Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 May 25.
Published in final edited form as: J Pharm Biomed Anal. 2019 Jun 28;174:650–654. doi: 10.1016/j.jpba.2019.06.038

Host cell protein quantification of an optimized purification method by mass spectrometry

Karine Reiter a, Motoshi Suzuki b, Lisa Renee Olano b, David L Narum a,*
PMCID: PMC11127253  NIHMSID: NIHMS1985565  PMID: 31279895

Abstract

Recombinant ExoProtein A (EPA), a detoxified form of Pseudomonas aeruginosa Exotoxin A, is used as a protein carrier in the vaccine field. A scaled manufacturing process, in which EPA was expressed in Escherichia coli, yielded a product that approached or exceeded our upper limit of E. coli host cell protein (HCP) content per human dose. The purification process was redeveloped to reduce HCP levels in the bulk product and HCP content was evaluated by orthogonal methods. Using a platform specific immunoassay, the HCP level from the original purification method was 1,830 ppm (0.18% w/w) while the revised purification process yielded the HCP below the detection limits of the assay. With a 2D/LC-MSE methodology the reference sample from the original process was found to contain 57 unique HCPs at a total level of 37,811 ppm (3.78% w/w). Two lots were tested after purification with the revised process and contained 730 and 598 ppm (0.07% and 0.06% w/w), respectively. To develop a high-throughput MS method, the samples were tested on a 1D/LC-MS/MS. The data sets from the two mass spectrometers correlated well. These improved HCP profiles support implementing the revised purification process for manufacturing the EPA protein carrier and 1D/LC-MS/MS for HCP analysis.

Keywords: ExoProtein A, Host cell proteins, Malaria, Quantitative mass spectroscopy, Vaccine

1. Introduction

Residual host cell proteins (HCPs) are process-related impurities derived from the cell substrate used to produce the target protein. Due to the potential immunogenicity and/or proteolytic activity of these extra proteins, HCPs must be brought down to low levels that ensure clinical safety [1,2]. Reducing HCP levels can also result in a more consistent manufacturing process due to the unknown variability of these other proteins and their amounts [3].

There is no universally available and accepted testing modality for HCP content and no assay standardization within industry. As a result, there is currently no single test or absolute control limits required by regulators during clinical trials and at registration. A recent report has indicated that the most likely range of HCPs in biotherapeutic products reviewed by the FDA is 1–100 ppm [4,5]. The level of acceptable HCPs is reviewed on a case-by-case basis by the regulatory authorities with an allowable HCP level generally higher for vaccines than for drugs with chronic administration [6].

The most common method for determining HCP content of a recombinant protein based therapeutic is by an immunoassay, often in the form of an enzyme-linked immunosorbent assay (ELISA) [2,5,7]. Mass spectrometry has emerged as an orthogonal method to quantify HCPs and it allows for identifying specific HCPs. A 2D-LC/MSE methodology with a data-independent acquisition mode, has been established as a method to identify and quantify proteins within several types of samples [8,9] and HCPs in purification eluate pools, in particular [10,11]. In addition, several reports using single dimension chromatographic separation in combination with data-dependent MS/MS acquisition have been presented as an HCP analytical method having a potentially higher sample throughput [12,13]. At present, there is no consensus on which approach for sample preparation, MS acquisition and data evaluation is best for HCP measurements by mass spectrometry [14].

Recombinant ExoProtein A (EPA) has been used as a carrier protein in the conjugate vaccine field against Staphylococcus aureus type 5 and 8, Shigella, typhoid and malaria [15–18]. It is a detoxified form of Pseudomonas aeruginosa Exotoxin A which is expressed and purified in high yields as a soluble protein from Escherichia coli [19]. Even though the quality and consistency of a bulk EPA was acceptable for our phase 1 clinical trials [18,20] during manufacturing campaigns the E. coli HCP content was marginally out of specifications. To prevent this deviation and avoid reprocessing the material, the purification methods were redeveloped with the aim to reduce the HCP content while maintaining the quality and quantity of purified EPA [21]. The revised process modified the column chromatography steps following a strong anion capture resin (Capto Q). Two mixed mode resins (Mercapto-Ethyl-Pyridine HyperCel™ and Hydroxyapatite) were used in place of a traditional hydrophobic interaction resin (Phenyl Sepharose), a strong anion resin (Q Sepharose) and a size exclusion chromatography step (Fig. 1). In our previous work, we showed that the biochemical and biophysical properties of the EPA protein were not altered between the two purification schemes [21]. In this study, we show that the revised purification process (Process B) achieved the goal of significantly and consistently reducing HCPs to acceptable levels.

