Spatiotemporal Control of Protein Refolding through Flash-Change Reaction Conditions.

Recombinant protein production is an essential aspect of biopharmaceutical manufacturing, with Escherichia coli serving as a primary host organism. Protein refolding is vital for protein production; however, conventional refolding methods face challenges such as scale-up limitations and difficulties in controlling protein conformational changes on a millisecond scale. In this study, we demonstrate the novel application of flow microreactors (FMR) in controlling protein conformational changes on a millisecond scale, enabling efficient refolding processes and opening up new avenues in the science of FMR technology. FMR technology has been primarily employed for small-molecule synthesis, but our novel approach successfully expands its application to protein refolding, offering precise control of the buffer pH and solvent content. Using interleukin-6 as a model, the system yielded an impressive 96% pure refolded protein and allowed for gram-scale production. This FMR system allows flash changes in the reaction conditions, effectively circumventing protein aggregation during refolding. To the best of our knowledge, this is the first study to use FMR for protein refolding, which offers a more efficient and scalable method for protein production. The study results highlight the utility of the FMR as a high-throughput screening tool for streamlined scale-up and emphasize the importance of understanding and controlling intermediates in the refolding process. The FMR technique offers a promising approach for enhancing protein refolding efficiency and has demonstrated its potential in streamlining the process from laboratory-scale research to industrial-scale production, making it a game-changing technology in the field.

Electrophoresis, a transport technology that transitioned from moving boundary method to zone method.

Gel electrophoresis, a transport technology, is one of the most widely used experimental methods in biochemical and pharmaceutical research and development. Transport technologies are used to determine hydrodynamic or electrophoretic properties of macromolecules. Gel electrophoresis is a zone technology, where a small volume of sample is applied to a large separation gel matrix. In contrast, a seldom-used electrophoresis technology is moving boundary electrophoresis, where the sample is present throughout the separation phase or gel matrix. While the zone method gives peaks of separating macromolecular solutes, the moving boundary method gives a boundary between solute-free and solute-containing phases. We will review electrophoresis as a transport technology of zone and moving boundary methods and describe its principles and applications.

Alternative downstream processes for production of antibodies and antibody fragments.

Protein-A or Protein-L affinity chromatography and virus inactivation are key processes for the manufacturing of therapeutic antibodies and antibody fragments. These two processes often involve exposure of therapeutic proteins to denaturing low pH conditions. Antibodies have been shown to undergo conformational changes at low pH, which can lead to irreversible damages on the final product. Here, we review alternative downstream approaches that can reduce the degree of low pH exposure and consequently damaged product. We and others have been developing technologies that minimize or eliminate such low pH processes. We here cover facilitated elution of antibodies using arginine in Protein-A and Protein-G affinity chromatography, a more positively charged amidated Protein-A, two Protein-A mimetics (MEP and Mabsorbent), mixed-mode and steric exclusion chromatography, and finally enhanced virus inactivation by solvents containing arginine. This article is part of a Special Issue entitled: Recent advances in molecular engineering of antibody.

Non-Affinity Purification of Antibodies.

Currently, purification of antibodies is mainly carried out using a platform technology composed primarily of Protein A chromatography as a capture step, regardless of the scale. However, Protein A chromatography has a number of drawbacks, which are summarized in this review. As an alternative, we propose a simple small-scale purification protocol without Protein A that uses novel agarose native gel electrophoresis and protein extraction. For large-scale antibody purification, we suggest mixed-mode chromatography that can in part mimic the properties of Protein A resin, focusing on 4-Mercapto-ethyl-pyridine (MEP) column chromatography.

Mirror-image streptavidin with specific binding to L-biotin, the unnatural enantiomer.

The streptavidin-biotin system is known to have a very high affinity and specificity and is widely used in biochemical immunoassays and diagnostics. However, this method is affected by endogenous D-biotin in serum sample measurements (biotin interference). While several efforts using alternative high-affinity binding systems (e.g., genetically modified streptavidin and biotin derivatives) have been attempted, these efforts have all led to reduction in affinity. To solve this interference issue, the enantiomer of streptavidin was synthesized, which enabled specific binding to L-biotin. We successfully obtained a functional streptavidin molecule by peptide synthesis using D-amino acids and an in vitro folding technique. Several characterizations, including size exclusion chromatography (SEC), circular dichroism spectra (CD), and heat denaturation experiments collectively confirmed the higher-order enantiomer of natural streptavidin had been formed with comparable stability to the natural protein. L-biotin specific binding of this novel molecule enabled us to avoid biotin interference in affinity measurements using the Biacore system and enzyme-linked immunosorbent assay (ELISA). We propose the enantiomer of streptavidin as a potential candidate to replace the natural streptavidin-biotin system, even for in vivo use.

