Search

Product-related impurities are challenging to remove during monoclonal antibody (mAb) purification process due to molecular similarity. Frontal chromatography on hydrophobic interaction resins has demonstrated its capability to effectively remove such impurities. However, process improvements geared towards purity level comes as a trade-off with the yield loss. In this work, we present a hydrophobic interaction chromatography process using multicolumn continuous chromatography (MCC) concept and frontal analysis to remove a high prevalence product related impurity. This design uses a two-column continuous system where the two columns are directly connected during product chase step to capture product wash loss without any in-process adjustment. This polish MCC operation resulted in a 10 % increase in yield while maintaining 99 % purity, despite the presence of 20 % product-related impurities in the feed material. One challenge associated with polish MCC design is that the accumulation of the impurities renders a non-steady state recycling. To surmount this issue and ensure a robust process, a mechanistic model was developed and validated to predict multicomponent breakthrough. This model was capable to predict multiple cycle behavior and accounts for increased impurity concentration. Assisted by the model, the optimized operation parameters and conditions can be determined to account for variation in product load quality. The simulated results demonstrate an effective doubling of productivity compared to conventional batch chromatography., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024. Published by Elsevier B.V.)

Fc-fusion proteins are an emerging class of protein therapeutics that combine the properties of biological ligands with the unique properties of the fragment crystallizable (Fc) domain of an immunoglobulin G (IgG). Due to their diverse higher-order structures (HOSs), Fc-fusion proteins remain challenging characterization targets within biopharmaceutical pipelines. While high-resolution biophysical tools are available for HOS characterization, they frequently demand extended time frames and substantial quantities of purified samples, rendering them impractical for swiftly screening candidate molecules. Herein, we describe the development of ion mobility-mass spectrometry (IM-MS) and collision-induced unfolding (CIU) workflows that aim to fill this technology gap, where we focus on probing the HOS of a model Fc-Interleukin-10 (Fc-IL-10) fusion protein engineered using flexible glycine-serine linkers. We evaluate the ability of these techniques to probe the flexibility of Fc-IL-10 in the absence of bulk solvent relative to other proteins of similar size, as well as localize structural changes of low charge state Fc-IL-10 ions to specific Fc and IL-10 unfolding events during CIU. We subsequently apply these tools to probe the local effects of glycine-serine linkers on the HOS and stability of IL-10 homodimer, which is the biologically active form of IL-10. Our data reveals that Fc-IL-10 produces significantly more structural transitions during CIU and broader IM profiles when compared to a wide range of model proteins, indicative of its exceptional structural dynamism. Furthermore, we use a combination of enzymatic approaches to annotate these intricate CIU data and localize specific transitions to the unfolding of domains within Fc-IL-10. Finally, we detect a strong positive, quadratic relationship between average linker mass and fusion protein stability, suggesting a cooperative influence between glycine-serine linkers and overall fusion protein stability. This is the first reported study on the use of IM-MS and CIU to characterize HOS of Fc-fusion proteins, illustrating the practical applicability of this approach.

Bispecific antibodies (bsAbs) represent a critically important class of emerging therapeutics capable of targeting two different antigens simultaneously. As such, bsAbs have been developed as effective treatment agents for diseases that remain challenging for conventional monoclonal antibody (mAb) therapeutics to access. Despite these advantages, bsAbs are intricate molecules, requiring both the appropriate engineering and pairing of heavy and light chains derived from separate parent mAbs. Current analytical tools for tracking the bsAb construction process have demonstrated a limited ability to robustly probe the higher-order structure (HOS) of bsAbs. Native ion mobility-mass spectrometry (IM-MS) and collision-induced unfolding (CIU) have proven to be useful tools in probing the HOS of mAb therapeutics. In this report, we describe a series of detailed and quantitative IM-MS and CIU data sets that reveal HOS details associated with a knob-into-hole (KiH) bsAb model system and its corresponding parent mAbs. We find that quantitative analysis of CIU data indicates that global KiH bsAb stability occupies an intermediate space between the stabilities recorded for its parent mAbs. Furthermore, our CIU data identify the hole-containing half of the KiH bsAb construct to be the least stable, thus driving much of the overall stability of the KiH bsAb. An analysis of both intact bsAb and enzymatic fragments allows us to associate the first and second CIU transitions observed for the intact KiH bsAb to the unfolding Fab and Fc domains, respectively. This result is likely general for CIU data collected for low charge state mAb ions and is supported by data acquired for deglycosylated KiH bsAb and mAb constructs, each of which indicates greater destabilization of the second CIU transition observed in our data. When integrated, our CIU analysis allows us to link changes in the first CIU transition primarily to the Fab region of the hole-containing halfmer, while the second CIU transition is likely strongly connected to the Fc region of the knob-containing halfmer. Taken together, our results provide an unprecedented road map for evaluating the domain-level stabilities and HOS of both KiH bsAb and mAb constructs using CIU.