Fig. 1. Flow diagram for the purification of EPA.

Fig. 1.

Open in a new tab

The original purification method (Process A) was redeveloped to reduce E. coli host cell proteins (Process B).

2. Material and methods

2.1. Slot blot immunoassay

A quantitative slot blot immunoassay was previously developed for determining the amount of E. coli HCPs in recombinantly expressed proteins [22] and was modified by using IRDye 800CW (LI-COR Biotechnology, Lincoln NE, USA) as the secondary antibody. Briefly, E. coli with no insert was fermented, the cells were isolated and disrupted using a microfluidizer. The soluble fraction from the cell lysate was used for generating a polyclonal antibody reagent. A standard curve was performed by 1:2 dilutions of the E. coli proteins and used to calculate the HCP level in the test sample. Fluorescence was measured at 800 nm with an Odyssey imaging instrument (LI-COR Biotechnology, Lincoln NE, USA).

2.2. HCP analysis by 2D/LC/MSE

The EPA protein was denatured at 0.1% RapiGest (Waters Corporation, Milford MA, USA) and then reduced with 5 mM DTT and alkylated with 10 mM iodoacetamide. Digestion was performed at a 1:40 trypsin:protein ratio overnight at 37 °C.

After digestion, the pH was adjusted to 10 by adding NH4OH for effective trapping on the first-dimension column. Samples were spiked with 50 fmol/μL of pre-digested alcohol dehydrogenase (ADH) standard (Waters Corporation) before injection for quantifying the results.

The sample was loaded onto a XBridge Peptide C18 NanoEase Column at 97% mobile phase A (20 mM ammonium formate, pH 10) and 3% mobile phase B (acetonitrile). A 7-step fractionation using 10%, 13%, 16%, 18%, 20%, 26% and 50% mobile phase B was performed in the first dimension. Sample was injected at a volume to ensure a TIC between 1e7 and 2e7 on the analytical column for each fraction, that is between 1 and 4 u L depending on the sample. The eluent is automatically diluted with 0.1% formic acid (FA) aqueous solution. This online dilution reduces the organic solvent percentage in the eluent and decreases the pH of the mobile phase such that all the peptides eluted from the first dimension can be effectively retained again on the trap column. At the end of the trapping step, peptides retained on the trap column are back-flushed onto the analytical column where a high-resolution RP separation is undertaken at a low pH.

The mobile phases for the second chromatographic dimension were 0.1% FA in water (A) and 0.1% FA in ACN (B). The second-dimension column was a nanoACQUITY UPLC HSS T3 1.8 um column. Separation in the second dimension was performed with a 70-minute gradient of 3%–50% mobile phase B. The column was washed at 85% B and re-equilibrated at 3% B before returning to the next step of fractionation. The MSE data were collected for each fraction during the second-dimension separations.

MS data were collected with a XEVO G2 quadrupole time-of-flight (Q-Tof) tandem mass spectrometer (Waters Corporation) operated in the data independent alternate scanning LC/MS mode (LC/MSE) with MassLynx software. An alternating low collision energy (6 V) and elevated collision energy (ramping from 15 to 35 V) acquisition was used to acquire peptide precursor (MS) and fragmentation (MSE) data. Spectra were collected from m/z 50–1,990 using a scan time of 1 s, a capillary voltage of 3.3 kV, a source temperature 120 °C and a cone voltage of 30 V.

The LC/MSE data was processed using PLGS 2.5 software (Waters Corporation) for HCP identification and protein quantification. For each sample replicate, all the MSE data from each fractionation step was combined into a single file using PLGS software. Processed spectra were searched against the proteome of E. coli strain B/BL21-DE3 (http://www.uniprot.org/taxonomy/469008) with the sequences of EPA and ADH proteins added. The database was reversed and the software searched a concatenation of the target database and the decoy database, incorporating the variable modification of methionine oxidation and the fixed modification of cysteine carbamidomethylation. Proteins with a false positive rate (FPR) of ≤2% and identified by 3 peptides or more per acquisition in at least two of the three replicates were included in the HCP count.

