Site-Specific Immobilization of Antibodies as a Platform Approach to Enable the Targeted Capture and Effective Removal of “Problematic” Host Cell Proteins (HCPs) from Complex Bioprocess Streams, Even at Sub-ppm Levels: Chinese Hamster Ovary (CHO) Phospholipase B-Like 2 (PLBL2)

Site-Specific Immobilization of Antibodies as a Platform Approach to Enable the Targeted Capture and Effective Removal of “Problematic” Host Cell Proteins (HCPs) from Complex Bioprocess Streams, Even at Sub-ppm Levels: Chinese Hamster Ovary (CHO) Phospholipase B-Like 2 (PLBL2) | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 12 June 2025 V2 Latest version Share on Site-Specific Immobilization of Antibodies as a Platform Approach to Enable the Targeted Capture and Effective Removal of “Problematic” Host Cell Proteins (HCPs) from Complex Bioprocess Streams, Even at Sub-ppm Levels: Chinese Hamster Ovary (CHO) Phospholipase B-Like 2 (PLBL2) Authors : Michael E. Dolan 0009-0009-4497-8725 [email protected] , Alexander Tedeschi , Sheldon F. Oppenheim , and Zhaohui Sunny Zhou Authors Info & Affiliations https://doi.org/10.22541/au.173372831.15992718/v2 804 views 300 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Removal of host cell proteins (HCPs) during biotherapeutics manufacturing is vital for ensuring patient safety and biotherapeutic stability and supply. Yet, this goal remains exceptionally challenging for some HCPs. For example, phospholipase B-like 2 (PLBL2) from Chinese hamster ovary (CHO) cells (the workhorse for therapeutic protein production) plagues engineers by evading typical purification strategies. New tools have emerged to aid HCP removal, but technologies directed at specific HCP species are still nascent, and an urgent, unmet need remains. Herein, we present a platform approach for the targeted removal of specific, challenging HCPs – even those present at sub-ppm levels. Using CHO PLBL2 as a model, we site-specifically modify and immobilize polyclonal antibodies directed against CHO PLBL2, which exists as multiple proteoforms (e.g., size and charge variants). The immobilized antibodies retain their antigen binding, enabling capture and clearance of CHO PLBL2 from an array of bioprocess streams, including IgG4 and IgG1 antibodies. Although centered on CHO PLBL2, our approach should be broadly applicable to numerous other HCPs across the increasingly diverse biotherapeutic landscape. Additionally, CHO PLBL2 recovered from the polyclonal antibodies exhibits multiple molecular size variants, opening the door to further characterization to identify other proteoforms. This insight can, in turn, guide purification development, even in processes without custom affinity anti-HCP steps. KEYWORDS: Phospholipase B-like 2 (PLBL2), host cell protein (HCP), downstream processing, affinity resin, goat antibody, antibody immobilization, proteoform, size and charge variants INTRODUCTION The removal of Chinese hamster ovarian (CHO) host cell proteins (HCPs) is a key objective of downstream processing for recombinant therapeutic proteins. As a broad class of proteins, HCPs exhibit diverse biophysical properties, which makes their separation from therapeutic proteins an arduous process, often requiring multiple orthogonal chromatography steps to meet industry and regulatory standards for residual HCP levels. Separation difficulties have primarily been attributed to noncovalent (e.g., charge-based or hydrophobicity-based) interactions between therapeutic proteins and HCPs (1,2). Historically, improving separation has relied on brute-force experimentation, such as screening mobile phase modifiers to disrupt putative therapeutic protein-HCP interactions (2,3). While such approaches have achieved lower HCP levels in the polished therapeutic, the nature of the putative protein-protein interactions and the identities of participating HCP species remained poorly understood. Advances in mass spectrometry and other analytical techniques in recent years have enabled the detection and identification of numerous HCPs across an array of CHO bioprocess streams (4-9). Although most HCP species present at the start of downstream processing are readily removed via chromatography or other purification techniques, numerous “problematic” species routinely resist separation (10,11). Moreover, while many co-purifying HCPs are innocuous, some exhibit harmful effects. For example, phospholipase B-like 2 (PLBL2) from Chinese hamster ovary (CHO) copurified at levels as high as about 330 ppm with Genentech’s monoclonal IgG4 antibody lebrikizumab and triggered immunogenic reactions in about 90% of subjects in a Phase III clinical trial (12). This finding necessitated an overhaul of the downstream process for lebrikizumab and delays to further clinical trials. Chromatographic and surface plasmon resonance spectroscopy studies have demonstrated the preferential binding of CHO PLBL2 to IgG4s over other IgG isotypes (13), which renders CHO PLBL2 removal a particular concern during the development of such therapeutics. Enzymatic HCPs – such as cathepsin D, lipoprotein lipase, and liposomal acid lipase, among others (14-20) – have been implicated in the excipient degradation of some drug substances and products, even when individual species are present at part-per-million (ppm, or ng of an individual HCP species per mg of therapeutic protein) or sub-ppm levels (19,21). By deleteriously impacting excipients (e.g., polysorbates), these enzymatic HCPs threaten the stability, shelf life, and clinical and potential commercial supplies of such products. Finally, although protein therapeutics recombinantly expressed in CHO cells have historically been at center stage, efforts at identifying and removing challenging HCPs from other host organisms for emerging modalities (e.g., adeno-associated virus (AAV) and lentiviral vectors for gene therapy) are gaining momentum and increased scrutiny (22-24). Given the potential for significant impact to drug stability, supply continuity, and patient safety, a robust suite of purification techniques to eliminate HCPs is paramount (11,25). In the last several years, new technologies have been brought to market to begin addressing this unmet need. For example, novel peptide ligands conjugated to chromatography resins (e.g., LigaGuard TM ) have been developed for both the targeted capture and removal of HCPs and as competitive alternatives to Protein A affinity resins (26-29). Additionally, resins which target specific, problematic HCP species are in development (e.g., CaptureSelect TM custom affinity resins from Thermo Fisher Scientific). However, platforms such as CaptureSelect TM require significant ligand screening and optimization to achieve suitable affinities for their target antigens and, consequently, many of these resins are still in their infancy and therefore not yet commercialized. Similarly, while LigaGuard TM exhibits the capability to remove a broad spectrum of HCPs, its ligands do not enable the targeted removal of individual impurities of interest, such as CHO PLBL2 or other “problematic” HCPs. Furthermore, as alluded to above, recent analytical efforts have focused singularly on HCP identification, while deeper exploration of the fundamental protein chemistry of HCPs has been overlooked. For example, it is widely recognized that therapeutic proteins (e.g., mAbs) expressed by mammalian cells possess myriad proteoforms, which manifest as diverse molecular size and charge variants, glycosylation, and numerous other attributes (30-35). This diversity in molecular architecture imparts biophysical diversity to the therapeutic protein proteoforms, which can promote (or hinder) the ability to isolate and remove (or enrich) these isoforms during downstream processing. By extension, because HCPs are endogenously expressed by the same hosts which express our biotherapeutics, we expect that HCPs also exhibit proteoforms diversity. Moreover, we expect that the presence of HCP proteoforms directly contributes to challenges with their separation, which has been confirmed by this and ongoing studies. Because of this diversity, chromatography resins which recognize and bind to numerous epitopes are most suited to effectively capturing and removing HCPs (and their expected variants). To that end, we report herein a platform approach using polyclonal antibodies to enable the targeted capture and removal of challenging HCPs – even those present even at sub-ppm levels in complex bioprocess streams. Using CHO PLBL2 as a proof of concept, polyclonal antibodies directed against CHO PLBL2 are chemo-enzymatically and site-specifically modified at their native Fc N-glycans to install azide handles, as previously described by our group (36). After undergoing click chemistry, the antibodies are readily immobilized to various resins; in our case, the azide-activated IgGs are conjugated to a dibenzocyclooctyne (DBCO)-functionalized biotin, followed by further coupling onto streptavidin-coated magnetic beads via the biotin-streptavidin interaction. We show that the immobilized IgGs retain their antigen binding capability and successfully capture and remove CHO PLBL2 from a diverse array of purified therapeutic mAbs (including IgG4 and IgG1), even when CHO PLBL2 is present at concentrations at least as low as 0.1 ppm. We also demonstrate that even from highly complex matrices (e.g., harvest cell culture fluid (HCCF), which is comprised of thousands of endogenous HCPs), clearance of CHO PLBL2 is substantial, and the therapeutic proteins achieve high recoveries. Finally, we show that, after recovering our