Determination and control of low-level amino acid misincorporation in human thioredoxin protein produced in a recombinant Escherichia coli production system

Escherichia coli is used extensively in the production of proteins within biotechnology for a number of therapeutic applications. Here, we discuss the production and overexpression of the potential biopharmaceutical human thioredoxin protein (rhTRX) within E. coli. Overexpression of foreign molecules within the cell can put an enormous amount of stress on the translation machinery. This can lead to a misfiring in the construction of a protein resulting in populations differing slightly in amino acid composition. Whilst this may still result in a population of active molecules being expressed, it does present significant problems with molecules that are destined for clinical applications. Amino acid misincorporation of this subset could potentially result in antibodies being raised to these unnatural proteins. Cross-reaction with a patient's endogenous thioredoxin could then lead to an autoimmune phenomena and serious health implications. Generally, the issue of misincorporation appears not to be a routine regulatory concern (see ICH Q6B guidelines). Therefore, amino acid misincorporation may not have been detected, much less explored in the clinic as the occurrence or absence of these random errors is not routinely reported. Using current technologies based on proteomics, the ability to find misincorporation critically depends upon the criteria for matching theoretical and experimental mass spectrometry data. Additionally, isolation and extraction of these mistranslated proteins from the production process is both difficult and expensive. Therefore, it is advantageous to find routes for removing their production during the upstream phase. In this study, we show how modern proteomic technology can be used to identify and quantify amino acid misincorporation. Using these techniques we have shown how manipulation of gene sequence and scoping of fermentation media composition can lead to the reduction and elimination of these misincorporations in rhTRX.     

Targeted codon optimization improves translational fidelity for an Fc fusion protein

High levels of translational errors, both truncation and misincorporation in an Fc‐fusion protein were observed. Here, we demonstrate the impact of several commercially available codon optimization services, and compare to a targeted strategy. Using the targeted strategy, only codons known to have translational errors are modified. For an Fc‐fusion protein expressed in Escherichia coli, the targeted strategy, in combination with appropriate fermentation conditions, virtually eliminated misincorporation (proteins produced with a wrong amino acid sequence), and reduced the level of truncation. The use of full optimization using commercially available strategies reduced the initial errors, but introduced different misincorporations. However, truncation was higher using the targeted strategy than for most of the full optimization strategies. This targeted approach, along with monitoring of translation fidelity and careful attention to fermentation conditions is key to minimizing translational error and ensuring high‐quality expression. These findings should be useful for other biopharmaceutical products, as well as any other transgenic constructs where protein quality is important. Biotechnol. Bioeng. 2012; 109: 2770–2777. © 2012 Wiley Periodicals, Inc.     

Assessing in vivo dynamics of multiple quality attributes from a therapeutic IgG4 monoclonal antibody circulating in cynomolgus monkey

Characterization of biopharmaceutical proteins and assessment and understanding of the critical quality attributes (CQAs) is a significant part of biopharmaceutical product development and is routinely performed in vitro. In contrast, systematic analysis of the quality attributes in vivo is not as widespread, although metabolism and clearance of multiple variants of therapeutic proteins administered to non-human primates and human subjects may have a different impact on safety, efficacy and exposure. The major hurdles of such studies are usually sample availability and technical capability. In this study, we used affinity purification coupled with liquid chromatography and mass spectrometric analysis of the digested protein for consistent and simultaneous detection of the full amino acid sequence of a therapeutic IgG4 monoclonal antibody, MAB1. This methodology allowed us to assess in vivo changes of all sequence-related modifications and quality attributes of MAB1 over the duration of a preclinical pharmacokinetic study in cynomolgus monkeys.     

