Bottom panel: TIC profile for HCP analysis of NIST mAb using PMLD. It is evident from the TIC profile that compared to direct digestion, most of the major mAb peptides have been reduced or eliminated after PMLD, while some peptides belonging to HCPs can be observed and are comparable to mAb peptides. Please click here to view a larger version of this figure.
Table 1: Parameters for nano LC. (A) Loading pump gradient. (B) NC pump gradient. (C) Ion source properties. Please click here to download this Table.
Table 2: Parameters for MS/MS. (A) Full scan properties. (B) MIPS properties. (C) Intensity properties. (D) Charge state properties. (E) Dynamic exclusion. (F) Data-dependent MS2 scan properties. Please click here to download this Table.
Table 3: Representative summary table of identified HCPs associated with NIST mAb analyzed by Proteome Discoverer software. The table includes information such as the accession number, HCP name, species, percentage of coverage, number of peptide spectrum matches (PSMs), number of unique peptides, molecular weight, expected isoelectric point (pI), Mascot score, Sequest HT score, and the number of peptides identified by Mascot and Sequest HT. Please click here to download this Table.
Supplementary Table 1: List of abbreviations used. Please click here to download this Table.
There are two versions of commercially available protein enrichment beads: one with a smaller capacity and the other with a larger capacity (see Table of Materials). Both versions of the enrichment beads contain ten preps in the package. The manufacturer’s instructions suggest that each prep from the small capacity kit can be used to enrich 10 mg of total protein. However, for optimal performance of host cell protein (HCP) enrichment from DS, each prep is good for five DS samples. Therefore, each kit can be used to enrich HCPs from fifty samples. For each prep from the large capacity kit, it is good for the enrichment of HCPs from twenty-five DS samples, allowing the detection of HCPs from a total of two hundred and fifty samples.
Step 3.1 advises against discarding the top or bottom caps, as they will be reused throughout the entire protocol. If beads settle in the top cap, replace them after removing the bottom plug and centrifuging while keeping the top cap on the column. To utilize the bottom cap as a plug, invert it and securely position it at the bottom of the spin column.
A critical step in the procedure is step 4.1, where the bead slurry settles at the bottom of the tube. Therefore, it is crucial to maintain continuous pipetting up and down while transferring the slurry into the monoclonal antibody (mAb) samples. Another critical step is step 4.6, where it is essential to pipette the slurry ten times within the tip while the beads are saturated with elution buffer. The duration of centrifugation in steps 4.3-4.5, 4.7, and 4.10 may vary depending on the amount of accumulated host cell proteins (HCPs). Therefore, it is crucial to check and ensure that the liquids in the tip have been properly spun down before proceeding to the next step. Again, when starting with an initial quantity of 15 mg of antibody-drug product, approximately 30 µg of antibody and enriched host cell proteins (HCPs) can be eluted. To achieve a protein-to-enzyme ratio of 400:1 for limited digestion, 75 ng of trypsin was added. When a lower amount of input material is used, measuring the concentration of the eluted protein sample is necessary. This measurement helps adjust the amount of trypsin that must be added accordingly. Step 4.9 shows a white precipitate after adding the 10% TFA. To prevent any interference from surfactants with the MS signal, it is crucial to check the pH after adding trifluoroacetic acid (TFA) to ensure the complete removal of the detergent from the supernatant.
Following the drying and resuspension of the eluent in water, it is necessary to perform a UV measurement of the peptide mixture. This measurement is crucial because the quantity of peptides loaded onto the column can influence the detection using MS. After evaluation, it was found that approximately 1 µg of the peptide mixture exhibited the best performance in MS. Consequently, this optimal amount needs to be injected into nano LC-MS/MS for further analysis.