purified therapeutic protein, CHO PLBL2 which was bound by the immobilized polyclonal antibodies can also be stripped from the resin and recovered for quantitation and for systematic and deep characterization to provide unprecedented insight into CHO PLBL2’s fundamental protein chemistry (e.g., proteoforms), such as the multiple molecular weight isoforms of CHO PLBL2 detailed in this work. This finding opens the door to further characterization to identify other proteoforms. Moreover, this insight can, in turn guide further purification development, especially for removing challenging HCPs, even in processes without custom affinity anti-HCP steps. To the best of our knowledge, ours is the first report of both the development of a custom affinity resin of this design, the subsequent demonstration of the removal of CHO PLBL2 (or any challenging CHO HCP) in this manner, and its application to the capture and identification of HCP proteoforms. Notably, the present work represents the first part in a series of studies which utilize site-specifically modified and immobilized anti-CHO PLBL2 antibodies for the affinity capture (and, as seen in the second part, subsequent detailed characterization) of CHO PLBL2 (37). Finally, our approach should be broadly applicable to enable the removal of numerous other difficult-to-remove HCPs across the increasingly diverse biotherapeutic landscape. MATERIALS AND METHODS Materials Goat anti-CHO PLBL2 polyclonal antibodies (GPLB-65ALY), mouse monoclonal anti-CHO PLBL2 antibodies (MPLB-65ALY-4D5), and CHO-S PLBL2 (E.C. 3.1.1; UniProt G3I6T1; AG65-0365-Z) were from Immunology Consultants Laboratory (Portland, OR, USA). Pembrolizumab (HY-P9902) was from MedChemExpress (Monmouth Junction, NJ, USA). Protein concentrations were determined via UV-Vis absorption and extinction coefficients of 1.4 mL mg -1 cm -1 for the goat anti-CHO PLBL2 polyclonal IgGs and mouse anti-CHO PLBL2 monoclonal IgG (at 280 nm; supplied by the vendor), 2.0 mL mg -1 cm -1 for the CHO-S PLBL2 (at 280 nm; supplied by the vendor), and 1.41 mL mg -1 cm -1 for pembrolizumab (at 280 nm; supplied by the vendor). UV-Vis spectra for the goat anti-CHO PLBL2 polyclonal IgGs, mouse anti-CHO PLBL2 monoclonal IgG, CHO-S PLBL2, and pembrolizumab are shown in Figure S.1 through Figure S.4, respectively. The GlyCLICK® Azide Activation kit to modify up to 2 mg of IgG (L1-AZ1-200) was from Genovis AB (Cambridge, MA, USA and Lund, Sweden) and contained the following materials: immobilized GlycINATOR® (EndoS2; E.C. 3.2.1.96; UniProt T1WGN1) microspin column, β-1,4-galactosyltransferase (GalT Y289L; E.C. 2.4.1.22; UniProt P08037), UDP-GalNAz, and 20X concentrate (0.5 M) of Tris-buffered saline (TBS, pH 7.4). DBCO-PEG 4 -biotin (CCT-A105-5) and high-capacity streptavidin beads (CCT-1497-5) were from Vector Laboratories (formerly Click Chemistry Tools; Newark, CA). The concentrations of DBCO-PEG 4 -biotin samples were determined via UV-Vis absorption and extinction coefficient of 20,000 M -1 cm -1 (at 307 nm; supplied by the vendor). The structure and UV-Vis spectra for DBCO-PEG 4 -biotin are shown in Figure S.5 through Figure S.6, respectively. The protein concentrator (0.5 mL, 30 kDa MWCO; 88502), NuPAGE 4X LDS sample buffer (NP0007), NuPAGE 4-12% Bis-Tris SDS-PAGE gel (NP0326BOX), NuPAGE 20X MES SDS running buffer (NP0002), NuPAGE 10X sample reducing agent (NP0004), DynaMag TM -2 magnet (12321D), bovine serum albumin (BSA; 23209), 20% sodium dodecyl sulfate (SDS) solution (BP1311-200), Novex TM IEF protein gel (pH 3-7; EC66452BOX), Novex TM IEF anode buffer (50X concentrate; LC5300), Novex TM IEF cathode buffer (pH 3-7; 10X concentrate; LC5370), Novex TM IEF sample buffer (pH 3-7; 2X concentrate; LC5371), Invitrolon TM PVDF Filter Paper Sandwich (LC2005), Xcell SureLock® Mini-Cell (EI0001), and Novex XCell II TM blot module (EI9051) were from Thermo Fisher Scientific (Waltham, MA, USA). Protein concentrations for the BSA were determined via UV-Vis absorption using an extinction coefficient of 0.67 mL mg -1 cm -1 (at 280 nm; supplied by the vendor); the UV-Vis spectra are shown in Figure S.7. Precision Plus Protein TM standard (1610373) was purchased from Bio-Rad Laboratories (Hercules, CA, USA). One-Step Blue® protein gel stain (21003-1L) was from Biotium (Fremont, CA, USA). Tris base (4109-02), hydrochloric acid (5619-02), glacial acetic acid (9522-05), sodium hydroxide (5000-03), polysorbate 80 (4125-04), and polysorbate 20 (4116-04) were from JT Baker (part of Avantor; Radnor, PA, USA). Formic acid (94318-250ML) was from Honeywell (Charlotte, NC, USA). The Criterion TM PowerPac Universal power supply (1656019) was from Bio-Rad Laboratories (Hercules, CA, USA). IEF fixing solution (2X concentrate; F7264) was from Sigma-Aldrich (St. Louis, MO, USA). IEF marker (pH 3-10; 39212.01) was from Serva Electrophoresis GmbH (Heidelberg, Germany). Methanol (A935-4) was from Fisher Scientific (Waltham, MA, USA). Donkey anti-mouse (H&L) polyclonal antibody peroxidase conjugated (610-703-002) was from Rockland Immunochemicals (Pottstown, PA, USA). ECL TM Western blotting detection reagents (RPN2106) were from Cytiva (Marlborough, MA, USA). Disposable square petri dishes (25378-045) were from VWR (part of Avantor; Radnor, PA, USA). Dulbecco’s phosphate-buffered saline (10X Concentrate; 14200-075) was from Gibco Waltham, MA, USA). Dry milk (M0841) was from Lab Scientific bioKEMIX (Danvers, MA, USA). Formic acid (94318-250ML) was from Honeywell (Charlotte, NC, USA). The 12-230 kDa Wes separation module (SM-W004) was from Protein Simple (San Jose, CA, USA) and contained the following materials: capillary cartridges (each containing 25 capillaries); 12 – 230 kDa pre-filled plates; 10X sample buffer; wash buffer; and lyophilized DTT, fluorescent master mix, and biotinylated ladder. The anti-mouse detection module (DM-002) was also from Protein Simple and contained the following materials: anti-mouse HRP conjugate secondary antibody, Luminol-S, peroxide, and antibody diluent. The hamster PLBL2 enzyme-linked immunosorbent assay (ELISA) kit (MBS564202) was from MyBioSource (San Diego, CA, USA) and included the following components: antibody-coated ELISA micro plate, detection antibody (100X concentrate), HRP-conjugated streptavidin (100X concentrate), calibrator, diluent solution, wash solution (20X concentrate), chromogen-substrate solution (3,3′,5,5′- tetramethylbenzidine dihydrochloride, TMB), and stop solution (0.3 M sulfuric acid). The CHO PLBL2 ELISA assay control was purified CHO-3E7 PLBL2 (AG65-0324) from Immunology Consultants Laboratory, diluted to a concentration 300 ng/mL. The diluent solution was from the hamster PLBL2 ELISA kit. Buffer Exchange of Goat Anti-CHO PLBL2 Polyclonal Antibody Prior to use, the protein concentrator was rinsed using 1X TBS (pH 7.4) according to vendor instructions. The contents of four vials of goat anti-CHO PLBL2 IgG (4 x 0.5 mg; supplied as a lyophilized product) were reconstituted using water into a total volume of 1 mL, achieving an IgG concentration of 2 mg/mL (or 13 µM). Immunology Consultants Laboratory supplies the goat anti-CHO PLBL2 IgG in phosphate-buffered saline plus a non-protein stabilizer. Therefore, the goat anti-CHO PLBL2 IgG was buffer exchanged prior to use in the chemo-enzymatic workflow. The goat anti-CHO PLBL2 IgG underwent concentration via centrifugation at 5,000 x g using a MySpin 12 mini centrifuge (75004081; Thermo Scientific, Waltham, MA), followed by dilution by ~2X using 1X TBS (pH 7.4) and additional concentration by ~2X. This process was repeated for a total of 6 exchanges, after which the goat anti-CHO PLBL2 IgG was recovered from the protein concentrator. The protein concentration was determined via UV-Vis absorption at 280 nm using a Lunatic microfluidic UV-Vis spectrophotometer from Unchained Labs (Pleasanton, CA, USA). Measurements were obtained using Lunatic Client software (version 5.1.0.213), while analyses were performed using Lunatic Analysis software (version 5.1.0.185). All measurements were performed in triplicate. Absorbances were normalized by the software to a 1 cm path length. About 1.8 mg of goat anti-CHO PLBL2 IgG (at 6.5 mg/mL or 43 µM) were recovered. Trimming the Native Fc N-Glycan via EndoS2 About 1.8 mg of goat anti-CHO PLBL2 IgG were added to the immobilized GlycINATOR® (EndoS2) microspin column, which had been pre-equilibrated using 1X TBS (pH 7.4). The column was incubated at room temperature (about 22°C) for 2 hours with end-over-end mixing. Trimmed goat anti-CHO PLBL2 IgG was recovered in a 1.5 mL Eppendorf tube via centrifugation at 1,000 x g using a MySpin 12 mini centrifuge. The EndoS2 column was washed three times using 100 µL of 1X TBS (pH 7.4) and centrifuged at 1,000 x g after each wash. Column effluents from the initial recovery and the washes were pooled in a single 1.5 mL Eppendorf tube and the protein concentration of the pool was determined. About 1.5 mg of goat anti-CHO PLBL2 IgG (at 2.6 mg/mL or 17 µM) were recovered for a yield of 85%. Azide Activation via GalT About 1.5 mg of goat anti-CHO PLBL2 IgG were mixed with 7 µL of buffer additive, UDP-GalNAz (reconstituted using 40 µL of 1X TBS, pH 7.4), and GalT. The mixture was incubated using a MultiTherm heating/cooling shaker (H5000-HC; Benchmark Scientific, Sayerville, NJ) at 30°C for 19 hours while agitating at 1,000 rpm and shielding from light. Excess UDP-GalNAz was removed after azide activation following the buffer-exchange procedure describe above; a total of 8 exchange were performed using 1X TBS (pH 7.4), after which the goat anti-CHO