2.3. HCP analysis by 1D/LC-MS/MS

The proteins were digested as described above. Digests were diluted 1:20 and spiked with 50 fmol/μL of pre-digested alcohol dehydrogenase (ADH) standard. LC-MS analysis was conducted using an Orbitrap Fusion Tribrid mass spectrometer (Thermo Electron) equipped with an EASY-Spray Ion Source and an Easy-nLC 1000. Liquid chromatography was carried out on a PepMap RSLC C18 column (size 3 μm, length 25 cm, inner diameter 75 μm) and a PepMap C18 Trap column (size 5 μm, length 2 cm, inner diame- ter 75 μm) operating at a 300 μL/min flow rate with mobile phases A: 98% water + 2% acetonitrile + 0.1% formic acid and B: 98% acetonitrile + 2% water + 0.1% formic acid. After 5 μL of the sample was loaded and the column is washed with 5 uL of the solvent A, the MS/MS data was acquired during the gradient from 0% B to 40% B for 80 min, from 40% B to 80% B for 2 min, and holding at 80% B for 2 min. The mass spectrometer was operated in standard data dependent acquisition mode with the following scan cycle. First, the Orbitrap F™S was used to acquire a high-resolution full MS scan of all ions from m/z 400 to mz/ 1800 at a resolution of 120,000 at m/z 190 at the target value of 4E5. The precursor ions are chosen with monoisotopic precursor selection, multiple charge states, and intensity threshold of 5E3. For the duration of 3 s, most intense precursors were subjected to a fragmentation in the Ion Trap, where the precursor (the target value of 1E4) is isolated with the window of m/z 1 and activated using the activation q at 0.25 and normalized collision energy at 35, then detected at the normal scan rate. Dynamic exclusion was enabled with the duration of 30 s with ±10 ppm mass tolerance. An extensive wash was carried out between each sample run.

The acquired data were analyzed using Proteome Discoverer, version 2.1.1.21 (ThermoFisher Scientific) using the custom database of E. coli strain B/BL21-DE3 with the sequences of EPA and ADH proteins added. Database search was carried out using Sequest HT algorithm, with the precursor mass tolerance at 15 ppm, the fragment mass tolerance at 0.8 Da, incorporating the dynamic modification of methionine oxidation and the static modification of cysteine carbamidomethylation. Proteins were identified with a 1% false discovery rate (FDR) using the Percolator algorithm as implemented in Proteome Discoverer. Proteins identified by 3 peptides or more per acquisition and in at least two of the three replicates were included in the HCP count. The protein concentration was calculated based on the top three peptides of the ADH spiked at 50 fmol.

3. Results and discussion

3.1. Platform slot-blot immunoassay

The accurate assessment and quantification of HCPs is important in the development of recombinantly produced biologics for clinical development. Recombinant EPA is a carrier protein used in chemically conjugated vaccines to effectively enhance poor immunogenicity of the active pharmaceutical ingredient [15–17,20,23]. The EPA used in the early malaria clinical trials and in rabbit toxicology studies has been derived from the Process A purification scheme (Fig. 1). Using a platform-specific immunoblot assay HCP assay, the amount of E. coli HCP was consistently at or above our upper limit for product release and for this reason we redeveloped the purification process. It was established that EPA derived from the original process (Process A) was highly comparable to that of the revised process (Process B) by various biophysical and biochemical assays, including a mass spectrometry analysis of post-translational modifications [21]. With this slot blot immunoblot assay, we demonstrated that EPA purified by Process B compared to Process A was reduced from an average of 0.18% w/w (1830 ppm) to below the sensitivity of the assay.

There are limitations to immunoassays which can be overcome using orthogonal methods such as mass spectrometry. Polyclonal antibodies raised against the HCP population upstream of the bioprocess may not be specific enough to the downstream process and thus, the amount of available antibodies may not be correlated with the amount of HCP in the drug substance [2]. In addition, certain host cell antigens may be less immunogenic, resulting in lower values for the immunoblot assay compared to MS methods [24].