Proteome analysis of the phenotypic variation process in Photorhabdus luminescens

Photorhabdus luminescens is an insect pathogen associated with specific soil nematodes. The bacterium has a complex life cycle with a symbiotic stage in which bacteria colonize the intestinal tract of the nematodes, and a pathogenic stage against susceptible larval-stage insect. Symbiosis-"deficient" phenotypic variants (known as secondary forms) arise during prolonged incubation. Correspondence analysis of the in silico proteome translated from the genome sequence of strain TT01 identified two major biases in the amino acid composition of the proteins. We analyzed the proteome, separating three classes of extracts: cellular, extracellular, and membrane-associated proteins, resolved by 2-DE. Approximately 450 spots matching the translation products of 231 different coding DNA sequences were identified by PMF. A comparative analysis was performed to characterize the protein content of both variants. Differences were evident during stationary growth phase. Very few proteins were found in variant II supernatants, and numerous proteins were lacking in the membrane-associated fraction. Proteins up-regulated by the phenotypic variation phenomenon were involved in oxidative stress, energy metabolism, and translation. The transport and binding of iron, sugars and amino acids were also affected and molecular chaperones were strongly down-regulated. A potential role for H-NS in phenotypic variation control is discussed.     

The primary structure of bovine cellular retinoic acid-binding protein

The complete amino acid sequence of bovine adrenal gland cellular retinoic acid-binding protein (CRABP) has been determined. The primary structure was established by analyses of cyanogen bromide fragments and peptides obtained by trypsin and Staphylococcus aureus protease digestions. The polypeptide chain of bovine CRABP comprises 136 amino acid residues. From partial sequence information, CRABP has been shown to be homologous to cellular retinol-binding protein, myelin protein P2, and the fatty acid-binding Z-protein. A comparison of the complete amino acid sequences of the members of this protein family, which also includes the rat intestinal fatty acid-binding protein, shows that CRABP is more similar to cellular retinol-binding protein and protein P2 than to the fatty acid-binding proteins. All five proteins are very similar in their NH2-terminal regions, suggesting that this part is important for a property common to the members of this protein family. This is the first report of a complete amino acid sequence of a CRABP.     

Rapid mapping of protein structure,interactions, and ligand binding by misincorporation proton-alkyl exchange

Understanding protein conformation, interactions, and ligand binding is essential to all biological inquiry. We report a novel biochemical technique, called misincorporation proton-alkyl exchange (MPAX), that can be used to footprint protein structure at single amino acid resolution. MPAX exploits translational misincorporation of cysteine residues to generate probes for physical analysis. We apply MPAX to the triosephosphate isomerase (beta/alpha)(8) barrel, accurately determining its substrate-binding site, a protein-protein interaction surface, the solvent-accessible protein surface, and the stability of the barrel. Because MPAX requires only microgram quantities of material and is not limited by protein size, it is ideally suited for proteins not amenable to conventional structural methods, such as membrane proteins, partially folded or insoluble proteins, and large protein complexes.     

Hepatitis B virus nucleocapsid assembly: primary structure requirements in the core protein.

As a step toward understanding the assembly of the hepatitis B virus (HBV) nucleocapsid at a molecular level, we sought to define the primary sequence requirements for assembly of the HBV core protein. This protein can self assemble upon expression in Escherichia coli. Applying this system to a series of C-terminally truncated core protein variants, we mapped the C-terminal limit for assembly to the region between amino acid residues 139 and 144. The size of this domain agrees well with the minimum length of RNA virus capsid proteins that fold into an eight-stranded beta-barrel structure. The entire Arg-rich C-terminal domain of the HBV core protein is not necessary for assembly. However, the nucleic acid content of particles formed by assembly-competent core protein variants correlates with the presence or absence of this region, as does particle stability. The nucleic acid found in the particles is RNA, between about 100 to some 3,000 nucleotides in length. In particles formed by the full-length protein, the core protein mRNA appears to be enriched over other, cellular RNAs. These data indicate that protein-protein interactions provided by the core protein domain from the N terminus to the region around amino acid 144 are the major factor in HBV capsid assembly, which proceeds without the need for substantial amounts of nucleic acid. The presence of the basic C terminus, however, greatly enhances encapsidation of nucleic acid and appears to make an important contribution to capsid stability via protein-nucleic acid interactions. The observation of low but detectable levels of nucleic acid in particles formed by core protein variants lacking the Arg-rich C terminus suggests the presence of a second nucleic acid-binding motif in the first 144 amino acids of the core protein. Based on these findings, the potential importance of the C-terminal core protein region during assembly in vivo into authentic, replication-competent nucleocapsids is discussed.     