PLBL2 IgG was recover ed from the protein concentrator and the protein concentration was determined. About 1.3 mg of azide-activated goat anti-CHO PLBL2 IgG (at 4.2 mg/mL or 28 µM) was recovered for a yield of 90%. Biotinylation via Site-Specific Conjugation to DBCO-PEG 4 -Biotin DBCO-PEG 4 -biotin (5 mg, 6.7 µmol) was reconstituted using 670 µL of water to a final concentration of 10 mM. About 1.3 mg of azide-activated anti-PLBL2 IgG was mixed with 10 mM DBCO-PEG 4 -biotin at a molar ratio of 50:1 (DBCO-PEG 4 -biotin:azide) and IgG concentration of 1.0 mg/mL (or 6.7 µM) in 1X TBS (pH 7.4). The reaction contents were incubated using a MultiTherm heating/cooling shaker for 22 hours at 25°C while agitating at 1,000 rpm and shielding from light. Excess DBCO-PEG 4 -biotin was removed after conjugation following the buffer-exchange procedure described above; a total of 8 exchanges were performed, after which the biotinylated goat anti-CHO PLBL2 IgG was recovered. About 1.1 mg of IgG (at 3.1 mg/mL or 7.3 µM) were recovered for a yield of 82%. Immobilization to Streptavidin Magnetic Beads To prepare for immobilization, high-capacity streptavidin magnetic beads were buffer exchanged into 1X TBS (pH 7.4) according to the following: 5 mL of beads were resuspended (10 mg of beads per mL of solution) and transferred in equal-volume aliquots to four clean 1.5 mL Eppendorf tubes, which were placed onto the DynaMag TM -2 magnet to magnetize and separate the beads from solution. The supernatant from each tube was discarded and replaced with 1,000 µL of 1X TBS (pH 7.4), into which the beads were resuspended. The tubes of beads were placed back onto the DynaMag TM -2 magnet to magnetize and separate the beads from solution. The supernatants were discarded, and the process was repeated four more times. After discarding the final supernatants, the beads in each tube were resuspended into 250 µL of 1X TBS (pH 7.4) and combined into a single tube. The tube was placed back onto the DynaMag TM -2 magnet to magnetize and separate the beads from solution, and the supernatant was discarded. The beads were then resuspended a final time in 267 µL of 1X TBS (pH 7.4) to achieve a concentration of ~150 mg of beads per mL of solution. According to the vendor, the high-capacity streptavidin beads are capable of binding >12 nmol of biotin per mg of bead. About 1.1 mg of biotinylated anti-CHO PLBL2 IgG were mixed with the streptavidin beads at a molar ratio of 50:1 (streptavidin:biotin) and IgG concentration of 1.0 mg/mL (or 6.7 µM). The coupling contents were incubated using a MultiTherm heating/cooling shakerfor 25 hours at 25°C while agitating at 1,000 rpm and shielding from light. Following immobilization, the beads were washed as described above to remove any potentially uncoupled goat anti-CHO PLBL2 IgG. Washes were performed according to the following: 3 bead volumes (BVs; i.e., 3 x 300 µL) of 50 mM Tris (pH 8.0), 3 BVs of 50 mM Tris with 2 M sodium chloride (pH 8.0), and 3 BVs of 50 mM Tris (pH 8.0). After washing, the beads were resuspended in 1.1 mL of 1X TBS (pH 7.4) to achieve an effective immobilized IgG concentration of 1.0 mg/mL (or 6.7 µM). SDS-PAGE and Densitometric Analysis To enable SDS-PAGE under reducing conditions, IgG samples were mixed with a 3X concentrate of NuPAGE LDS sample buffer containing 50 mM DTT. For lanes assigned to samples containing a known concentration of goat anti-CHO PLBL2 IgG, 2 µg of IgG were loaded. For lanes assigned to supernatant and wash samples from the post-immobilization wash steps, 10 µL of sample were loaded. All samples were heated at 70°C for 15 minutes to promote denaturation. Precision Plus Protein TM standard (10 µL) and IgG samples were loaded onto a NuPAGE 4-12% Bis-Tris SDS gel. Electrophoresis was conducted using an XCell SureLock® Mini-Cell (Thermo Fisher Scientific, EI0001) connected to a Criterion TM Cell power supply (Bio-Rad Laboratories, 1656019) by running at 200 V using NuPAGE MES SDS running buffer. Protein bands were stained by incubating with One-Step Blue® protein gel stain at room temperature (about 22°C) overnight and destaining in water for 2 hours. Bands were visualized using a GS-900 Calibrated Densitometer equipped with Image Lab 6.1 (Bio-Rad Laboratories). Bands were visualized using the Coomassie Brilliant Blue R-250 setting, employing a transmissive scanning mode and red filter. Band intensities for the biotinylated heavy chain and unmodified light chain for the immobilization load material and post-immobilization supernatant were quantitated using ImageJ version 1.53e (National Institutes of Health; Bethesda, MD, USA). The band intensities were used to calculate the immobilization efficiency. Figure S.8 shows the SDS-PAGE gel for the immobilization load material, post-immobilization supernatant, and post-immobilization washes. Case Study 1: Removal of Recombinant CHO-S PLBL2 from Monoclonal IgG1 Antibody (mAb A) Expression and Purification of mAb A A humanized monoclonal IgG1 antibody (mAb A) was expressed in CHO-K1 cells via suspension culture, clarified to remove cells and cellular debris, purified, buffer exchanged into PBS (pH 7.2), and stored at ≤-65°C according to internal procedures. The UV-Vis spectra for mAb A are shown in Figure S.9. Removal of Recombinant CHO-S PLBL2 from mAb A Binary mixtures of CHO-S PLBL2 and mAb A were generated by spiking CHO-S PLBL2 into solutions of mAb A to achieve relative PLBL2 concentrations of 10 ppm, 1 ppm, and 0.1 ppm according to the following procedure. First, two serial dilutions were produced: the first containing 25 µg/mL of CHO-S PLBL2 with 10,000 µg/mL of mAb A in PBS (pH 7.2), and the second containing 2.5 µg/mL of CHO-S PLBL2 with of 10,000 µg/mL mAb A in PBS (pH 7.2). To produce the 10 ppm CHO-S PLBL2 mixture, the second serial dilution was further mixed with a mAb A stock solution and PBS (pH 7.2) to achieve a total mAb A content of 1.5 mg and CHO-S PLBL2 content of 15 ng in a total volume of 3 mL. The 1 ppm and 0.1 ppm binary mixtures were prepared similarly to achieve total mAb A contents of 15 mg and 150 mg, respectively, and CHO-S PLBL2 contents of 15 ng each in a volume of 3 mL each. A 0 ppm sample (i.e., containing no CHO-S PLBL2) was prepared by mixing mAb A stock solution and PBS (pH 7.2) to achieve a total mAb A content of 150 mg in a total volume of 3 mL, mimicking the conditions of the 0.1 ppm sample but without CHO-S PLBL2. The capture and removal of CHO-S PLBL2 proceeded by incubating the immobilized goat anti-CHO PLBL2 IgG with CHO-S PLBL2 at sub-saturating conditions: 10 µg (or 66.7 picomol in 10 µL of bead solution) of immobilized goat anti-CHO PLBL2 IgG were mixed with each CHO-S PLBL2-mAb A mixture to achieve a molar ratio of ~0.003 moles of CHO-S PLBL2 to 1 mole of immobilized IgG. The solution was mixed end over end for 18 hours at room temperature (about 22°C) while shielding from light. Following incubation, 1 mL of the solution was added to a clean 1.5 mL Eppendorf tube, which was placed onto the DynaMag TM -2 to magnetize and separate the beads from solution. The supernatant from the Eppendorf tube was collected into a separate 15 mL conical tube and replaced with another 1 mL of the post-incubation solution. This process was repeated until all supernatant from the incubation had been separated from the beads and collected. The beads were then washed using 4 x 100 µL of PBS (pH 7.2), 3 x PBS with 2 M sodium chloride (pH 6.8), and 3 x 100 µL of Tris-buffered saline (pH 7.4). Bound CHO-S PLBL2 was stripped from the immobilized anti-CHO PLBL2 IgG using 3 x 15 µL of 2% (w/v) SDS. Capillary Western Blot Analysis Residual CHO-S PLBL2 content of the load material, supernatant, wash, and elution samples was analyzed via capillary Western blot. The analysis was performed using the Wes (Protein Simple,004-600) using a 12 – 230 kDa separation module and the anti-mouse detection module. Samples were prepared either undiluted or diluted (using PBS (pH 7.2)) by mixing with the 5X fluorescent master mix (containing DTT) and were denatured by heating at 95°C for 5 minutes, after which they were centrifuged at 5,000 x g using a MySpin 12 mini centrifuge and stored on ice. A standard curve of CHO-S PLBL2 was prepared in PBS (pH 7.2) by diluting the commercial, recombinant CHO-S PLBL2 stock to concentrations of 3.125 ng/µL, 0.3125 ng/µL, and 0.03125 ng/µL. All CHO-S PLBL2 standards were prepared with 100 ng/µL BSA to function as a carrier protein. The CHO-S PLBL2 standards were mixed with the 5X fluorescent master mix (containing DTT) and denatured by heating at 95°C for 5 minutes, after which they were centrifuged at 5,000 x g using a MySpin 12 mini centrifuge and stored on ice. The CHO-S PLBL2 concentrations for each standard were chosen to enable the delivery of 50 pg, 25 pg, 10 pg, 5 pg, and 1 pg of CHO-S PLBL2 upon loading the capillary. Mouse anti-CHO PLBL2 monoclonal IgG was prepared by diluting the IgG to 1.0 µg/mL using antibody diluent. Anti-mouse HRP conjugate secondary polyclonal antibodies were used as supplied in the anti-mouse detection module. The Luminol-S and peroxide substrate mixture was prepared according to vendor recommendations and stored on ice. The biotinylated ladder, CHO-S PLBL2 standards, mAb A strip samples, primary antibody, secondary antibody, antibody diluent, wash buffer, and Luminol-S/peroxide substrates were then dispensed onto the plate following vendor instructions. Three negative controls were included: first, a control capillary in which the 25 pg CHO-S PLBL2 standard served as the antigen but the primary mouse anti-CHO