3.2. Mass spectrometry

To support our regulatory filing regarding the process changes with the understanding of the limited sensitivity of the immunoassay, we initially performed two-dimensional liquid chromatography/mass spectrometry on a Xevo G2 QTof mass spectrometer. This 2D–LC/MSE methodology has been well-established for quantitation of host cell proteins [10,11,25]. It is a label-free quantification procedure based on the “top 3” intensity tryptic peptides per protein. The fractionation by 2D-LC allows for better mining of low abundant proteins. First, the digest is loaded onto the first-dimension reversed-phase column at high pH (pH 10) and then eluted in seven step gradients onto the second-dimension column run at low pH (pH 2.5). A gradient is performed on the second column for each step and as the peptides elute from the second-dimension high-resolution separation, a quadrupole time-of-flight mass spectrometer detects the peptides and their fragments by alternating the collision cell energy between a low and an elevated energy (MSE methodology). This 2D-LC method has a run time of approximately 10 h per sample. Following collection of the MSE data from the seven analytical injections, the data are digitally combined and analyzed for the content of E. coli HCPs.

A comparison was performed on EPA purified by Process A and Process B. The 2D–LC/MSE methodology quantified 57 different E. coli proteins with a range of 40 ppm to 5,541 ppm (Fig. 2) for a total of 37,811 ppm (3.78%) HCPs in the Process A material. Only one protein, diaminopimealate decarboxylase (Accession number A0A140N6 V2), was quantified in the EPA from Process B (Table 1 and Fig. 2). The EPA purified by the Process B method contained 730 ppm of this protein and a second purified lot contained 598 ppm.

Fig. 2. HCPs quantitatively identified in EPA.

Fig. 2.

Open in a new tab

Reference standard derived either from Process A (black circles) or Process B (red circle) by 2D/LC-MSE.

Table 1.

HCPs in two lots of EPA purified by Process B quantified by DIA and DDA mass spectroscopya.

Accession number 2D/MSE Lot 1 1D/MS2 Lot 1 2D/MSE Lot 2 1D/MS2 Lot 2
A0A140N6 V2 730 937 598 457
A0A140NFP9* 193 39
A0A140N9K7 58 24
A0A140NBC5 36 28
A0A140NGG2 17
A0A140NCG0* 14
A0A140N7C6 58
Total HCP (ppm) 730 1255 598 607
Open in a new tab
a

HCPs quantified using the 3 top ionizing peptides in at least two of three technical replicates.

*

indicates proteins with 3 peptides in only one of the replicates for the 2D/MSE methodology.

To develop a more high-throughput mass spectroscopy method for the purposes of HCP analysis, the 2D-LC/MSE method was transferred to a 1D/LC-MS/MS method on an Orbitrap Fusion Tribrid mass spectrometer. This data dependent mode of acquisition (DDA) differs from MSE in that no ion transmission window is applied with the first mass analyzer prior to collision induced disassociation. This approach fragments one peptide at a time and lets the software identify the respective peptide [12].

The samples were handled in the same manner for both MS methods and were spiked with 50 fmol ADH as a label-free internal standard. For both methods, proteins were quantified from the top three ionizing peptides and were included in the calculation if seen in two of the three replicates. Fifty-nine HCPs were quantified by 1D/LC-MS/MS with a range of 21 ppm to 3,436 ppm for a total of 31,265 ppm (3.13% w/w) HCPs. The two instruments quantified 38 proteins in common and a regression analysis of those proteins gave an r2 of 0.84 (Fig. 3).

Fig. 3. Scatterplot of the HCPs quantitatively identified by DIA and DDA mass spectroscopy and the association of HCPs identified by both instruments.

Fig. 3.

Open in a new tab

Thirty-eight commonly quantified proteins were used to derive the r2 of 0.84.

Material purified by Process B and analyzed by 1D/LC-MS/MS had six HCPs for a total of 1,255 ppm (0.13% w/w). The second EPA lot purified by Process B had five HCPs quantified at 607 ppm. Two of the proteins (A0A140N9K7 and A0A140NCG0) quantified by the Orbitrap Fusion were also quantified (≥ 3 peptides) in one of the three replicates on the Xevo G2 QTof. Additionally, two proteins quantified by this method in the Process B EPA were also present in the Process A EPA (A0A140NFP9 and A0A140NCG0).