Engineered protein function by selective amino acid diversification

Almost all protein engineering methods rely upon making changes to naturally occurring proteins that already possess some of the desired properties. This will probably remain the case as long as we lack a complete understanding of the way that an amino acid sequence gives rise to a protein with a precisely defined biological function. Common to all methods for altering an existing protein is the selection of a subset of amino acids in the protein for variation and a choice of which substitutions to make at each position. Variants are then tested empirically and further variants are created based upon their performance. Differences between protein engineering methods are the ways in which amino acids are chosen for variation, the protocols followed for creating the variants, and how information regarding variant properties is used in creating subsequent variants. In this article, we describe these differences and provide examples of how the experimental parameters of specific projects determine which method is most suitable.     

Cloning and sequencing of a full length cDNA corresponding to human cellular retinol-binding protein

We have isolated and sequenced a cDNA clone corresponding to the human cellular retinol-binding protein (CRBP). The deduced amino acid sequence, which encompasses 134 amino acid residues, shows significant homology with several low molecular weight proteins which bind hydrophobic ligands. No homology to the plasma retinol-binding protein was observed. Southern and Northern blot analyses suggest that the CRBP gene is present in a single copy in the haploid genome and that it is transcribed in a single mRNA species.     

A de novo designed protein protein interface

As an approach to both explore the physical/chemical parameters that drive molecular self-assembly and to generate novel protein oligomers, we have developed a procedure to generate protein dimers from monomeric proteins using computational protein docking and amino acid sequence design. A fast Fourier transform-based docking algorithm was used to generate a model for a dimeric version of the 56-amino-acid beta1 domain of streptococcal protein G. Computational amino acid sequence design of 24 residues at the dimer interface resulted in a heterodimer comprised of 12-fold and eightfold variants of the wild-type protein. The designed proteins were expressed, purified, and characterized using analytical ultracentrifugation and heteronuclear NMR techniques. Although the measured dissociation constant was modest ( approximately 300 microM), 2D-[(1)H,(15)N]-HSQC NMR spectra of one of the designed proteins in the absence and presence of its binding partner showed clear evidence of specific dimer formation.     

Many amino acid substitution variants identified in DNA repair genes during human population screenings are predicted to impact protein function

Over 520 different amino acid substitution variants have been previously identified in the systematic screening of 91 human DNA repair genes for sequence variation. Two algorithms were employed to predict the impact of these amino acid substitutions on protein activity. Sorting Intolerant from Tolerant (SIFT) classified 226 of 508 variants (44%) as "Intolerant." Polymorphism Phenotyping (PolyPhen) classed 165 of 489 amino acid substitutions (34%) as "Probably or possibly damaging." Another 9-15% of the variants were classed as "Potentially intolerant or damaging." The results from the two algorithms are highly associated, with concordance in predicted impact observed for approximately 62% of the variants. Twenty-one to thirty-one percent of the variant proteins are predicted to exhibit reduced activity by both algorithms. These variants occur at slightly lower individual allele frequency than do the variants classified as "Tolerant" or "Benign." Both algorithms correctly predicted the impact of 26 functionally characterized amino acid substitutions in the APE1 protein on biochemical activity, with one exception. It is concluded that a substantial fraction of the missense variants observed in the general human population are functionally relevant. These variants are expected to be the molecular genetic and biochemical basis for the associations of reduced DNA repair capacity phenotypes with elevated cancer risk.     

Brassica napus plastid and mitochondrial chaperonin-60 proteins contain multiple distinct polypeptides.