Background: Bispecific antibodies (BsAb) belong to a novel antibody category with advantages over traditional mono-specific therapeutic antibodies. However, product variants are also commonly seen during the production of BsAb, which poses significant challenges to downstream processing. In this study, the adsorptive characteristics of a BsAb product and its variants were investigated for a set of depth filters during primary recovery of the cell culture fluid., Methods: The retention of the BsAb product and its variants on a set of Millistak+® D0HC and X0HC depth filters were first investigated, followed by studying the mechanism of their adsorption on the depth filters. The chemical and structural properties of depth filters along with the molecular properties of the product and its variants were studied subsequently., Results: The X0HC filter was found to be able to retain a significant amount of low molecular weight (LMW) variants along with a low amount of main product retained. Different levels of retention, observed for these variants, were correlated to their different hydrophobic and charge characteristics in relation with the adsorptive properties of the depth filters used. Electrostatic, hydrophobic, and hydrogen bonding interactions were found to be the key forces to keep product variants retained on the depth filter where the higher hydrophobicity of the LMW variants may cause them to be preferentially retained., Conclusion: Harvest depth filters potentially can be utilized for retaining the BsAb variants, which depends on relative molecular properties of the product and its variants and adsorptive properties of the depth filters used., (© The Author(s) 2022. Published by Oxford University Press on behalf of Antibody Therapeutics. All rights reserved. For Permissions, please email: journals.permissions@oup.com.)

Detergent-mediated virus inactivation (VI) provides a valuable orthogonal strategy for viral clearance in mammalian processes, in particular for next-generation continuous manufacturing. Furthermore, there exists an industry-wide need to replace the conventionally employed detergent Triton X-100 with eco-friendly alternatives. However, given Triton X-100 has been the gold standard for VI due its minimal impact on protein stability and high inactivation efficacy, inactivation by other eco-friendly detergents and its impact on protein stability is not well understood. In this study, the sugar-based detergent commonly used in membrane protein purification, n-dodecyl-β- d-maltoside was found to be a promising alternative for VI. We investigated a panel of detergents to compare the relative VI efficacy, impact on therapeutic quality attributes, and clearance of the VI agent and other impurities through subsequent chromatographic steps. Detergent-mediated inactivation and protein stability showed comparable trends to low pH inactivation. Using experimental and modeling data, we found detergent-mediated product aggregation and its kinetics to be driven by extrinsic factors such as detergent and protein concentration. Detergent-mediated aggregation was also impacted by an initial aggregation level as well as intrinsic factors such as the protein sequence and detergent hydrophobicity, and critical micelle concentration. Knowledge gained here on factors driving product stability and VI provides valuable insight to design, standardize, and optimize conditions (concentration and duration of inactivation) for screening of detergent-mediated VI., (© 2022 Wiley Periodicals LLC.)

This study describes the characterization of conjugation sites for a random, lysine conjugated 2-iminothiolane (2-IT) based antibody-drug-conjugate synthesized from an IgG1 antibody and a duocarmycin analog-based payload-linker. Of the 80 putative lysine sites, 78 were found to be conjugated via tryptic peptide mapping and LC-HRMS. Surprisingly, seven cysteine-linked conjugated peptides were also detected resulting from the conjugation of cysteine residues derived from the four inter-chain disulfide bonds during the reaction. This unexpected finding could be attributed to the free thiols of the 2-IT thiolated antibody intermediates and/or the 4-mercaptobutanamide by-product resulting from the hydrolysis of 2-IT. These free thiols could cause the four inter-chain disulfide bonds of the antibody to scramble via intra- or inter-molecular attack. The presence of only pair of non-reactive (unconjugated) lysine residues, along with the four intact intra-chain disulfide bonds, is attributed to their poor accessibility, which is consistent with solvent accessibility modeling analysis. We also discovered a major by-product derived from the hydrolysis of the amidine moiety of the N -terminus conjugate. In contrast, the amidine moiety in lysine-linked conjugates appeared stable. Based on our results, we propose plausible formation mechanisms of cysteine-linked conjugates and the hydrolysis of the N -terminus conjugate, which provide scientific insights that are beneficial to process development and drug quality control.