A novel protein refolding system using lauroyl-l-glutamate as a solubilizing detergent and arginine as a folding assisting agent.

More than 50 detergents, including acylated amino acid derivatives, were screened for their ability to solubilize and refold recombinant proteins expressed as inclusion bodies. Two model proteins, human interleukin-6 and microbial transglutaminase, were solubilized by these detergents and the solubilized proteins were rapidly diluted for testing their solubilization and refolding effectiveness. Long chain-acylated amino acid derivatives having dicarboxylic acid moieties were found to be superior to others under the conditions tested. In particular, lauroyl-l-glutamate (C12-l-Glu) showed the highest recovery of the native proteins. The effectiveness of dilution refolding was greatly improved by adding aggregation suppressive arginine into the refolding solvents. To gain understanding how this detergent works, interactions between detergents and proteins were examined using spectroscopic and native gel electrophoretic analyses, showing ideal properties for C12-l-Glu as a solubilzing agent, i.e. highly reversible nature of the detergent binding to the model globular proteins and of the conformational changes. These properties most likely have contributed to the effective protein solubilzation and refolding of inclusion bodies using C12-l-Glu and arginine.

Refolding single-chain antibody (scFv) using lauroyl-L-glutamate as a solubilization detergent and arginine as a refolding additive.

Therapeutic potential of immunoconjugates has opened a new window for antibody-based biopharmaceuticals. Greater tissue penetration and hence enhanced cell toxicity are obtained with a smaller version of antibodies. While the whole antibody can be readily produced via mammalian expression system, antibody fragments often require refolding of insoluble proteins. Here we report a new refolding method for antibody fragments using a novel amino acid-based detergent as a solubilizing agent and arginine-assisted refolding. Inclusion bodies of antibody fragments were solubilized by 2.5% lauroyl-L-Glu (C12-L-Glu) and successfully refolded by multi-step dilution into a buffer solution containing arginine hydrochloride and thiol/disulfide-exchange reagents. Adjustment of temperature was found to be critical for increase in the refolding yield. Although each protein requires appropriate optimization, solubilization by C12-L-Glu and dilution refolding assisted by arginine can generate the native, functional antibody fragments. The procedure should enable us to utilize bacterial expression systems for the large-scale manufacturing.

Screening of effective column rinse solvent for Protein-A chromatography.

Mildly acidic arginine solution is highly effective in elution of bound proteins from Protein-A columns. Although Protein-A is specific in antibody capture, it does bind other proteins, which must then be removed before elution by aqueous arginine solution. If they are not removed, a strong elution property of aqueous arginine solutions will elute the contaminating proteins along with antibodies. Here we have examined various salt solutions as a column rinse solvent. We screened various solvents for their effects on binding of purified antibodies to Protein-A, instead of their effectiveness to elute the bound contaminants. Those solvents that result in a slight flow-through of the antibodies during loading should be effective in eluting non-specifically bound proteins that have weaker affinity for Protein-A than antibodies: namely, if a particular solvent reduces antibody binding to Protein-A, it is expected to be effective in reducing binding of contaminants and hence eluting them. Such screening showed a few compounds, including arginine and sodium acetate, as potential column rinse agents. A combination of arginine and sodium acetate was tested for a few crude materials containing antibodies.

MEP HyperCel chromatography II: binding, washing and elution.