The PMLD approach offers several advantages in sample preparation, including sufficient monoclonal antibody (mAb) DS reduction while preserving the majority of low-abundance host cell proteins (HCPs). Unlike immunoprecipitation (IP), this technique does not rely on anti-HCP antibodies, thus eliminating the bias associated with predefined antibodies and reducing the labor and time required for producing anti-HCP antibodies. Furthermore, unlike filtration methods based on molecular weight cut-off17, PMLD enriches HCPs regardless of their size. It can be applied to enrich HCPs from various drug modules, such as antibodies, fusion proteins, ScFv, and more, making it a versatile approach compared to Protein A deletion methods21.
In addition to these advantages, PMLD exhibits a lower detection limit and enables the detection of a greater number of HCPs from NISTmAb, a widely characterized reference standard, compared to other methods. However, it is worth noting that PMLD requires homemade equipment, which limits its application for automation. To expand its usability, replacing certain equipment, such as the tip with frit, with commercially available alternatives is possible. Additionally, performing desalting on 96-well plates can enable higher throughput using this approach. Exploring automation or semi-automation in future experiments can be a logical next step to expand the wider application of PMLD.
The authors have nothing to disclose.
None.
16 G, Metal Hub Needle, 2 in, point style 3 | Hamilton | 91016 | |
Acclaim PepMap 100 C18 trap column (20 cm × 0.075 mm) | Thermo Fisher | 164535 | |
Acetonitrile | Fisher-Scientific | A955 | |
Acetonitrile with 0.1% Formic Acid (v/v), Optima LC/MS Grade | Fisher-Scientific | LS120-4 | |
Amicon Ultra-0.5 Centrifugal Filter Unit | Millipore Sigma | UFC5010 | |
C18 analytical column (0.075 mm × 1.7 μm × 30 cm, 100 Å) | CoAnn Technologies | HEB07503001718I | |
Centrifuge 5424 | Eppendorf | 5405000646 | |
Dithiothreitol (DTT) | Thermo Fisher | A39255 | |
Frit for SPE cartridges, 9.5 mm, 3 mL, 100/pk | Agilent | 12131020 | |
GL-Tip GC | GL Sciences Inc | 7820-11201 | |
in-house mAb | Regeneron | concentration 200 mg/mL | |
Iodoacetamide (30 x 9.3 mg) | Thermo Fisher | A39271 | |
Isopropanol | Fisher-Scientific | 149320025 | |
L-Histidine | Sigma Aldrich | H6034 | |
L-Histidine monohydrochloride monohydrate | Sigma Aldrich | 53370 | |
Methanol | Fisher-Scientific | A456-4 | |
Milli-Q | Millpore | 30035 | |
NanoDrop 2000 | Thermo Scientific | ND-2000 | |
Orbitrap Exploris 480 | Thermo Fisher | BRE725539 | |
Protein LoBind Tube 0.5 mL | Eppendorf (VWR) | 22431064 | |
Protein LoBind Tube 2.0 mL | Eppendorf (VWR) | 22431102 | |
Proteome Discoverer software 2.4 | Thermo Scientific | ||
ProteoMiner Protein Enrichment Large-Capacity Kit | Bio-Rad | 1633007 | |
ProteoMiner Protein Enrichment Small-Capacity Kit | Bio-Rad | 1633006 | |
Sodium deoxycholate (SDC) | Sigma Aldrich | D6750 | |
Sodium lauroyl sarcosinate (SLS) | Sigma Aldrich | L5777 | |
SpeedVac | Labconco | 7970010 | |
Thermomixer R | Eppendorf | 22670107 | |
Trifluoracetic acid (TFA) | Fisher-Scientific | 28904 | |
Trypsin (Sequencing Grade Modified) (5 x 20 ug) | Promega | V5111 | |
Tube Revolver Rotator | Thermo Fisher | 88881001 | |
UltiMate 3000 RSLC nano system | Thermo Fisher | ULTIM3000RSLCNANO | |
UltraPure 1 M Tris-HCl pH 8.0 | Thermo Fisher | 15568-025 | |
Vortex Genie 2 | VWR | 102091-234 | |
Water with 0.1% Formic Acid (v/v), Optima LC/MS Grade | Fisher-Scientific | LS118-4 |