PLBL2 monoclonal antibody was replaced with antibody diluent; second, a control capillary in which the 25 pg CHO-S PLBL2 standard served as the antigen but the anti-mouse HRP conjugate was replaced with antibody diluent; and, finally, a control capillary in which the antigen was replaced by 0.1X sample buffer while the primary mouse anti-CHO PLBL2 monoclonal antibody and anti-mouse HRP conjugate were unchanged. Separation was performed using default settings: separation at 375 V for 25 minutes, photo-immobilization, antibody diluent blocking for 5 minutes, primary and secondary antibody incubation for 30 minutes each, and luminol/peroxide chemiluminescence detection with exposures of 1, 2, 4, 8, 16, 32, 64, 128, and 512 seconds. The separation and immunoblotting were analyzed using Compass for Simple Western (version 4.0.0). Peaks were detected and identified automatically by the software, but each electropherogram was inspected and manual corrections to standards were performed, where necessary. For the CHO-S PLBL2 standards, peak areas of the dominant PLBL2 peak (~80 kDa) were determined, plotted, and fitted using linear regression. Using the standard curve, the limit of quantitation (LOQ) for the capillary Western blot was determined to be <1 pg CHO-S PLBL2. Areas of the analogous ~80 kDa peaks in the mAb A samples were then used to determine their CHO-S PLBL2 content. A vial of HCCF containing mAb B (at 4.5 mg/mL or 30 µM) was removed from storage at ≤-65°C and thawed at room temperature. The removal of endogenous CHO PLBL2 proceeded by incubating the immobilized goat anti-CHO PLBL2 IgGs with the mab B HCCF at sub-saturating conditions for the CHO PLBL2: 250 µg (or 1.66 nmol in 250 µL of bead solution) of immobilized goat anti-CHO PLBL2 IgGs were mixed with HCCF containing 55 µg (or 0.8 nmol) of endogenous CHO PLBL2 to achieve a molar ratio of 0.5 moles of CHO PLBL2 to 1 mole of immobilized IgG. The solution was mixed end over end for 19.5 hours at 2-8°C while shielding from light Case Study 2: Removal of Endogenous CHO PLBL2 from Pembrolizumab (IgG4) Quantitation of Endogenous CHO PLBL2 Content of Pembrolizumab The endogenous CHO PLBL2 content of pembrolizumab was determined via capillary Western blot. The analysis was performed on a Jess system (Protein Simple, 004-650) using a 12 – 230 kDa separation module and the anti-mouse detection module. Pembrolizumab (formulated in 100 mM proline-acetate and 20 mM arginine (pH 5.0) at a protein concentration of 7.29 mg/mL (or 49.7 µM)) was prepared undiluted and diluted at 2X, 5X, 10X, 20X, 50X, and 100X using PBS (pH 7.2) prior to mixing with the 5X fluorescent master mix and proceeding as described above. The separation and immunoblotting were analyzed using Compass for Simple Western (version 6.0.0) and peak detection followed the procedure described above. The endogenous CHO PLBL2 content of pembrolizumab was quantitated against a standard curve of CHO-S PLBL2 and was determined to be about 2-8 µg/mL. Figure S.10 shows the electropherograms of the 10X, 20X, 50X, and 100X diluted pembrolizumab samples. Removal of Endogenous CHO PLBL2 from Pembrolizumab A 50 mg vial of pembrolizumab (formulated in 100 mM proline-acetate and 20 mM arginine (pH 5.0) at a protein concentration of 7.3 mg/mL (or 50 µM)) was removed from storage at ≤-65°C and thawed at room temperature. To prepare for the separation, the pembrolizumab was adjusted to pH 7.0 using 1 M Tris (pH 8.0), rendering a solution containing 6.2 mg/mL (or 42 µM). The removal of endogenous CHO PLBL2 proceeded by incubating the immobilized goat anti-CHO PLBL2 IgGs with the pembrolizumab at sub-saturating conditions for the CHO PLBL2: 250 µg (or 1.66 nmol in 250 µL of bead solution) of immobilized goat anti-CHO PLBL2 IgG were mixed with 8.45 mL of adjusted pembrolizumab solution containing 54 µg (or 0.8 nmol) of endogenous CHO PLBL2 to achieve a molar ratio of 0.5 moles of CHO PLBL2 to 1 mole of immobilized IgG (assuming an endogenous CHO PLBL2 concentration of 8 µg/mL). The solution was mixed end over end for 19.5 hours at 2-8°C while shielding from light. Following incubation, 1 mL of the solution was added to a clean 1.5 mL Eppendorf tube, which was placed onto the DynaMag TM -2 to magnetize and separate the beads from solution. The supernatant from the Eppendorf tube was collected into a separate 15 mL conical tube and replaced with another 1 mL of the post-incubation solution. This process was repeated until all supernatant from the incubation had been separated from the beads and collected. The beads were then washed using 5 BVs (i.e., 5 x 250 µL) of 50 mM Tris (pH 8.0), 3 BVs of 50 mM Tris with 1% (w/v) polysorbate 80 (pH 9.0), and 3 BVs of 50 mM Tris (pH 8.0). Bound CHO PLBL2 was stripped from the immobilized anti-CHO PLBL2 IgG using 3 x 80 µL (~1/3 BV) of 1% (w/v) formic acid (pH 2.2). Capillary Western Blot Analysis The load material, supernatant, wash, and elution samples were analyzed via capillary Western blot (using the Jess) similar to the procedure described above. Alongside the pembrolizumab samples, a standard curve of CHO-S PLBL2 was analyzed to enable quantitation of the endogenous CHO PLBL2 content of the pembrolizumab samples. Figure S.11 shows the electropherograms of the pre-capture starting material, polished pembrolizumab, and strip fractions. CHO PLBL2 ELISA Analysis The CHO PLBL2 content of the pembrolizumab samples was determined via CHO PLBL2 ELISA. The ELISA kit was removed from storage at 2-8°C and equilibrated at room temperature for 1 hour prior to use. The pembrolizumab samples and PLBL2 assay control were removed from storage at ≤-65°C and thawed at room temperature. The assay control was diluted 100X using diluent solution for a final working concentration of 3 ng/mL. The load sample of pembrolizumab was diluted 500X, then further serially diluted 1000X, 2000X, and 4000X using the diluent solution. The polished pembrolizumab sample was diluted 20X, then further serially diluted 40X, 80X, and 160X using the diluent solution. Both the 100X detection antibody and 100X HRP-streptavidin solutions were diluted to working concentrations of 1X by adding 120 µL of the respective 100X solutions to 11.88 mL of diluent solution. The 1X wash solution was prepared by adding 50 mL of 20X wash solution to 950 mL of Milli-Q water. The calibrator used to prepare the working standard curve was reconstituted using Milli-Q water, then diluted to a concentration of 20 ng/mL CHO PLBL2 in diluent solution according to the lot-specific certificate of analysis provided in the kit. Intermediate standard stock was prepared using a 2X dilution of the 20 ng/mL CHO PLBL2 calibrator in diluent solution for a final concentration of 10 ng/mL. The working standard series was prepared by further diluting the 10 ng/mL intermediate standard stock in diluent solution to final standard concentrations of 5, 2.5, 1.25, 0.63, 0.31, 0.16, and 0.08 ng/mL. 100 µL of prepared standards, diluent blank, assay control, and pembrolizumab were transferred in triplicate to the microtiter strips in the ELISA plate, covered, and incubated on a plate shaker at 25°C for 2 hours while agitating at 500 rpm. Following the incubation, each well of the ELISA plate was washed 3 times with 375 µL of 1X wash solution using a Bio-Tek automated plate washer (ELX405TS from Agilent; Santa Clara, CA, USA). 100 µL of 1X detection antibody were added to each well of the microtiter strips in the ELISA plate. The plate was covered with a foil plate seal and incubated on a plate shaker at 25°C for 20 minutes while agitating at 500 rpm. Following the incubation, each well of the ELISA plate was washed using the same wash procedure described above. 100 µL of 1X HRP-streptavidin were added to each well of the microtiter strips in the ELISA plate. The plate was covered with a foil plate seal and incubated on a plate shaker at 25°C for 20 minutes while agitating at 500 rpm. Following the incubation, each well of the ELISA plate was washed using the same wash procedure described above. 100 µL of TMB chromogen substrate solution were added to each well of the microtiter strips in the ELISA plate. The plate was covered with a foil plate seal and incubated on a plate shaker at 25°C for 10 minutes while agitating at 500 rpm. Following the incubation, the reaction was stopped by adding 100 µL of stop solution to each well of the ELISA plate. The absorbance at 450 nm of each well was measured using a SpectraMax M2E microplate reader (Molecular Devices; San Jose, CA, USA). The data were analyzed using SoftMax Pro (Molecular Devices, version 7.0.3). The standard curve was generated by plotting the absorbance at 450 nm against the CHO PLBL2 concentration of the standards and applying a 4-parameter logistic fit. Each sample and assay control replicate were interpolated against this fitted standard curve and multiplied by the initial dilution factor to determine the CHO PLBL2 concentration. All replicates that fell within the standard curve range of 5 to 0.08 ng/mL were used to determine the final back-calculated CHO PLBL2 concentration of the samples and assay control. A minimum of two dilutional levels were on curve and included in the analysis for each of the pembrolizumab samples. Case Study 3: Removal of Endogenous CHO PLBL2 from Harvest Cell Culture Fluid (HCCF) Containing Monoclonal Antibody IgG1 (mAb B) Expression of mAb B A humanized monoclonal IgG1 antibody (mAb B) was expressed in CHO-K1 cells via suspension culture, clarified to remove cells and cellular debris, and stored at ≤-65°C according to internal