Viral clearance is an important performance metric for the downstream process of monoclonal antibodies (mAbs) due to its impact on patient safety. Anion exchange chromatography (AEX) has been well-accepted in the industry as one of the workhorse techniques for removing viruses, and is considered to be able to achieve high log clearance values under most operating conditions. However, it is not uncommon for viral clearance results on AEX to fall below the desired level despite operating under conditions that should achieve high clearance levels according to conventional wisdom of how this mode of chromatography operates. In this study, a design of experiment (DoE) approach was used to develop a more fundamental understanding of viral clearance during AEX chromatography using Minute Virus of Mice (MVM) on POROS HQ resin. Load pH, conductivity and virus concentration were evaluated as design factors for three mAbs with varying physical and chemical properties. The hydrophobicity and surface charge distributions of the molecules were found to be the most significant factors in influencing viral clearance performance, and the viral clearance trends did not seem to fit with conventional wisdom. To explain this seemingly unconventional behavior, we propose a new mechanism that suggests that interactions between the mAb and the virus have a major contribution on retention of the virus on the resin. This furthered understanding may help improve the predictability, performance and robustness of viral clearance during AEX chromatography., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2020. Published by Elsevier B.V.)

During the development of a therapeutic monoclonal antibody (mAb-1), the charge variant profile obtained by pH-gradient cation exchange chromatography (CEX) contained two main peaks, each of which exhibited a unique intrinsic fluorescence profile and demonstrated inter-convertibility upon reinjection of isolated peak fractions. Domain analysis of mAb-1 by CEX and liquid chromatography-mass spectrometry indicated that the antigen-binding fragment chromatographed as two separate peaks that had identical mass. Surface plasmon resonance binding analysis to antigen demonstrated comparable kinetics/affinity between these fractionated peaks and unfractionated starting material. Subsequent molecular modeling studies revealed that the relatively long and flexible complementarity-determining region 3 (CDR3) loop on the heavy chain could adopt two discrete pH-dependent conformations: an "open" conformation at neutral pH where the HC-CDR3 is largely solvent exposed, and a "closed" conformation at lower pH where the solvent exposure of a neighboring tryptophan in the light chain is reduced and two aspartic acid residues near the ends of the HC-CDR3 loop have atypical pKa values. The pH-dependent equilibrium between "open" and "closed" conformations of the HC-CDR3, and its proposed role in the anomalous charge variant profile of mAb-1, were supported by further CEX and hydrophobic interaction chromatography studies. This work is an example of how pH-dependent conformational changes and conformation-dependent changes to net charge can unexpectedly contribute to perceived instability and require thorough analytical, biophysical, and functional characterization during biopharmaceutical drug product development.

Pyroglutamic acid (pyroGlu) is commonly observed at the N-terminus of therapeutic monoclonal antibodies. Notably, the term "pyroGlu" refers to a single product that could originate from the cyclization of either an N-terminal glutamine or an N-terminal glutamic acid. This is an important and easily overlooked distinction that has major implications on the charge variant nature of a pyroGlu relative to its uncyclized form. Cyclization of an N-terminal glutamine for instance clearly produces an acidic variant with a lower isoelectric point owing to the loss of the positively charged N-terminal amine. In this report, we demonstrate that cyclization of an N-terminal glutamic acid on the other hand produces a basic variant with a higher isoelectric point contrary to the typical assumption that the simultaneous loss of the N-terminal amine and the carboxylic acid side-chain would negate the formation of a charge variant. The results of our investigation demonstrate the need to consider the relative strengths of the acidic and basic functional groups which are altered when assessing whether the product will be a charge variant. This study also adds new knowledge and experimental evidence to understand charge heterogeneity in monoclonal antibodies., (Copyright © 2019 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.)

When developing purification processes for monoclonal antibodies (mAbs), ensuring the effective removal of high molecular weight (HMW) species is often challenging and labor intensive. In this work, we present a bottom-up characterization approach to achieve streamlined polishing step development as well as a more fundamental understanding of the protein of interest. Prior to physicochemical characterization, in-process HMW species of two IgG4 mAbs (mAb A and mAb B) were isolated via semi-preparative size exclusion chromatography (SEC). Key differences in approximate molecular weight, net charge, and native surface hydrophobicity were then identified using multi-angle light scattering (SEC-MALS), analytical-scale chromatographic screening, isoelectric focusing, and structural aggregation propensity modeling. SEC-MALS revealed two main HMW isoforms for each mAb: dimers and 1.7-mers for mAb A, and tetramers and dimers for mAb B. Analytical-scale chromatographic screening showed promising trends in charge-based separation for mAb A, and hydrophobic-based separation for mAb B. Isoelectric focusing data detected a 30% increase in acidic variants for mAb A HMW species relative to monomer, and a 20% increase in basic variants for mAb B HMW species. Lastly, analytical-scale characterization data was successfully applied to preparative scale purification conditions, producing results highly similar to those observed during analytical characterization of the isolated species. By using this high-throughput approach as a template for preparative-scale process development, key physicochemical differences between aggregate and monomer species were utilized to determine optimal polishing steps for HMW removal., (Copyright © 2019 Elsevier B.V. All rights reserved.)