Binding, washing and elution conditions for MEP HyperCel chromatography were examined using two conditioned media (CM) containing monoclonal antibodies (humanized IgG1) and bovine serum albumin (BSA). Monoclonal antibodies derived from mammalian expression system bound to the column without pretreatment, although a majority of contaminating proteins present in the CM also showed binding. Inorganic salts, ethanol and glycerol were ineffective in eluting proteins under the conditions examined, suggesting that electrostatic or hydrophobic interactions are not a major factor for antibody binding to the MEP resin. Ethylene glycol, 2-propanol, urea and arginine were effective, to varying degrees, in elution of the bound proteins. The bound contaminating proteins and BSA were effectively eluted with ethylene glycol and the bound antibodies were finally eluted with aqueous arginine solutions at neutral pH. MEP showed selectivity toward BSA and hence utility for removing BSA from the samples. Interestingly, Fc-fusion proteins derived from silkworm larvae showed no detectable binding. Serum proteins present in silkworm larvae strongly competed with the Fc-fusion proteins and monoclonal antibody for binding to MEP resin, while the same Fc-fusion proteins can be readily purified in one-step by Protein-A resin, again confirming weak selectivity of the MEP resin.

MEP chromatography of antibody and Fc-fusion protein using aqueous arginine solution.

MEP HyperCel resin, one of the Protein-A mimetic columns, is designed to bind antibodies at physiological pH and elutes the bound antibodies at mildly acidic pH. We have tested aqueous arginine solution for washing and elution of the resin. To our surprise, bound antibody and Fc-fusion protein eluted at pH 7.0 using 1M arginine solution. Various solvent additives were then examined at pH 7.0. Among the tested additives, urea and arginine were the only additives that were effective in elution. Thus, urea and arginine at low concentrations were effectively used for washing the resin. NaCl and MgCl(2) at 0.1-1M and ethanol at 5-20% were not effective. Based on these observations, it appears that protein binds to MEP resin through both polar and hydrophobic interactions with some contribution of electrostatic interaction, which can be simultaneously reduced by arginine or urea. On the other hand, Mabsorbent, another Protein-A mimetic column, appears to be more non-specific and non-selective.

Utilization of Arg-elution method for FLAG-tag based chromatography.

FLAG-tag is one of the commonly used purification technologies for recombinant proteins. An antibody, M2, specifically binds to the FLAG-tag whether it is attached to N- or C-terminus of proteins to be purified. The bound proteins are generally eluted by competition with a large excess of free FLAG peptide. This requires synthetic FLAG peptide and also removal of bound FLAG peptide for M2 column regeneration. We have shown before that arginine at mild pH can effectively dissociate protein-protein or protein-ligand interactions, e.g. in Protein-A, antigen and dye-affinity chromatography. We have tested here elution of FLAG-fused proteins by arginine for columns of M2-immobilized resin using several proteins in comparison with competitive elution by FLAG peptide or low pH glycine buffer. Active and folded proteins were successfully and effectively eluted using 0.5-1M arginine at pH 3.5-4.4, as reported in this paper.

Co-operative thermal inactivation of herpes simplex virus and influenza virus by arginine and NaCl.

Elevated temperatures have been used to inactivate viruses for plasma-derived biopharmaceuticals. This paper describes the effects of arginine and NaCl in conjunction with elevated temperature for inactivation of two enveloped viruses, i.e., herpes simplex virus type 1 (HSV-1) and influenza virus type A at neutral pH. In phosphate-buffered saline, a significant inactivation of HSV-1 occurred above 40 degrees C, resulting in less than 10% surviving virus (over 90% virus inactivation) at 50 degrees C. Arginine concentration dependently decreased the temperature required for virus inactivation, leading to temperature shift by almost 17 degrees C at 1.2M. NaCl also decreased the inactivation temperature, but to a considerably lesser extent, indicating that virus inactivation effect of arginine is not simply due to ionic strength. Influenza virus was also inactivated by high temperature, but its responses to arginine and NaCl were different from those on HSV-1, suggesting that virus inactivation mechanism is different between these two viruses, i.e., the effects of these reagents are virus specific.

Protein aggregation suppressor arginine as an effective mouth cleaning agent.

We have tested here whether or not arginine, a well-known aggregation suppressor, is effective in removing bacterial cells, which may present a potential risk of accidental pneumonia infection in aged individuals, from the oral mucosal membranes. This is based on the ability of arginine to suppress protein-protein interaction and surface adsorption and increase the solubility of organic compounds. Twelve student volunteers were subjected to mouthwashes with saline, citrate buffer (pH 3.5), arginine (pH 3.5) and a commercial Listerine. Insignificant effects were observed with saline and citrate buffer, whereas arginine and Listerine mouthwashes led to significant reduction of bacterial cells from the dorsal side of the volunteer's tongue. Arginine also appeared to disrupt biofilms present in the mouth.