procedures. Removal of Endogenous CHO PLBL2 from mAb B A vial of HCCF containing mAb B (at 4.5 mg/mL or 30 µM) was removed from storage at ≤-65°C and thawed at room temperature. The removal of endogenous CHO PLBL2 proceeded by incubating the immobilized goat anti-CHO PLBL2 IgGs with the mab B HCCF at sub-saturating conditions for the CHO PLBL2: 250 µg (or 1.66 nmol in 250 µL of bead solution) of immobilized goat anti-CHO PLBL2 IgGs were mixed with HCCF containing 55 µg (or 0.8 nmol) of endogenous CHO PLBL2 to achieve a molar ratio of 0.5 moles of CHO PLBL2 to 1 mole of immobilized IgG. The solution was mixed end over end for 19.5 hours at 2-8°C while shielding from light. Following incubation, the mixture was treated as described for the pembrolizumab samples to separate the beads and recover the supernatant. The beads were then washed using 5 BVs (i.e., 5 x 250 µL) of 50 mM Tris (pH 8.0), 3 BVs of 50 mM Tris with 1% (w/v) polysorbate 80 (pH 9.0), and 3 BVs of 50 mM Tris (pH 8.0). Bound CHO PLBL2 was stripped from the immobilized anti-CHO PLBL2 IgGs using 3 x 80 µL (~1/3 BV) of 1% (w/v) formic acid (pH 2.2). Capillary Western Blot Analysis The load material, supernatant, wash, and elution samples were analyzed via capillary Western blot (using the Jess) similar to the procedure described above. Alongside the mAb B samples, a standard curve of CHO-S PLBL2 was analyzed to enable quantitation of the endogenous CHO PLBL2 content of the mAb B samples. Figure S.12 shows the electropherograms of the pre-capture starting material, post-capture supernatant, and three strips. CHO PLBL2 ELISA Analysis Analysis of the CHO PLBL2 content for the mAb B HCCF samples was performed according to the procedure described above for the pembrolizumab samples. Analysis of Charge Isoforms via Isoelectric Focusing (IEF) Gel and Western Blot To enable IEF under native conditions, two Novex TM IEF gels (pH 3-7) were prepared. The three strip fractions from the affinity capture of CHO PLBL2 from pembrolizumab were each mixed with a 2X concentrate of IEF sample buffer (pH 3-7). Lanes assigned to these samples were loaded with 20 µL of prepared sample. In contrast, lanes assigned to IEF marker (pH 3-10) were loaded with 10 µL of IEF marker. Electrophoresis was conducted using an XCell SureLock® Mini-Cell connected to a Criterion TM PowerPac Universal power supply by running at 100 V for 1 hour, 200 V for 1 hour, and 500 V for 30 minutes. After the separation, the IEF gels were removed and rinsed using deionized water. One IEF gel was incubated in 1X IEF fixing solution while gently agitating at room temperature (about 22°C) for 30 minutes. After fixing, the IEF gel was rinsed using water. Protein bands were stained by incubating the gel with One-Step Blue® protein gel stain at room temperature overnight, followed by destaining in water for 2 hours. Bands were visualized using a GS-900 Calibrated Densitometer from Bio-Rad. The imaging system used the Coomassie Brilliant Blue R-250 setting, employing a transmissive scanning mode and a red filter. The second IEF gel was equilibrated in transfer buffer (0.7% (w/v) acetic acid, pH 3.0) for 10 minutes while preparing the Western blot. Transfer of the proteins from the IEF gel to the PVDF membrane was performed via wet transfer. Prior to use, the PVDF membrane was immersed in methanol for 2 minutes before dripping to remove excess methanol and being transferred to the transfer buffer for 5 minutes. Similarly, the filter papers and sponges were immersed in transfer buffer for 5 minutes. The transfer sandwich was assembled in the XCell II TM Blot Module with the PVDF membrane on the negatively charged electrode side of the stack and the IEF gel on the positively charged electrode side of the stack, according to vendor instructions. The XCell II TM Blot Module was then assembled in the XCell SureLock® Mini-Cell, with the upper chamber filled with transfer buffer and the lower chamber filled with deionized water. Using the Criterion TM PowerPac Universal power supply, protein transfer proceeded at 20 V for 1 hour. Blocking was performed by incubating the PVDF membrane in phosphate-buffered saline containing 0.1% polysorbate 20 (i.e., PBST) with 5% dry milk while gently agitating at 2-8°C overnight. After decanting the blocking solution, the primary antibody solution (0.1 µg/mL of the mouse anti-CHO PLBL2 monoclonal IgG in PBST) was applied and the PVDF membrane was incubated while gently agitating at room temperature for 60 minutes. After incubation, the primary antibody solution was decanted and the PVDF membrane was washed three times using PBST while gently agitating at room temperature for 5 minutes per wash. After the final wash, the secondary antibody solution (0.1 µg/mL of the donkey anti-mouse polyclonal IgG-HRP conjugate in PBST) was applied and the PVDF membrane was incubated while gently agitating at room temperature for 60 minutes. After incubation, the secondary antibody solution was decanted and the PVDF membrane was washed three times using PBST while gently agitating at room temperature for 30 seconds per wash, followed by three more washes in the same solution for 10 minutes per wash. Immediately prior to use, the working ECL reagent was prepared by mixing ECL Reagent 1 and Reagent 2 in equal-volume amounts. After decanting the final PBST wash following secondary antibody incubation, the working ECL reagent was added to the PVDF membrane, and the membrane was incubated while gently agitating at room temperature for 2 minutes to ensure universal coverage of the membrane by the ECL reagent. The membrane was then removed using forceps and excess ECL reagent was allowed to drip off just prior to imaging. Imaging employed a G:Box Chemi-XX6 imaging system equipped with Genesis 1.8.2.0 (Syngene). Protein bands were visualized using the ECL chemiluminescent blot setting and exposure times in an additive series which included 1 second, 2 seconds, 5 seconds, 10 seconds, 30 seconds, 1 minute, 2 minutes, and 5 minutes. RESULTS AND DISCUSSION Site-Specific Modification and Immobilization Proceed with High Yields and Coupling Efficiency As depicted in Scheme 1, the polyclonal anti-CHO PLBL2 IgGs underwent successful chemo-enzymatic modification following our published approach (36): first, cleavage of the Fc N-glycan to the core N-acetylglucosamine via EndoS2 (E.C. 3.2.1.96); second, installation of an azide handle via an engineered β-1,4-galactosyltransferase (GalT Y289L; E.C. 2.4.1.22); and, finally, click chemistry (via strain-promoted azide-alkyne cycloaddition) with a dibenzocyclooctyne (DBCO)-functionalized biotin. The biotinylated IgGs were then immobilized by coupling their biotin moieties to streptavidin-functionalized magnetic beads, rendering our anti-CHO PLBL2 resin. As previously observed and as described in the Materials and Methods, all chemo-enzymatic steps proceeded cleanly at mild conditions and achieved high yields. Immobilization was confirmed via SDS-PAGE gel. Under reducing conditions, the biotinylated IgGs (i.e., the load material for the immobilization) exhibit biotinylated heavy chain (~50 kDa; Figure 1, Lane 3) and unmodified light chain (~25 kDa; Figure 1, Lane 3). No biotinylated heavy chain or unmodified light chain bands are observed in the supernatant recovered from the immobilization reaction (Figure 1, Lane 4) or from subsequent wash steps (Figure S.8, Lanes 5 – 13 in the Supporting Information). Densitometric analysis of the post-immobilization supernatant determined that >95% of the biotinylated anti-CHO PLBL2 IgGs were immobilized onto the resin. Moreover, the anti-CHO PLBL2 IgGs were only recoverable from the beads upon exposure to high concentrations of detergents or to high heat (data not shown), consistent with the exceptional strength of the biotin-streptavidin interaction (38). Successful Removal of CHO PLBL2 from a Diverse Array of Protein Mixtures and Determination of Its Molecular Size Variants The anti-CHO PLBL2 resin demonstrated a robust ability to remove CHO PLBL2 from a diverse array of bioprocessing streams and therapeutic protein modalities. Scheme 2 depicts the general workflow: Anti-CHO PLBL2 IgGs immobilized on the resin exposed to the protein matrix (containing both CHO PLBL2 and a model therapeutic mAb) recognize and bind with high affinity to the contaminating CHO PLBL2. Introduction of a magnetic field removes the anti-CHO PLBL2 resin (and the bound CHO PLBL2) from suspension, enabling the recovery of the polished therapeutic mAb. Application of a strip solution disrupts the antigen-antibody interaction and enables the recovery of the captured CHO PLBL2 separate from the purified mAb. In case study 1, we demonstrated that our resin clears CHO PLBL2 even when already present at very trace levels (e.g., sub-ppm) typical of drug substances but which are nevertheless a concern for stability of protein products and, moreover, even when CHO PLBL2 is present as a mixture of distinct isoforms. Binary mixtures of highly purified humanized monoclonal IgG1 antibody (mAb A) and commercial, recombinant CHO-S PLBL2 were prepared by spiking in the commercial CHO-S PLBL2 to concentrations of 10 ppm, 1 ppm, and 0.1 ppm (relative to mAb A) alongside a 0 ppm control (i.e., no CHO-S PLBL2), which was intended to otherwise mimic the 0.1 ppm condition. When analyzed via capillary Western blot (Figure 2), the commercial CHO-S PLBL2 exhibits multiple distinct peaks, including a major peak at ~80 kDa and a minor peak at ~45 kDa; the major peak at ~80 kDa was used to quantitate the CHO PLBL2 content in each sample. This pattern