Tryptophan (Trp) oxidation in proteins leads to a number of events, including changes in color, higher order structure (HOS), and biological activity. We describe here a number of new findings through a comprehensive characterization of 6 monoclonal antibodies (mAbs) following selective oxidation of Trp residues by 2,2'-azobis(2-amidinopropane) dihydrochloride. Fluorescence spectroscopy, in combination with second derivative analysis, demonstrates that the loss of Trp fluorescence intensity is a sensitive indicator of Trp oxidation in mAbs. Size-exclusion chromatography with UV and intrinsic Trp fluorescence detection was demonstrated to be a useful method to monitor Trp oxidation levels in mAbs. Furthermore, the Trp oxidation levels measured by size-exclusion chromatography with UV and intrinsic Trp fluorescence detection were found to be in agreement with the values obtained from tryptic peptide mapping by liquid chromatography with mass spectrometric detection and correlate with the total solvent accessible surface area of the exposed Trp residues from in silico modeling. Finally, near-UV circular dichroism and Raman spectroscopy were used to evaluate the impact of Trp oxidation on HOS and identify specific oxidation products, respectively. This work demonstrates that protein HOS is altered on Trp oxidation in mAbs and multiple spectroscopic markers can be used to monitor the molecule-dependent Trp oxidation behavior., (Copyright © 2019 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.)

The higher order structure (HOS) of proteins plays a critical role in the efficacy and stability of biological drugs. Perturbation of the regional structure of proteins can affect biological activity and cause instability. Characterization of HOS has become an integral part of biological drug development and is expected from regulatory agencies. The commonly used techniques for HOS characterization, such as circular dichroism, Fourier-transform infrared, differential scanning calorimetry, intrinsic fluorescence, and hydrogen-deuterium exchange mass spectrometry, have their limitations ranging from lack of sensitivity and specificity to the need of high-level expertise and poor access to instrumentation due to high cost. In this study, we demonstrated a novel controlled proteolysis-based LC-QDa method for the detection of HOS change. By digesting proteins directly without denaturation and reduction, the HOS information can be revealed through the digested peptides. After optimizing the digestion conditions and the detection procedures, we identified 13 signature peptides that can monitor various antibody domains for any HOS changes caused by external stress. By comparing the peptide peak areas between unknown samples and a native control sample, any regional structural changes in unknown samples can be detected. The method was subsequently applied to a wide range of forced degradation samples to demonstrate higher sensitivity compared to the near-UV CD method that is frequently used for monitoring tertiary structural changes. By further reducing the number of signature peptides to five and optimizing liquid chromatography gradient duration, a streamlined, high-throughput, and controlled proteolysis method was successfully established. This method can be used to support process and formulation development as well as potentially for stability testing.

Protein higher order structure (HOS) is an essential quality attribute to ensure protein stability and proper biological function. Protein HOS characterization is performed during comparability assessments for product consistency as well as during forced degradation studies for structural alteration upon stress. Circular dichroism (CD) spectroscopy is a widely used technique for measuring protein HOS, but it remains difficult to assess HOS with a high degree of accuracy and precision. Moreover, once spectral changes are detected, interpreting the differences in terms of specific structural attributes is challenging. Spectral normalization by the protein concentration remains one of the largest sources of error and reduces the ability to confidently detect differences in CD spectra. This work develops a simple method to enhance the precision of the CD spectral measurements through normalization of the CD spectra by the protein concentration determined directly from the CD measurement. This method is implemented to successfully detect small CD spectral changes in multiple forced degradation studies as well as comparability assessments during biologics drug development. Furthermore, the interpretation of CD spectral changes in terms of HOS differences are provided based on orthogonal data in conjunction with structural insights gained through in silico homology modeling of the protein structure., (Copyright © 2018 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.)

Capillary gel electrophoresis using sodium dodecyl sulfate (CE-SDS) is used commercially to provide quantitative purity data for therapeutic protein characterization and release. In CE-SDS, proteins are denatured under reducing or nonreducing conditions in the presence of SDS and electrophoretically separated by molecular weight and hydrodynamic radius through a sieving polymer matrix. Acceptable performance of this method would yield protein peaks that are baseline resolved and symmetrical. Nominal CE-SDS conditions and parameters are not optimal for all therapeutic proteins, specifically for Recombinant Therapeutic Protein-1 (RTP-1), where acceptable resolution and peak symmetry were not achieved. The application of longer alkyl chain detergents in the running buffer matrix substantially improved assay performance. Matrix running buffer containing sodium hexadecyl sulfate (SHS) increased peak resolution and plate count 3- and 8-fold, respectively, compared to a traditional SDS-based running gel matrix. At Bristol-Myers Squibb (BMS), we developed and qualified a viable method for the characterization and release of RTP-1 using an SHS-containing running buffer matrix. This work underscores the potential of detergents other than SDS to enhance the resolution and separation power of CE-based separation methods.