The differences in the limits of quantitation between the two mass spectrometers may relate to the differences in the instrumentations, acquisition methods and data analysis methods, such as a false positive rate of 2% versus a false discovery rate of 1%. Detection of low-abundant HCPs by bottom-up mass spectrometry is limited by both acquisition and bioinformatics components [26]. Beyond the difference in acquisition method (DDA vs DIA), both the Waters and Thermo Fisher instruments were operated with a different peptide mass window that determined which peptides are fragmented. The impact of this, and other, instrument parameters should be mitigated by the fact that identifications are not determined by a single peptide identification. However, some proteins will not produce many peptides using a traditional trypsin digestion protocol which might result in them being missed in the acquisitions. The choices when processing the data through various bioinformatics pipelines is critical to a complete set of identifications. Each tool uses a basic algorithm for scoring potential spectral matches and determining a cutoff for false discoveries. The choice of the cutoff can also be handled by using an additional algorithm that rescores and ranks peptide identifications using multiple statistics from the original search engine results [27]. An example of this approach is the Percolator semi-machine learning algorithm used in the analysis of the DDA data. Each database search/scoring method results in the identification of some unique peptides not determined with a different approach, and the limited statistics available for samples with small numbers of spectra may result in these differences being more profound.

4. Conclusions

The revised purification process for recombinant EPA successfully reduced total HCP levels from 37,811 ppm to 730 ppm (3.78% to 0.0073%) as shown by 2D/LC-MSE. The 2D/LC-MSE and 1D/LC-MS/MS methodologies were shown to yield similar levels of HCP content from bulk EPA derived by the two purification processes. One significant benefit of the 1D/LC-MS/MS is a reduction in the run time from 10 h by 2D/LC-MSE to a run time of 80 min without loss of sensitivity, allowing for a high throughput measurement of HCP content. The ability to detect HCPs by mass spectrometry is limited by the limit of detection (LOD) of the instrument, composition of low-abundant proteins and their ability to produce peptides that are amenable to mass spectrometric acquisition upon proteolytic digestion, and the proper configuration of the bioinformatics pipeline.

As observed by others, immunoblot assays have limitations which can be addressed by orthogonal methods such as mass spectrometry [2,24]. Other strategies for HCP characterization include using immunoblot and ELISA assays to quantify HCP levels with LC-MS/MS methods used solely for identification of the HCPs [28]. It will be important to understand the measurable differences in HCP content for setting specifications for safety when developing recombinant biological products.

Acknowledgements

We thank Weili Dai for her technical support in performing the immunoblot assays. We also thank Vu Nguyen, Martin Burkhardt and Richard Shimp for developing downstream purification methods for the EPA protein. We appreciate the editorial support of Patrick Gorres. This work was supported by the Intra- mural Research Program of the National Institute for Allergy and Infectious Diseases, National Institutes of Health.