Structure-based analysis reveals hydration changes induced by arginine hydrochloride.

Arginine hydrochloride has been used to suppress protein aggregation during refolding and in various other applications. We investigated the structure of hen egg-white lysozyme (HEL) and solvent molecules in arginine hydrochloride solution by X-ray crystallography. Neither the backbone nor side-chain structure of HEL was altered by the presence of arginine hydrochloride. In addition, no stably bound arginine molecules were observed. The number of hydration water molecules, however, changed with the arginine hydrochloride concentration. We suggest that arginine hydrochloride suppresses protein aggregation by altering the hydration structure and the transient binding of arginine molecules that could not be observed.

Solvent modulation of column chromatography.

A majority of column chromatographies use only selected salts, e.g., ammonium sulfate, NaCl, Citrate and phosphate in hydrophobic interaction chromatography (HIC) and NaCl in ion exchange and dye affinity chromatographies. Alternatively, a pH range below or above the neutral value is often used to reduce affinity interactions, e.g., in Protein-A or dye affinity column chromatography. Although these parameters are easily manipulated, they are not necessarily the optimal conditions for high recovery and resolution of the proteins. So-called co-solvents have been used, although to a limited extent, to manipulate performance of column chromatography. Here the term co-solvent is used to indicate its relatively high concentrations required for these applications, meaning that it also serves as solvent along with water. Ethylene glycol and MgCl(2) have been used to elute specific antibodies from antigen-affinity column. Arginine has also been used for the same purpose. Arginine has much wider applications for various column chromatographies, including size exclusion chromatography (SEC), HIC and affinity chromatography. Polyethylene glycol and glycine have also been used to improve the performance of HIC and hydroxyapatite chromatography. This review summarizes these applications of co-solvents for column chromatographies.

Butyroyl-arginine as a potent virus inactivation agent.

Virus inactivation is a critical step in the manufacturing of recombinant therapeutic proteins, in particular antibodies, using mammalian expression systems. We have shown in the previous paper that arginine is effective in inactivation of herpes simplex virus type 1 (HSV-1) and influenza virus at low temperature under mildly acidic pH, i.e., above pH 4.0; above this pH, conformational changes of most antibodies are negligible. We have here extended virus inactivation study of arginine to other enveloped viruses, such as Sendai virus and Newcastle Disease Virus (NDV), and observed that arginine was ineffective against both viruses under the similar conditions, i.e., on ice and above pH 4.0. However, an arginine derivative, butyroyl-arginine, showed a strong virucidal potency against Sendai virus, leading to a 4log reduction in virus yield at pH 4.0, but not against NDV. In addition, although arginine and butyroyl-arginine were equally effective against influenza virus having a cleaved form of hemagglutinin spike proteins, only butyroyl-arginine was significantly effective against the same virus, but having an uncleaved hemagglutinin spike proteins. Furthermore, butyroyl-arginine was more effective than arginine against HSV-1 at pH 4.5; i.e., it has a broader pH spectrum than does arginine.

Long-Term Stability and Reversible Thermal Unfolding of Antibody Structure at Low pH: Case Study.

We have here observed that the differential scanning calorimetry profiles and melting temperatures of a humanized antibody were unchanged over a 10-year span when stored at 4°C and at different pH values, even at pH 2.7. This is somewhat surprising, as this particular antibody undergoes conformational changes below pH 4.0. Differential scanning calorimetry analysis showed that melting of the antibody at pH 2.7 was highly reversible, suggesting a possibility that the observed reversibility is at least in part responsible for a 10-year stability at low pH. Conversely, it showed thermal unfolding followed by aggregation at higher pH.

Small molecule pharmacological chaperones: From thermodynamic stabilization to pharmaceutical drugs.