Cation-exchange chromatography (CEX) of a structurally unstable Fc-fusion protein exhibited multi-peak elution profile upon a salt-step elution due to protein aggregation during intra-column buffer transition where low pH and high salt coexisted. The protein exhibited a single-peak elution behavior during a pH-step elution; nevertheless, the levels of soluble aggregates (i.e. high molecular weight species, HMW) in the CEX eluate were still found up to 12-fold higher than that for the load material. The amount of the aggregates formed upon the pH-step elution was dependent on column loading with maximum HMW achieved at intermediate loading levels, supporting the hypothesis that the aggregation was the result of both the conformational changes of the bound protein and the solution concentration of the aggregation-susceptible proteins during elution. Factors such as high load pH, short protein/resin contact time, hydrophilic resin surface, and weak ionizable ligand were effective, to some extent, to reduce aggregate formation by improving the structural integrity of the bound protein. An orthogonal technique, differential scanning fluorimetry (DSF) using Sypro Orange dye confirmed that the bound protein exposed more hydrophobic area than the native molecule in free solution, especially in the pH 4-5 range. The Sypro Orange dye study of resin surface property also demonstrated that the poly[styrene-divinylbenzene]-based Poros XS with polyhydroxyl surface coating is more hydrophobic compared to the agarose-based CM Sepharose FF and SP Sepharose FF. The hydrophobic property of Poros XS contributed to stronger interactions with the partially unfolded bound protein and consequently to the higher aggregate levels seen in Poros XS eluate. This work also investigates the aggregation reversibility in CEX eluate where up to 66% of the aggregates were observed to dissociate into native monomers over a period of 120h, and links the aggregate stability to such conditions as resin surface properties and charged ligand type. Experimental data was correlated semi-quantitatively with theoretical protein charge and hydrophobicity calculations using homology modeling within the BIOVIA Discovery Studio software. Finally, an arginine-sulphopropyl (Arg-SP) agarose resin immobilized with multi-functional ligands was prepared to verify the proposed hypothesis and to eliminate the aggregate formation. The findings of this work provide general insights in understanding aggregate formation and dissociation for structurally unstable proteins in the CEX step., (Copyright © 2016 Elsevier B.V. All rights reserved.)

We report the use of molecular modeling to predict the oxidation propensity of methionine residues in proteins. Oxidation of methionine to the sulfoxide form is one of the major degradation pathways for therapeutic proteins. Oxidation can occur during production, formulation, or storage of pharmaceuticals and it often reduces or eliminates biological activity. We use a molecular model based on atomistic simulations called 2-shell water coordination number to predict the oxidation rates for several model proteins and therapeutic candidates. In addition, we implement models that are based on static and simulation average of the solvent-accessible area (SAA) for either the side chain or the sulfur atom in the methionine residue. We then compare the results from the different models against the experimentally measured relative rates of methionine oxidation. We find that both the 2-shell model and the simulation-averaged SAA models are accurate in predicting the oxidation propensity of methionine residues for the proteins tested. We also find the appropriate parameter ranges where the models are most accurate. These models have significant predictive power and can be used to enable further protein engineering or to guide formulation approaches in stabilizing the unstable methionine residues., (© 2015 Wiley Periodicals, Inc. and the American Pharmacists Association.)

Determining the aggregation propensity of protein-based biotherapeutics is an important step in the drug development process. Typically, a great deal of data collected over a large period of time is needed to estimate the aggregation propensity of biotherapeutics. Thus, candidates cannot be screened early on for aggregation propensity, but early screening is desirable to help streamline drug development. Here, we present a simple molecular computational method to predict the aggregation propensity via hydrophobic interactions, thought to be the most common mechanism of aggregation, and electrostatic interactions. This method uses a new quantity termed Developability Index. It is a function of an antibody's net charge, calculated on the full-length antibody structure, and the spatial aggregation propensity, calculated on the complementarity-determining region structure. Its accuracy is due to the molecular level details and the incorporation of the tertiary structure of the antibody. It is particularly applicable to antibodies or other proteins for which structures are available or could be determined accurately using homology modeling. Applications include the selection of molecules in the discovery or early development process, selection of mutants for stability, and estimation of resources needed for development of a given biomolecule., (Copyright © 2011 Wiley-Liss, Inc.)