of multiple molecular size isoforms is consistent with previous reports for CHO PLBL2 produced recombinantly and for its mouse and human orthologs and may stem from limited proteolysis of PLBL2 following translation (39-41). As seen in Figure 3, none of the prepared binary mixtures of mAb A and commercial CHO-S PLBL2 shows a visible peak at 80 kDa, indicating that the concentration of the spiked-in CHO-S PLBL2is below the limit of detection for the assay. Nevertheless, the electropherograms for the strip fractions of the 10 ppm, 1 ppm, and 0.1 ppm conditions all exhibit major and minor peaks and profiles consistent with those of the commercial CHO-S PLBL2 in Figure 2. These findings indicate that even when CHO-S PLBL2 is present (but not detectable) in the starting material down to at least the 0.1 ppm level, our resin can capture and separate the CHO-S PLBL2 from mAb A. Moreover, the custom affinity resin recognizes, captures, and enables the subsequent recovery of multiple molecular weight isoforms of commercial CHO-S PLBL2, thereby serving as a comprehensive tool for the removal of diverse PLBL2 variants. Finally, the CHO-S PLBL2 recovered in the strip represented ~70% of that loaded into the 10 ppm, 1 ppm, and 0.1 ppm binary mixtures, indicating that at least 70% of the CHO-S PLBL2 was removed from mAb A by the resin to afford higher purity mAb A in the flow through. As expected, similar behavior is not observed for the 0 ppm control, where no commercial CHO-S PLBL2 was spiked in. As discussed in the introduction, previous chromatographic and surface plasmon resonance spectroscopy investigations have demonstrated the preferential binding of CHO PLBL2 to IgG4s over other IgG isotypes (13), which renders CHO PLBL2 removal a particular concern during the development of IgG4-based therapeutics. Our second case study utilized pembrolizumab – a humanized IgG4 which inhibits PD-1 and is administered to treat various cancers (42,43) – as a model IgG4 to assess CHO PLBL2 removal by our custom resin. Notably, the pembrolizumab used in case study 2 differs from the binary mixtures engineered in case study 1, in that pembrolizumab containing endogenous (rather than spiked-in recombinant) CHO PLBL2 was utilized. That is, the pembrolizumab (purchased from MedChemExpress) possessed residual CHO PLBL2 which had been expressed endogenously in CHO cells alongside the pembrolizumab. This pembrolizumab exhibited >520 ppm of residual, endogenous CHO PLBL2 even after undergoing extensive purification via harvest clarification, Protein A affinity chromatography, cation-exchange chromatography, and ultrafiltration/diafiltration by MedChemExpress (as seen in Figure 5 and Figure S.10 in the Supporting Information).As seen in Lane 2 of Figure 4, the pembrolizumab starting material exhibits at least three molecular size isoforms, including a major isoform which migrates to ~80 kDa and two other isoforms which migrate to ~45 kDa and ~30 kDa. Nevertheless, our anti-CHO PLBL2 resin achieved >99% clearance of the CHO PLBL2 to afford pembrolizumab possessing only 0.7 ppm (as confirmed by CHO PLBL2 ELISA, Figure 5) at >85% yield (Figure 7), even without substantial optimization to the separation. As seen in Lane 3 of Figure 4 and in Figure S.11a in the Supporting Information, the recovered pembrolizumab is entirely depleted of the ~80 kDa and ~45 kDa isoforms. Similarly, the resulting strip fractions containing CHO PLBL2 recovered from the resin exhibit primarily the ~80 kDa and ~45 kDa isoforms (Lanes 4-6, Figure 4), along with a small peak corresponding to the ~30 kDa isoform (Figure S.11b in the Supporting Information). Interestingly, this suggests that the ~30 kDa isoform may not be bound by the resin to a significant degree, and this finding highlights the challenges in developing broad yet selective solutions to enable the separation of HCPs from therapeutic proteins. In our final case study, we again investigated clearance of endogenous CHO PLBL2, this time from harvest cell culture fluid (HCCF) containing a second IgG1 antibody (mAb B, case study 3). As the feedstock for subsequent downstream unit operations, HCCF is a highly complex matrix which, alongside the therapeutic protein of interest, contains thousands of other endogenously expressed HCPs (including CHO PLBL2) that are necessary for CHO cellular function. Due to the enormous number of HCP species present in the matrix and the diversity of their physicochemical properties, there is high potential for overlap with those of CHO PLBL2 and, consequently, significant potential to impact CHO PLBL2 binding to the polyclonal antibodies immobilized on our resin. Therefore, HCCF represents an extreme but cogent case to interrogate the selectivity of our anti-CHO PLBL2 resin for its target antigen. As seen in Figure 6, the mAb B HCCF possesses an endogenous CHO PLBL2 content of nearly 600 ppm relative to the mAb B concentration. However, even without substantial optimization, our resin captured and removed >60% of the CHO PLBL2 to achieve <240 ppm in the recovered mAb B product. Similarly to case study 2, the endogenous CHO PLBL2 in case study 3 exhibits three molecular weight isoforms, with the resin effectively capturing the isoforms which migrate to ~80 kDa and ~45 kDa when analyzed via capillary Western blot (Figure S.12 in the Supporting Information). Moreover, as seen in Figure 7, the separation achieved nearly 100% yield for mAb B while drastically reducing the burden of CHO PLBL2 removal for subsequent downstream unit operations. CHO PLBL2 Also Exhibits Numerous Charge Isoforms Although it is widely recognized that eukaryotic expression systems such as CHO cells introduce to their expressed proteins heterogeneity through post-translational and other modifications, this knowledge has largely been overlooked for CHO HCPs, including PLBL2. Analysis via isoelectric focusing (IEF) gel electrophoresis followed by a Western blot revealed that, in addition to the molecular weight isoforms described above, CHO PLBL2 exhibits numerous and distinct charge isoforms. As seen in Lanes 3 through 5 of Figure 8 (and Figure S.13 in the Supporting Information), CHO PLBL2 which was captured and recovered from pembrolizumab demonstrates several distinct protein bands between isoelectric points (pIs) of about 5.0 and about 6.0, indicating a diverse array of charge isoforms within this pI range. For reference, the theoretical pI for CHO PLBL2 (calculated based on the amino acid sequence of UniProt entry G3I6T1) is 5.63 (39). Bands are also observed at pIs >6.0, although these were not fully resolved on the IEF gel or in the Western blot. CONCLUSION To the best of our knowledge, ours is the first publication of a custom affinity resin, especially one which employs polyclonal antibodies, for the targeted and robust removal of CHO PLBL2 from bioprocessing streams. From a downstream processing perspective, our battery of testing demonstrates that our resin effectively binds and separates CHO PLBL2 from myriad feedstocks, including those possessing thousands of other HCPs which may compete for binding to the immobilized polyclonal antibodies. Our case studies showcased that our custom anti-CHO PLBL2 resin magnificently isolates and removes CHO PLBL2 even when present at sub-ppm levels and, crucially, even when CHO PLBL2 exists as a diverse collection of charge and molecular size variants, rather than a single molecular species. Nevertheless, the capture and substantial removal of CHO PLBL2 across all three case studies suggest that our resin provides broad coverage across these proteoforms (and likely many others, as discussed below), adding a potent arrow in the quiver to tackle a notorious HCP during bioprocess development. Although our research utilized goat polyclonal IgGs for the capture of our desired HCP, antibodies derived from other animal species which meet the requirements of our chemo-enzymatic workflow (36) may be equally suitable for generating other custom resins. Because many non-goat antibodies have been developed to target other challenging HCPs, successful application of our workflow to these antibodies will broaden the HCP species which can be subjected to selective capture and removal by such resins. We envision that our approach will enable exquisite tuning to address an array of bioprocess needs, such as the mixing of multiple custom resins – each targeting a distinct HCP – to capture and clear multiple problematic HCPs simultaneously in a single, tailored separation. While this publication showcases a proof of concept using streptavidin-functionalized magnetic beads, our ongoing research focuses on adapting our chemo-enzymatic workflow to conjugate the anti-HCP antibodies directly to the chromatography resin. Additionally, we are exploring the use of porous (e.g., agarose) beads to enable removal in a packed-bed format operated in flow-through mode, in which the CHO PLBL2 is bound by the resin while the therapeutic protein flows through the packed bed and is recovered at substantially higher purity. These adaptations will aid in the adoption of our technology at bench and manufacturing scales for both therapeutic proteins and newer biological modalities (e.g., adeno-associated virus and lentiviral vectors) where HCPs are emerging concerns. Finally, we have demonstrated that the captured CHO PLBL2 is readily released from the immobilized antibodies in high yields; in isolating it from the therapeutic protein and any other HCPs, it is also highly enriched. As described in the introduction, we expect that CHO PLBL2 exists as a diverse