References

  • [1].International Conference on Harmonisation, Guidance on specifications: test procedures and acceptance criteria for biotechnological/biological products, Notice. Food and Drug Administration, HHS, Fed Regist 64 (159) (1999) 44928–44935. [PubMed] [Google Scholar]
  • [2].Wang FQ, Richardson D, Shameem M, Host-cell protein measurement and control, Biopharm International 28 (6) (2015) 32–38. [Google Scholar]
  • [3].Graham L, Khrenov A, Risk Assessments and Control Strategies for Host Cell Proteins, CMC Strategy Forum, 2015.
  • [4].Champion K, Madden H, Dougherty J, Shacter E, Defining your product profile and maintaining control over it, part 2: challenges of monitoring host cell protein impurities, Bioprocess Int. 3 (8) (2005) 52–57.29899681 [Google Scholar]
  • [5].Wang X, Hunter AK, Mozier NM, Host cell proteins in biologics development: identification, quantitation and risk assessment, Biotechnol. Bioeng. 103 (3) (2009) 446–458. [DOI] [PubMed] [Google Scholar]
  • [6].Tscheliessnig AL, Konrath J, Bates R, Jungbauer A, Host cell protein analysis in therapeutic protein bioprocessing - methods and applications, Biotechnol. J 8 (6) (2013) 655–670. [DOI] [PubMed] [Google Scholar]
  • [7].Bracewell DG, Francis R, Smales CM, The future of host cell protein (HCP) identification during process development and manufacturing linked to a risk-based management for their control, Biotechnol. Bioeng 112 (9) (2015) 1727–1737. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [8].Mbeunkui F, Goshe MB, Investigation of solubilization and digestion methods for microsomal membrane proteome analysis using data-independent LC-MSE, Proteomics 11 (5) (2011) 898–911. [DOI] [PubMed] [Google Scholar]
  • [9].Silva JC, Gorenstein MV, Li GZ, Vissers JP, Geromanos SJ, Absolute quantification of proteins by LCMSE: a virtue of parallel MS acquisition, Mol. Cell Proteomics 5 (1) (2006) 144–156. [DOI] [PubMed] [Google Scholar]
  • [10].Doneanu CE, Anderson M, Williams BJ, Lauber MA, Chakraborty A, Chen W, Enhanced detection of low-abundance host cell protein impurities in high-purity monoclonal antibodies down to 1 ppm using ion mobility mass spectrometry coupled with multidimensional liquid chromatography, Anal. Chem 87 (20) (2015) 10283–10291. [DOI] [PubMed] [Google Scholar]
  • [11].Farrell A, Mittermayr S, Morrissey B, Mc Loughlin N, Navas Iglesias N, Marison IW, Bones J, Quantitative host cell protein analysis using two dimensional data independent LC-MS(E), Anal. Chem 87 (18) (2015) 9186–9193. [DOI] [PubMed] [Google Scholar]
  • [12].Reisinger V, Toll H, Mayer RE, Visser J, Wolschin F, A mass spectrometry-based approach to host cell protein identification and its application in a comparability exercise, Anal. Biochem 463 (2014) 1–6. [DOI] [PubMed] [Google Scholar]
  • [13].Walker DE, Yang F, Carver J, Joe K, Michels DA, Yu XC, A modular and adaptive mass spectrometry-based platform for support of bioprocess development toward optimal host cell protein clearance, MAbs 9 (4) (2017) 654–663. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Hada V, Bagdi A, Bihari Z, Timari SB, Fizil A, Szantay C Jr., Recent advancements, challenges, and practical considerations in the mass spectrometry-based analytics of protein biotherapeutics: a viewpoint from the biosimilar industry, J. Pharm. Biomed. Anal 161 (2018) 214–238. [DOI] [PubMed] [Google Scholar]
  • [15].Fattom A, Schneerson R, Watson DC, Karakawa WW, Fitzgerald D, Pastan I, Li X, Shiloach J, Bryla DA, Robbins JB, Laboratory and clinical evaluation of conjugate vaccines composed of Staphylococcus aureus type 5 and type 8 capsular polysaccharides bound to Pseudomonas aeruginosa recombinant exoprotein A, Infect. Immun 61 (3) (1993) 1023–1032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [16].Mai NL, Phan VB, Vo AH, Tran CT, Lin FY, Bryla DA, Chu C, Schiloach J, Robbins JB, Schneerson R, Szu SC, Persistent efficacy of Vi conjugate vaccine against typhoid fever in young children, N. Engl. J. Med 349 (14) (2003) 1390–1391. [DOI] [PubMed] [Google Scholar]