The lack of a fast selection method to identify the most stable protein is one of the major challenges for developing successful therapeutic protein formulations more rapidly. The swift and accurate detection of small amounts of aggregates is another problem since aggregates may trigger an immunological response and the aggregation decreases the biological activity of the antibody. Here we present an alternative method for initial screening of the aggregation propensity of proteins, using monoclonal antibodies (mAb) as an example and thioflavin T (ThT) binding. The major advantage of ThT binding is the short duration of testing compared with size-exclusion chromatography (SEC) measurements that can take 6 months or more even under accelerated conditions. The tendency to aggregate of each therapeutic human mAb probed with the ThT assay, together with SEC, is employed to formulate the ranking of mAb aggregation. ThT binding can determine the propensity of proteins to aggregate in a few days, illustrating that ThT binding would be a valuable screening tool., (© 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.)

Characterization of aggregation profiles of monoclonal antibodies (mAb) is gaining importance because an increasing number of mAb-based therapeutics are entering clinical studies and gaining marketing approval. To develop a successful formulation, it is imperative to identify the critical biochemical properties of each potential mAb drug candidate. We investigated the conformational change and aggregation of a human IgG1 using external dye-binding experiments with fluorescence spectroscopy and compared the aggregation profiles obtained to the results of size-exclusion chromatography. We show that using an appropriate dye at selected mAb concentration, unfolding or aggregation can be studied. In addition, dye-binding experiments may be used as conventional assays to study therapeutic mAb stability.

Understanding antibody aggregation is of great significance for the pharmaceutical industry. We studied the aggregation of five different therapeutic monoclonal antibodies (mAbs) with size-exclusion chromatography-high-performance liquid chromatography (SEC-HPLC), fluorescence spectroscopy, electron microscopy, and light scattering methods at various temperatures with the aim of gaining insight into the aggregation process and developing models of it. In particular, we find that the kinetics can be described by a second-order model and are non-Arrhenius. Thus, we develop a non-Arrhenius model to connect accelerated aggregation experiments at high temperature to long-term storage experiments at low temperature. We evaluate our model by predicting mAb aggregation and comparing it with long-term behavior. Our results suggest that the number of monomers and mAb conformations within aggregates vary with the size and age of the aggregates, and that only certain sizes of aggregates are populated in the solution. We also propose a kinetic model based on conformational changes of proteins and monomer peak loss kinetics from SEC-HPLC. This model could be employed for a detail analysis of mAb aggregation kinetics., (Copyright © 2011 Wiley-Liss, Inc. and the American Pharmacists Association)

Identifying protein binding sites provides important clues to the function of a protein. Experimental methods to identify the binding sites such as determining the crystal structures of protein complexes are extremely laborious and expensive. Here, we present a computational technique called spatial aggregation propensity (SAP) based on molecular simulations to predict protein binding sites. We apply this technique to two model proteins, an IgG1 antibody and epidermal growth factor receptor (EGFR) and demonstrate that SAP predicts protein binding regions with very good accuracy. In the case of the IgG1 antibody, SAP accurately predicts binding regions with the Fc-receptor, protein-A, and protein-G. For EGFR, SAP accurately predicts binding regions with EGF, TGFα, and with another EGFR. The resolution of SAP is varied to obtain a detailed picture of these binding sites. We also show that some of these binding sites overlap with protein self-aggregation prone regions. We demonstrate how SAP analysis can be used to engineer the protein to remove unfavorable aggregation prone regions without disturbing protein binding regions. The SAP technique could be also used to predict the yet unknown binding sites of numerous proteins, thereby providing clues to their function., (Copyright © 2010 Wiley-Liss, Inc.)

Intrinsic tryptophan (Trp) fluorescence is often used to determine conformational changes of proteins. The fluorescence of multi-Trp proteins is generally assumed to be additive. This assumption usually holds well if Trp residues are situated at long distances from each other in the absence of any excited state reactions involving these residues and therefore when energy transfer does not occur. Here, we experimentally demonstrate energy transfer among Trp residues and support it by a Master Equation kinetic model applied to a therapeutic monoclonal antibody (mAb). The mAbs are one of the most studied and important biologics for the pharmaceutical industry, and they contain many Trp residues in close proximity. Understanding mAb fluorescence is critical for interpreting fluorescence data and protein-structure relationships. We propose that Trp residues could be categorized into three types of emitters in the mAbs. Experimentally, we categorize them according to solvent accessibility based on dependence of their fluorescence lifetime on the external quencher concentration and their emission wavelength. Theoretically, we categorize with molecular dynamics simulations according to their solvent accessibility. This method of combinatorial mapping of fluorescence characteristics can be utilized to illuminate structural aspects as well as make comparisons of drug formulations for these pharmaceutical proteins.