collection of proteoforms. Indeed, as seen in the second part of this series of studies, we confirm that this enrichment opens the door to deeper and more sensitive analyses of CHO PLBL2 (and, by extension, to other HCPs) via mass spectrometry, such as the determination of post-translational modifications (e.g., glycosylation, phosphorylation, deamidation, oxidation, etc.) (37). Armed with a stronger understanding of the impurities which plague our bioprocesses, we can develop novel and more targeted purification strategies to ensure their removal and safeguard both patient safety and clinical and commercial supplies. AUTHOR CONTRIBUTIONS M.E.D. contributed to the conceptualization and methodology, carried out experiments, interpreted the results, led the literature review, and prepared the manuscript. A.T. carried out experiments, interpreted the results, and prepared the manuscript. S.F.O. contributed to the conceptualization and methodology, interpreted the results, and reviewed and edited the manuscript. Z.S.Z. contributed to the conceptualization and methodology, interpreted the results, and reviewed and edited the manuscript. All authors agreed to the final version of the manuscript. The authors kindly thank Elly Klages and Drs. Christina Alves and Tracy Sioussat for their helpful reviews of the manuscript. The authors also thank Drs. Leo Wang, Yan Wang, Chris Barton, Olga Paley and Professors Alexander Ivanov and Mary Jo Ondrechen for their technical insights and helpful discussions. FUNDING The funding for this research was provided by Takeda Development Center Americas (a wholly owned subsidiary of Takeda Pharmaceutical Company Ltd). CONFLICT OF INTEREST STATEMENT M.E.D., A.T., and S.F.O. are employees of Takeda Development Center Americas, which is a wholly owned subsidiary of Takeda Pharmaceutical Company Ltd. All individuals may own Takeda stock, restricted stock units, and/or stock options. DATA AVAILABILITY STATEMENT All data are incorporated into the article and its online supplementary material. Graphic Abstract. A novel, facile, and platform-able approach using antibodies to create custom affinity resins for the selective and robust capture and removal of “difficult-to-remove” host cell proteins (HCPs) from various bioprocess streams. Scheme 1. Chemo-enzymatic, glycan-mediated, site-specific modification of goat anti-CHO PLBL2 polyclonal IgGs and immobilization onto solid supports and resins (e.g., magnetic beads). Figure 1. Analysis of immobilized goat polyclonal anti-CHO PLBL2 IgG under reducing conditions on an SDS-PAGE gel. Lanes 1 and 2: molecular weight markers. Lane 3: biotinylated goat anti-CHO PLBL2 IgG, exhibiting biotinylated heavy chain (HC) and unmodified light chain (LC) bands. Lane 4: supernatant recovered after the immobilization step, exhibiting no HC or unmodified LC bands, thereby confirming immobilization of the IgG onto the streptavidin-functionalized beads. Scheme 2. Binding of a target host cell protein (e.g., CHO PLBL2) to the immobilized polyclonal antibodies directed against the host cell protein of interest and subsequent removal from a bioprocessing stream containing a highly pure therapeutic protein (e.g., monoclonal antibody). Figure 2. Commercial, recombinant CHO-S PLBL2 analyzed via capillary Western blot at multiple protein masses. The electropherograms exhibit a major peak at ~80 kDa and a minor peak at ~45 kDa. Figure 3. Successful capture and removal of the commercial CHO-S PLBL2 from a binary mixture containing mAb A using the immobilized polyclonal anti-CHO PLBL2 IgGs (case study 1). Even when PLBL2 is present (but not detectable) in the starting material down to at least the 0.1 ppm level, CHO-S PLBL2 removal was achieved at relative CHO-S PLBL2 spiking levels of (a) 10 ppm (i.e., 10 ng of CHO-S PLBL2 per mg of mAb A), (b) 1 ppm, and (c) 0.1 ppm. As expected, no removal (or subsequent recovery in the strip) was observed for condition (d) 0 ppm, which experienced no commercial CHO-S PLBL2 spike. For each condition, electropherograms show the load material and three strip fractions. Figure 4. Affinity capture of CHO PLBL2 from pembrolizumab and subsequent enrichment, as detected via capillary Western blot. Lane 1: molecular weight marker. Lane 2: pembrolizumab starting material (pre-enrichment, “load”), which exhibits multiple molecular weight isoforms (~80 kDa, ~45 kDa, and ~30 kDa) of endogenous CHO PLBL2. Lane 3: pembrolizumab after affinity capture (“depleted”), which exhibits depletion of the ~80 kDa and ~45 kDa isoforms of endogenous CHO PLBL2 but retention of the ~30 kDa isoform. Lanes 4 – 6: enriched fractions, which exhibit recovery and enrichment of primarily the ~80 kDa and ~45 kDa isoforms of endogenous CHO PLBL2. Figure 5. Substantial reduction of CHO PLBL2 from a solution of pembrolizumab (an IgG4; case study 2). The initial pembrolizumab solution (“load”) exhibited >520 ppm (or >520 ng of CHO PLBL2 per mg of pembrolizumab) of residual CHO PLBL2. However, our anti-CHO PLBL2 resin removed >99% of the residual PLBL2 to achieve 0.7 ppm in the recovered pembrolizumab (“polished pembrolizumab”). Figure 6. Capture and reduction of CHO PLBL2 from harvest cell culture fluid (HCCF) containing mAb B (case study 3). Because the HCCF also contains thousands of other endogenous HCPs, it is an extreme case to investigate the selectivity of the immobilized anti-CHO PLBL2 IgGs for their target antigen over other competing HCP species. The initial mAb B HCCF (“load”) exhibited nearly 600 ppm of residual CHO PLBL2. However, our anti-CHO PLBL2 resin removed >60% of the CHO PLBL2 to achieve <240 ppm of CHO PLBL2 in the mAb B solution after exposure to the resin (“polished mAb B”), thereby significantly reducing the burden of CHO PLBL2 removal for subsequent downstream unit operations. Figure 7. Recoveries of pembrolizumab (86%, case study 2) and mAb B (99%, case study 3) following removal of endogenous CHO PLBL2 using the custom affinity anti-CHO PLBL2 resin. Notably, high recoveries were achieved without significant process optimization. Figure 8. Analysis via IEF gel and Western blot of the charge variants of CHO PLBL2 captured from pembrolizumab. Lanes 1 – 2: isoelectric point (pI) markers. Lanes 3 – 5: enriched CHO PLBL2. Numerous distinct protein bands are observed between pIs of about 5.0 and about 6.0, indicating a diverse array of charge isoforms for CHO PLBL2 within this range. Bands are also observed at pIs >7.0, which were not fully resolved. For reference, the theoretical pI for CHO PLBL2 (calculated based on the amino acid sequence of UniProt entry G3I6T1) is 5.63. Figure 8. Analysis via IEF gel and Western blot of the charge variants of CHO PLBL2 captured from pembrolizumab. Lanes 1 – 2: isoelectric point (pI) markers. Lanes 3 – 5: enriched CHO PLBL2. Numerous distinct protein bands are observed between pIs of about 5.0 and about 6.0, indicating a diverse array of charge isoforms for CHO PLBL2 within this range. Bands are also observed at pIs >7.0, which were not fully resolved. For reference, the theoretical pI for CHO PLBL2 (calculated based on the amino acid sequence of UniProt entry G3I6T1) is 5.63. REFERENCES: 1. Levy, N.E., Valente, K.N., Choe, L.H., Lee, K.H. and Lenhoff, A.M. (2014) Identification and characterization of host cell protein product-associated impurities in monoclonal antibody bioprocessing. Biotechnol Bioeng , 111 , 904-912. 2. Shukla, A.A. and Hinckley, P. (2008) Host cell protein clearance during protein A chromatography: development of an improved column wash step. Biotechnol Prog , 24 , 1115-1121. 3. Tarrant, R.D., Velez-Suberbie, M.L., Tait, A.S., Smales, C.M. and Bracewell, D.G. (2012) Host cell protein adsorption characteristics during protein A chromatography. Biotechnol Prog , 28 , 1037-1044. 4. Krawitz, D.C., Forrest, W., Moreno, G.T., Kittleson, J. and Champion, K.M. (2006) Proteomic studies support the use of multi-product immunoassays to monitor host cell protein impurities. Proteomics , 6 , 94-110. 5. Chen, I.H., Xiao, H., Daly, T. and Li, N. (2020) Improved Host Cell Protein Analysis in Monoclonal Antibody Products through Molecular Weight Cutoff Enrichment. Anal Chem , 92 , 3751-3757. 6. Johnson, R.O., Greer, T., Cejkov, M., Zheng, X. and Li, N. (2020) Combination of FAIMS, Protein A Depletion, and Native Digest Conditions Enables Deep Proteomic Profiling of Host Cell Proteins in Monoclonal Antibodies. Anal Chem , 92 , 10478-10484. 7. Hamaker, N.K., Min, L. and Lee, K.H. (2022) Comprehensive assessment of host cell protein expression after extended culture and bioreactor production of CHO cell lines. Biotechnol Bioeng , 119 , 2221-2238. 8. Carvalho, S.B., Profit, L., Krishnan, S., Gomes, R.A., Alexandre, B.M., Clavier, S., Hoffman, M., Brower, K. and Gomes-Alves, P. (2024) SWATH-MS as a strategy for CHO host cell protein identification and quantification supporting the characterization of mAb purification platforms. J Biotechnol , 384 , 1-11. 9. Wilson, L.J., Lewis, W., Kucia-Tran, R. and Bracewell, D.G. (2022) Identification and classification of host cell proteins during biopharmaceutical process development. Biotechnol Prog , 38 , e3224. 10. Valente, K.N., Lenhoff, A.M. and Lee, K.H. (2015) Expression of difficult-to-remove host cell protein impurities during extended Chinese hamster ovary cell culture and their impact on continuous bioprocessing. Biotechnol Bioeng , 112 , 1232-1242. 11. Jones, M., Palackal, N., Wang, F., Gaza-Bulseco, G., Hurkmans, K., Zhao, Y., Chitikila, C., Clavier, S., Liu, S., Menesale, E. et al. (2021) "High-risk" host cell proteins (HCPs): A multi-company collaborative view. Biotechnol Bioeng , 118 , 2870-2885. 