  • [17].Passwell JH, Harlev E, Ashkenazi S, Chu C, Miron D, Ramon R, Farzan N, Shiloach J, Bryla DA, Majadly F, Roberson R, Robbins JB, Schneerson R, Safety and immunogenicity of improved Shigella O-specific polysaccharide-protein conjugate vaccines in adults in Israel, Infect. Immun 69 (3) (2001) 1351–1357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [18].Talaat KR, Ellis RD, Hurd J, Hentrich A, Gabriel E, Hynes NA, Rausch KM, Zhu D, Muratova O, Herrera R, Anderson C, Jones D, Aebig J, Brockley S, MacDonald NJ, Wang X, Fay MP, Healy SA, Durbin AP, Narum DL, Wu Y, Duffy PE, Safety and immunogenicity of Pfs25-EPA/Alhydrogel(R), a transmission blocking vaccine against Plasmodium falciparum: an open label study in malaria naive adults, PLoS One 11 (10) (2016), e0163144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [19].Fass R, van de Walle M, Shiloach A, Joslyn A, Kaufman J, Shiloach J, Use of high density cultures of Escherichia coli for high level production of recombinant Pseudomonas aeruginosa exotoxin A, Appl. Microbiol. Biotechnol 36 (1) (1991) 65–69. [DOI] [PubMed] [Google Scholar]
  • [20].Sagara I, Healy SA, Assadou MH, Gabriel EE, Kone M, Sissoko K, Tembine I, Guindo MA, Doucoure M, Niare K, Dolo A, Rausch KM, Narum DL, Jones DL, MacDonald NJ, Zhu D, Mohan R, Muratova O, Baber I, Coulibaly MB, Fay MP, Anderson C, Wu Y, Traore SF, Doumbo OK, Duffy PE, Safety and immunogenicity of Pfs25H-EPA/Alhydrogel, a transmission-blocking vaccine against Plasmodium falciparum: a randomised, double-blind, comparator-controlled, dose-escalation study in healthy Malian adults, Lancet Infect. Dis 18 (9) (2018) 969–982. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Burkhardt M, Reiter K, Nguyen V, Suzuki M, Herrera R, Duffy PE, Shimp RJ, MacDonald NJ, Olano LR, Narum DL, Assessment of the impact of manufacturing changes on the physicochemical properties of the recombinant vaccine carrier ExoProtein A, Vaccine (In press). 10.1016/j.vaccine.2018.09.037. [DOI] [PMC free article] [PubMed]
  • [22].Zhu D, Saul AJ, Miles AP, A quantitative slot blot assay for host cell protein impurities in recombinant proteins expressed in E. Coli, J. Immunol. Methods 306 (1–2) (2005) 40–50. [DOI] [PubMed] [Google Scholar]
  • [23].Qian F, Aebig JA, Reiter K, Barnafo E, Zhang Y, Shimp RL Jr., Rausch KM, Jones DS, Zhu D, Lambert L, Mullen GE, Narum DL, Miller LH, Wu Y, Enhanced antibody responses to Plasmodium falciparum Pfs28 induced in mice by conjugation to ExoProtein A of Pseudomonas aeruginosa with an improved procedure, Microbes Infect. 11 (3) (2009) 408–412. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Zhu-Shimoni J, Yu C, Nishihara J, Wong RM, Gunawan F, Lin M, Krawitz D, Liu P, Sandoval W, Vanderlaan M, Host cell protein testing by ELISAs and the use of orthogonal methods, Biotechnol. Bioeng 111 (12) (2014) 2367–2379. [DOI] [PubMed] [Google Scholar]
  • [25].Schenauer MR, Flynn GC, Goetze AM, Profiling the effects of process changes on residual host cell proteins in biotherapeutics by mass spectrometry, Biotechnol. Prog 29 (4) (2013) 951–957. [DOI] [PubMed] [Google Scholar]
  • [26].Tabb DL, Vega-Montoto L, Rudnick PA, Variyath AM, Ham AJ, Bunk DM, Kilpatrick LE, Billheimer DD, Blackman RK, Cardasis HL, Carr SA, Clauser KR, Jaffe JD, Kowalski KA, Neubert TA, Regnier FE, Schilling B, Tegeler TJ, Wang M, Wang P, Whiteaker JR, Zimmerman LJ, Fisher SJ, Gibson BW, Kinsinger CR, Mesri M, Rodriguez H, Stein SE, Tempst P, Paulovich AG, Liebler DC, Spiegelman C, Repeatability and reproducibility in proteomic identifications by liquid chromatography-tandem mass spectrometry, J. Proteome Res 9 (2) (2010) 761–776. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [27].Spivak M, Weston J, Bottou L, Kall L, Noble WS, Improvements to the percolator algorithm for Peptide identification from shotgun proteomics data sets, J. Proteome Res 8 (7) (2009) 3737–3745. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [28].Ahluwalia D, Dhillon H, Slaney T, Song H, Boux H, Mehta S, Zhang L, Valdez A, Krishnamurthy G, Identification of a host cell protein impurity in therapeutic protein, P1, J. Pharm. Biomed. Anal 141 (2017) 32–38. [DOI] [PubMed] [Google Scholar]

RESOURCES