Monoclonal antibodies are the fastest growing class of biologics in the pharmaceutical industry. The correlation between mAb glycosylation and aggregation has not been elucidated in detail, yet understanding the structure-stability relationship involving glycosylation is critical for developing successful drug formulations. We conducted studies of temperature-induced aggregation and compared the stability of both glycosylated and aglycosylated forms of a human IgG1. In parallel, we also performed molecular dynamics simulations of the glycosylated full antibody to gain an understanding of the polysaccharide surroundings at the molecular level. Aglycosylated mAbs are somewhat less stable and therefore aggregate more easily than the glycosylated form at the temperatures studied. Glycosylation seems to enhance solubility and stability of these therapeutics and thus might be important for long-term storage.

Therapeutic proteins such as antibodies are playing an increasingly prominent role in the treatment of numerous diseases including cancer and rheumatoid arthritis. However, these proteins tend to degrade due to aggregation during manufacture and storage. Aggregation decreases protein activity and raises concerns about an immunological response. We have recently developed a method based on full antibody atomistic simulations to predict antibody aggregation prone regions [Proc. Natl. Acac. Sci. 2009, 106, 11937]. This method is based on "spatial-aggregation-propensity (SAP)", a measure of the dynamic exposure of hydrophobic patches. In the present paper, we expand on this method to analyze the aggregation prone regions over a wide parameter range. We also explore the effect of different hydrophilic mutations on these predicted aggregation prone regions to engineer antibodies with enhanced stability. The mutation to lysine is more effective than serine but less effective than glutamic acid in enhancing antibody stability. Furthermore, we show that multiple simultaneous mutations on different SAP peaks can have a cumulative effect on enhancing protein stability. We also investigate the accuracy of various cheaper alternatives for SAP evaluation because the full antibody atomistic simulations are highly computationally expensive. These cheaper alternatives include antibody fragment (Fab, Fc) simulations, implicit solvent models, or direct computations from a static structure (i.e., a structure from X-ray or homology modeling). The SAP evaluation from the static structure is 200,000 times faster but less accurate compared to the SAP from explicit atom simulations. Nevertheless, the SAP from a static structure still predicts most of the major aggregation prone regions, making it a potential approach for use in high-throughput applications. Thus, the SAP technology described here could be employed either in high-throughput developability screening of therapeutic protein candidates or to improve their stability at later stages of manufacturing.

Antibodies are multidomain proteins that are extensively used as a research tool in molecular biology and as therapeutics in medicine. In many cases, antibodies are engineered to contain surface cysteines for the site-specific conjugation of payloads. These antibodies can serve as payload vehicles in targeting a diseased cell to which the conjugated molecules exercise their activity. Here, we design and analyze a set of fourteen new IgG1 cysteine variants, with at least one variant per immunoglobulin fold domain. The cross-linking propensity of these mutants correlates very well with a tool we have developed for measuring aggregation propensity in silico, called spatial aggregation propensity (SAP). Our results indicate the utility of the SAP technology in selecting antibody cysteine variants with desired properties. Moreover, the different oligomerization propensity of the variants suggests a variety of applications in molecular biology and medicine, such as payload delivery, structural analysis, electrophoresis, and chromatography.

Protein-carbohydrate interactions are important for glycoprotein structure and function. Antibodies of the IgG class, with increasing significance as therapeutics, are glycosylated at a conserved site in the constant Fc region. We hypothesized that disruption of protein-carbohydrate interactions in the glycosylated domain of antibodies leads to the exposure of aggregation-prone motifs. Aggregation is one of the main problems in protein-based therapeutics because of immunogenicity concerns and decreased efficacy. To explore the significance of intramolecular interactions between aromatic amino acids and carbohydrates in the IgG glycosylated domain, we utilized computer simulations, fluorescence analysis, and site-directed mutagenesis. We find that the surface exposure of one aromatic amino acid increases due to dynamic fluctuations. Moreover, protein-carbohydrate interactions decrease upon stress, while protein-protein and carbohydrate-carbohydrate interactions increase. Substitution of the carbohydrate-interacting aromatic amino acids with non-aromatic residues leads to a significantly lower stability than wild type, and to compromised binding to Fc receptors. Our results support a mechanism for antibody aggregation via decreased protein-carbohydrate interactions, leading to the exposure of aggregation-prone regions, and to aggregation.

Monoclonal antibodies represent the fastest growing class of pharmaceuticals. A major problem, however, is that the proteins are susceptible to aggregation at the high concentration commonly used during manufacturing and storage. Our recent publication describes a technology based on molecular simulations to identify aggregation-prone regions of proteins in silico. The technology, called spatial aggregation propensity (SAP), identifies hot-spots for aggregation based on the dynamic exposure of spatially-adjacent hydrophobic amino acids. Monoclonal antibodies (mAbs) in which patches with high-SAP scores are changed to patches with significantly reduced SAP scores via a single mutation are more stable than wild type, thus validating the SAP method for mapping aggregation-prone regions on proteins. We propose that the SAP technology will be useful for protein stabilization, and as a screening tool to bridge discovery and development of protein-based therapeutics by a rational assessment of the developability of candidate protein drugs.