A great deal of attention has been paid to so-called amyloid diseases, in which the proteins responsible for the cell death and resultant diseases undergo conformational changes and aggregate in vivo, although whether aggregate formation is the cause or the result of the cell death is controversial. Recently, an increasing attention is given to protein folding diseases tightly associated with mutations. These mutations result in temperature-dependent misfolding and hence inactivation of the proteins, leading to loss of function, at physiological temperature; at low so-called permissive temperatures, the mutant proteins correctly fold and acquire functional structure. Alternatively, activation can be induced by use of osmolytes, which restores the folding of the mutant proteins and hence are called chemical chaperones. The osmolytes are compatible with macromolecular function and do stabilize the native protein structure. However, chemical chaperones require high concentrations for effective folding of mutant proteins and hence are too toxic in in-vivo applications. This limitation can be overcome by pharmacological chaperones, whose functions are similar to the chemical chaperones, but occur at much lower concentrations, i.e., physiologically acceptable concentrations. Although the research and clinical importance of pharmacological chaperones has been emphasized, the initial and central concept of osmolytes is largely ignored. Here we attempt to bridge the concept of osmolytes to applications of pharmacological chaperones.

Effects of acid exposure on the conformation, stability, and aggregation of monoclonal antibodies.

Exposure of antibodies to low pH is often unavoidable for purification and viral clearance. The conformation and stability of two humanized monoclonal antibodies (hIgG4-A and -B) directed against different antigens and a mouse monoclonal antibody (mIgG1) in 0.1M citrate at acidic pH were studied using circular dichroism (CD), differential scanning calorimetry (DSC), and sedimentation velocity. Near- and far-UV CD spectra showed that exposure of these antibodies to pH 2.7-3.9 induced only limited conformational changes, although the changes were greater at the lower pH. However, the acid conformation is far from unfolded or so-called molten globule structure. Incubation of hIgG4-A at pH 2.7 and 3.5 at 4 degrees C over the course of 24 h caused little change in the near-UV CD spectra, indicating that the acid conformation is stable. Sedimentation velocity showed that the hIgG4-A is largely monomeric at pH 2.7 and 3.5 as well as at pH 6.0. No time-dependent changes in sedimentation profile occurred upon incubation at these low pHs, consistent with the conformational stability observed by CD. The sedimentation coefficient of the monomer at pH 2.7 or 3.5 again suggested that no gross conformational changes occur at these pHs. DSC analysis of the antibodies showed thermal unfolding at pH 2.7-3.9 as well as at pH 6.0, but with decreased melting temperatures at the lower pH. These results are consistent with the view that the antibodies undergo limited conformational change, and that incubation at 4 degrees C at low pH results in no time-dependent conformational changes. Titration of hIgG4-A from pH 3.5 to 6.0 resulted in recovery of native monomeric proteins whose CD and DSC profiles resembled those of the original sample. However, titration from pH 2.7 resulted in lower recovery of monomeric antibody, indicating that the greater conformational changes observed at this pH cannot be fully reversed to the native structure by a simple pH titration.

Suppression of protein interactions by arginine: a proposed mechanism of the arginine effects.

Arginine has been used to suppress protein aggregation and protein-protein or protein-surface interactions during protein refolding and purification. While its biotechnology applications are gradually expanding, the mechanism of these effects of arginine has not been fully elucidated. Arginine is more effective at higher concentrations, an indication of weak interactions with the proteins. The effects of weakly interacting additives, such as arginine, on protein solubility, stability and aggregation have been explained from three different approaches: i.e., (1) the effects of additives on the structure of water, (2) the interactions of additives with the amino acid side chains and peptide bonds and (3) the preferential interactions of additives with the proteins. Here we have examined these properties of arginine and compared with those of other additives, e.g., guanidine hydrochloride (GdnHCl) and certain amino acids and amines. GdnHCl is a strong salting-in agent and denatures proteins, while betaine is a protein stabilizer. Several amino acids and amine compounds, including betaine, which stabilize the proteins, are strongly excluded; i.e., the proteins are preferentially hydrated in these solutions. On the other hand, GdnHCl preferentially binds to the proteins. Arginine is intermediate between these two extreme cases and shows a more complicated pattern of interactions with the proteins. The effects of additives on water structure, e.g., the surface tension of aqueous solution of the additives and the solubility of amino acids in the presence of additives also shed light on the mechanism of the effects of the additives on protein aggregation. While arginine increases the surface tension of water, it favorably interacts with most amino acid side chains and the peptide bonds, a property shared with GdnHCl. Thus, we propose that while arginine is similar to GdnHCl in the amino acid level, arginine interacts with the proteins differently from GdnHCl.