12. Fischer, S.K., Cheu, M., Peng, K., Lowe, J., Araujo, J., Murray, E., McClintock, D., Matthews, J., Siguenza, P. and Song, A. (2017) Specific Immune Response to Phospholipase B-Like 2 Protein, a Host Cell Impurity in Lebrikizumab Clinical Material. AAPS J , 19 , 254-263. 13. Tran, B., Grosskopf, V., Wang, X., Yang, J., Walker, D., Jr., Yu, C. and McDonald, P. (2016) Investigating interactions between phospholipase B-Like 2 and antibodies during Protein A chromatography. J Chromatogr A , 1438 , 31-38. 14. Robert, F., Bierau, H., Rossi, M., Agugiaro, D., Soranzo, T., Broly, H. and Mitchell-Logean, C. (2009) Degradation of an Fc-fusion recombinant protein by host cell proteases: Identification of a CHO cathepsin D protease. Biotechnol Bioeng , 104 , 1132-1141. 15. Bee, J.S., Tie, L., Johnson, D., Dimitrova, M.N., Jusino, K.C. and Afdahl, C.D. (2015) Trace levels of the CHO host cell protease cathepsin D caused particle formation in a monoclonal antibody product. Biotechnol Prog , 31 , 1360-1369. 16. Ranjan, S., Chung, W.K., Hofele, R., Heidbrink Thompson, J., Bee, J., Zhang, L., Robbins, D. and Cramer, S.M. (2019) Investigation of cathepsin D-mAb interactions using a combined experimental and computational tool set. Biotechnol Bioeng , 116 , 1684-1697. 17. Ranjan, S., Chung, W.K., Zhu, M., Robbins, D. and Cramer, S.M. (2019) Implementation of an experimental and computational tool set to study protein-mAb interactions. Biotechnol Prog , 35 , e2825. 18. Chiu, J., Valente, K.N., Levy, N.E., Min, L., Lenhoff, A.M. and Lee, K.H. (2017) Knockout of a difficult-to-remove CHO host cell protein, lipoprotein lipase, for improved polysorbate stability in monoclonal antibody formulations. Biotechnol Bioeng , 114 , 1006-1015. 19. Graf, T., Tomlinson, A., Yuk, I.H., Kufer, R., Spensberger, B., Falkenstein, R., Shen, A., Li, H., Duan, D., Liu, W. et al. (2021) Identification and Characterization of Polysorbate-Degrading Enzymes in a Monoclonal Antibody Formulation. J Pharm Sci , 110 , 3558-3567. 20. Chen, Y., Xu, C.F., Stanley, B., Evangelist, G., Brinkmann, A., Liu, S., McCarthy, S., Xiong, L., Jones, E., Sosic, Z. et al. (2021) A Highly Sensitive LC-MS/MS Method for Targeted Quantitation of Lipase Host Cell Proteins in Biotherapeutics. J Pharm Sci , 110 , 3811-3818. 21. Hall, T., Sandefur, S.L., Frye, C.C., Tuley, T.L. and Huang, L. (2016) Polysorbates 20 and 80 Degradation by Group XV Lysosomal Phospholipase A2 Isomer X1 in Monoclonal Antibody Formulations. J Pharm Sci , 105 , 1633-1642. 22. Zhang, S., Xiao, H. and Li, N. (2024) Analysis of Host Cell Proteins in AAV Products with ProteoMiner Protein Enrichment Technology. Anal Chem , 96 , 1890-1897. 23. Bracewell, D.G., Smith, V., Delahaye, M. and Smales, C.M. (2021) Analytics of host cell proteins (HCPs): lessons from biopharmaceutical mAb analysis for Gene therapy products. Curr Opin Biotechnol , 71 , 98-104. 24. Johnson, S., Wheeler, J.X., Thorpe, R., Collins, M., Takeuchi, Y. and Zhao, Y. (2018) Mass spectrometry analysis reveals differences in the host cell protein species found in pseudotyped lentiviral vectors. Biologicals , 52 , 59-66. 25. Bracewell, D.G., Francis, R. and Smales, C.M. (2015) 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 , 1727-1737. 26. Reese, H.R., Xiao, X., Shanahan, C.C., Chu, W., Van Den Driessche, G.A., Fourches, D., Carbonell, R.G., Hall, C.K. and Menegatti, S. (2020) Novel peptide ligands for antibody purification provide superior clearance of host cell protein impurities. J Chromatogr A , 1625 , 461237. 27. Lavoie, R.A., di Fazio, A., Blackburn, R.K., Goshe, M.B., Carbonell, R.G. and Menegatti, S. (2019) Targeted Capture of Chinese Hamster Ovary Host Cell Proteins: Peptide Ligand Discovery. Int J Mol Sci , 20 . 28. Lavoie, R.A., di Fazio, A., Williams, T.I., Carbonell, R. and Menegatti, S. (2020) Targeted capture of Chinese hamster ovary host cell proteins: Peptide ligand binding by proteomic analysis. Biotechnol Bioeng , 117 , 438-452. 29. Sripada, S.A., Chu, W., Williams, T.I., Teten, M.A., Mosley, B.J., Carbonell, R.G., Lenhoff, A.M., Cramer, S.M., Bill, J., Yigzaw, Y. et al. (2022) Towards continuous mAb purification: Clearance of host cell proteins from CHO cell culture harvests via "flow-through affinity chromatography" using peptide-based adsorbents. Biotechnol Bioeng , 119 , 1873-1889. 30. Smith, L.M., Kelleher, N.L. and Consortium for Top Down, P. (2013) Proteoform: a single term describing protein complexity. Nat Methods , 10 , 186-187. 31. Smith, L.M. and Kelleher, N.L. (2018) Proteoforms as the next proteomics currency. Science , 359 , 1106-1107. 32. Chumsae, C., Gifford, K., Lian, W., Liu, H., Radziejewski, C.H. and Zhou, Z.S. (2013) Arginine modifications by methylglyoxal: discovery in a recombinant monoclonal antibody and contribution to acidic species. Anal Chem , 85 , 11401-11409. 33. Chumsae, C., Hossler, P., Raharimampionona, H., Zhou, Y., McDermott, S., Racicot, C., Radziejewski, C. and Zhou, Z.S. (2015) When Good Intentions Go Awry: Modification of a Recombinant Monoclonal Antibody in Chemically Defined Cell Culture by Xylosone, an Oxidative Product of Ascorbic Acid. Anal Chem , 87 , 7529-7534. 34. Liu, M., Zhang, Z., Cheetham, J., Ren, D. and Zhou, Z.S. (2014) Discovery and characterization of a photo-oxidative histidine-histidine cross-link in IgG1 antibody utilizing (1)(8)O-labeling and mass spectrometry. Anal Chem , 86 , 4940-4948. 35. Liu, M., Zhang, Z., Zang, T., Spahr, C., Cheetham, J., Ren, D. and Zhou, Z.S. (2013) Discovery of undefined protein cross-linking chemistry: a comprehensive methodology utilizing 18O-labeling and mass spectrometry. Anal Chem , 85 , 5900-5908. 36. Dolan, M.E., Sadiki, A., Wang, L.L., Wang, Y., Barton, C., Oppenheim, S.F. and Zhou, Z.S. (2024) First site-specific conjugation method for native goat IgG antibodies via glycan remodeling at the conserved Fc region. Antib Ther , 7 , 233-248. 37. Dolan, M.E., Wang, L., Wang, Y., Barton, C., Oppenheim, S.F. and Zhou, Z.S. (2025) First Elucidation of Proteoforms of Challenging Host Cell Proteins (HCPs) in Biomanufacturing, Enabled by Affinity Enrichment: Chinese Hamster Ovary (CHO) Phospholipase B-Like 2 (PLBL2) Captured from a Monoclonal Antibody. Authorea . 38. Chivers, C.E., Koner, A.L., Lowe, E.D. and Howarth, M. (2011) How the biotin-streptavidin interaction was made even stronger: investigation via crystallography and a chimaeric tetramer. Biochem J , 435 , 55-63. 39. Martin Vanderlaan, W.S., Peter Liu, Julie Nishihara, George Tsui, Margaret Lin, Feny Gunawan, Sara Parker, Robert Ming Wong, Justin Low, Xiangdan Wang, Jihong Yang, Karthik Veeravalli, Patrick McKay, Chris Yu, Lori O'Connell, Benjamin Tran, Rajesh Vij, Chris Fong, Kathleen Francissen, Judith Zhu-Shimoni, Valerie Quarmby, Denise Krawitz. (2015) Hamster Phospholipase B-Like 2 (PLBL2): A Host-Cell Protein Impurity in Therapeutic Monoclonal Antibodies Derived from Chinese Hamster Ovary Cells. Bioprocess International , 13 , 18, 20, 22-24, 26, 28-29, 55. 40. Deuschl, F., Kollmann, K., von Figura, K. and Lubke, T. (2006) Molecular characterization of the hypothetical 66.3-kDa protein in mouse: lysosomal targeting, glycosylation, processing and tissue distribution. FEBS Lett , 580 , 5747-5752. 41. Jensen, A.G., Chemali, M., Chapel, A., Kieffer-Jaquinod, S., Jadot, M., Garin, J. and Journet, A. (2007) Biochemical characterization and lysosomal localization of the mannose-6-phosphate protein p76 (hypothetical protein LOC196463). Biochem J , 402 , 449-458. 42. Garon, E.B., Rizvi, N.A., Hui, R., Leighl, N., Balmanoukian, A.S., Eder, J.P., Patnaik, A., Aggarwal, C., Gubens, M., Horn, L. et al. (2015) Pembrolizumab for the treatment of non-small-cell lung cancer. N Engl J Med , 372 , 2018-2028. 43. Robert, C., Schachter, J., Long, G.V., Arance, A., Grob, J.J., Mortier, L., Daud, A., Carlino, M.S., McNeil, C., Lotem, M. et al. (2015) Pembrolizumab versus Ipilimumab in Advanced Melanoma. N Engl J Med , 372 , 2521-2532. Information & Authors Information Version history V1 Version 1 09 December 2024 V2 Version 2 12 June 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords affinity resin antibody immobilization charge variants downstream processing goat antibody host cell protein phospholipase b-like 2 (plbl2) proteoforms size variants Authors Affiliations Michael E. Dolan 0009-0009-4497-8725 [email protected] Northeastern University Department of Chemistry and Chemical Biology View all articles by this author Alexander Tedeschi Takeda Development Center Americas Inc View all articles by this author Sheldon F. Oppenheim Takeda Development Center Americas Inc View all articles by this author Zhaohui Sunny Zhou Northeastern University Department of Chemistry and Chemical Biology View all articles by this author Metrics & Citations Metrics Article Usage 804 views 300 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Michael E. Dolan, Alexander Tedeschi, Sheldon F. Oppenheim, et al. Site-Specific Immobilization of Antibodies as a Platform Approach to Enable the Targeted Capture and Effective Removal of “Problematic” Host Cell Proteins (HCPs) from Complex Bioprocess Streams, Even at Sub-ppm Levels: Chinese Hamster Ovary (CHO) Phospholipase B-Like 2 (PLBL2). Authorea . 12 June 2025. DOI: https://doi.org/10.22541/au.173372831.15992718/v2 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.173372831.15992718/v2","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a003cff019514807',t:'MTc3OTUzNjcwNA=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();