Therapeutic antibodies of many different IgG subclasses (IgG1, IgG2 and IgG4) are used in the treatment of various cancers, rheumatoid arthritis and other inflammatory and infectious diseases. These antibodies are stored for long durations under high concentrations as required in the disease treatment. Unfortunately, these antibodies aggregate under these storage conditions, leading to a decrease in antibody activity and raising concerns about causing an immunological response. Thus, there is a tremendous need to identify the aggregation-prone regions in different classes of antibodies. We use the SAP (spatial-aggregation-propensity) technology based on molecular simulations to determine the aggregation-prone motifs in the constant regions of IgG1 classes of antibodies. Mutations engineered on these aggregation-prone motif regions led to antibodies of enhanced stability. Fourteen aggregation-prone motifs are identified, with each motif containing one to seven residues. While some of these motifs contain residues that are neighbors in primary sequence, others contain residues that are far apart in primary sequence but are close together in the tertiary structure. Comparison of the IgG1 sequence with those of other subclasses (IgG2, IgG3 and IgG4) showed that these aggregation-prone motifs are largely preserved among all IgG subclasses. Other broader classes of antibodies (IgA1, IgD, IgE and IgM), however, differed in these motif regions. The aggregation-prone motifs identified were therefore common to all IgG subclasses, but differ from those of non-IgG classes. Moreover, since the motifs identified are in the constant regions, they are applicable for all antibodies within the IgG class irrespective of the variable region. Thus, the motif regions identified could be modified on all IgGs to yield antibodies of enhanced stability.

Therapeutic proteins such as antibodies constitute the most rapidly growing class of pharmaceuticals for use in diverse clinical settings including cancer, chronic inflammatory diseases, kidney transplantation, cardiovascular medicine, and infectious diseases. Unfortunately, they tend to aggregate when stored under the concentrated conditions required in their usage. Aggregation leads to a decrease in antibody activity and could elicit an immunological response. Using full antibody atomistic molecular dynamics simulations, we identify the antibody regions prone to aggregation by using a technology that we developed called spatial aggregation propensity (SAP). SAP identifies the location and size of these aggregation prone regions, and allows us to perform target mutations of those regions to engineer antibodies for stability. We apply this method to therapeutic antibodies and demonstrate the significantly enhanced stability of our mutants compared with the wild type. The technology described here could be used to incorporate developability in a rational way during the screening of antibodies in the discovery phase for several diseases.

This paper is devoted to equilibrium molecular-dynamics (MD) simulations of a fully atomistic model of binary mixtures of water (component 1) and ethanol (component 2). We investigate ways to extract from these simulations effective, pairwise additive potentials suitable to describe the interactions between coarse-grained molecules (i.e., beads) in corresponding mesoscale dissipative particle-dynamics simulations. The fully atomistic model employed in MD simulations is mapped onto an implicit water model, where the internal degrees of freedom of ethanol and all the degrees of freedom of water are integrated out. This gives us an effective one-component system consisting only of ethanol beads. The effective interaction potential between a pair of ethanol beads, Phi(R), is approximated at three levels of sophistication. At the lowest one, we approximate Phi(R) by the potential of mean force between the centers of mass of two ethanol beads calculated in the fully atomistic MD simulations; at the second level, we take Phi(R) to be the potential linked to total and direct correlation functions in the hypernetted-chain closure of the Ornstein-Zernike equation. At the third level we approximate Phi(R) numerically by improving it iteratively through the Boltzmann inversion scheme. Our results indicate that the level-one approach works only at the lowest (8 wt %) concentration; the level-two approach works only up to intermediate ethanol concentrations (ca. 50 wt %). Only the Boltzmann inversion scheme works for all, up to the highest concentration considered (70 wt %).

The impact of alcohol additives on the self-assembly of surfactants in supercritical carbon dioxide is investigated using lattice Monte Carlo simulations. We observe that all studied (model) alcohols reduce the critical micelle concentration. The reduction is stronger the longer the hydrocarbon chain of the alcohol, and the higher the alcohol concentration. Short-chain alcohols are found to concentrate in the surfactant layer of the aggregates, replacing surfactant molecules and leading to a strong decrease of the aggregation number and a large increase of the number of aggregates. On the other hand, only a small number of alcohol molecules with longer chain length are found in the aggregates, leading to a slight increase in the aggregation number. However, structural properties such as size and density profiles of aggregates at the same aggregation number are not influenced markedly. Consequently, short-chain alcohols act as cosurfactants, directly influencing the properties of the aggregates, while alcohols with longer hydrocarbon chains work as cosolvents, altering the properties of the solvent. However, the transition between both extremes is gradual.