Plastid chaperonin-60 protein was purified to apparent homogeneity from Brassica napus using a novel protocol. The purified protein, which migrated as a single species by nondenaturing polyacrylamide gel electrophoresis, contained four polypeptides: three variants of p60cpn60 alpha and p60cpn60 beta. Partial amino acid sequence determination demonstrated that each variant of p60cpn60 alpha is a distinct translation product. During this study, additional chaperonin-60 proteins were purified. These proteins, which were free from contaminating plastid chaperonin-60, were separated into at least two high molecular weight species that were resolved only by nondenaturing polyacrylamide gel electrophoresis. These proteins contained three 60-kD polypeptides. Two of these polypeptides were recognized by existing antisera, whereas the third was not. Partial amino acid sequence data revealed that each of these, including the immunologically distinct polypeptide, is a chaperonin-60 subunit of putative mitochondrial origin. The behavior of chaperonin-60 proteins during blue A Dyematrex chromatography suggests that this matrix may be generally useful for the identification of chaperonin-60 proteins.     

Identification of a matrix-binding domain in MAGP1 and MAGP2 and intracellular localization of alternative splice forms.

MAGP1 is a small molecular mass protein associated with microfibrils in the extracellular matrix (ECM). To identify the molecular basis of its interaction with other microfibrillar proteins, deletion constructs of MAGP1 were expressed in a mammalian cell system that served as a model for microfibril assembly. This study identified a 54-amino acid sequence in the carboxyl-terminal region of the protein that defines a matrix-binding domain that is sufficient to target MAGP1 to the ECM. Site-directed mutagenesis demonstrated that binding activity is dependent on the presence of 7 cysteine residues in this sequence. MAGP2 contains a sequence similar to the matrix-binding domain of MAGP1, but could not associate with the ECM because of a single amino acid change. Two naturally occurring MAGP1 splice variants, MAGP1B (human-specific) and MAGP1D (found in mice), localized intracellularly when expressed as chimeric proteins with green fluorescent protein in rat lung fibroblasts. This suggests a second action site for MAGP1.     

Targeted purification development enabled by computational biophysical modeling

Chromatographic and non‐chromatographic purification of biopharmaceuticals depend on the interactions between protein molecules and a solid–liquid interface. These interactions are dominated by the protein–surface properties, which are a function of protein sequence, structure, and dynamics. In addition, protein–surface properties are critical for in vivo recognition and activation, thus, purification strategies should strive to preserve structural integrity and retain desired pharmacological efficacy. Other factors such as surface diffusion, pore diffusion, and film mass transfer can impact chromatographic separation and resin design. The key factors that impact non‐chromatographic separations (e.g., solubility, ligand affinity, charges and hydrophobic clusters, and molecular dynamics) are readily amenable to computational modeling and can enhance the understanding of protein chromatographic. Previously published studies have used computational methods such as quantitative structure–activity relationship (QSAR) or quantitative structure–property relationship (QSPR) to identify and rank order affinity ligands based on their potential to effectively bind and separate a desired biopharmaceutical from host cell protein (HCP) and other impurities. The challenge in the application of such an approach is to discern key yet subtle differences in ligands and proteins that influence biologics purification. Using a relatively small molecular weight protein (insulin), this research overcame limitations of previous modeling efforts by utilizing atomic level detail for the modeling of protein–ligand interactions, effectively leveraging and extending previous research on drug target discovery. These principles were applied to the purification of different commercially available insulin variants. The ability of these computational models to correlate directionally with empirical observation is demonstrated for several insulin systems over a range of purification challenges including resolution of subtle product variants (amino acid misincorporations). Broader application of this methodology in bioprocess development may enhance and speed the development of a robust purification platform. © 2014 American Institute of Chemical Engineers Biotechnol. Prog., 31:154–164, 2015     

Effects of randomizing the Sup35NM prion domain sequence on formation of amyloid fibrils in vitro

The mechanism by which proteins aggregate and form amyloid fibrils is still elusive. In order to preclude interference by cellular factors and to clarify the role of the primary sequence of Sup35p prion domain in formation of amyloid fibrils, we generated five Sup35NM variants by randomizing amino acid sequences in PrDs without altering the amino acid composition and analyzed the in vitro process of amyloid fibril formation. The results showed that each of the five Sup35NM variants polymerized into amyloid fibrils in vitro under native conditions. Furthermore, the Sup35NM variants showed differences in their aggregation time courses. These findings indicate that specific amino acid sequence features in PrD can modify the rate of conversion of Sup35p into amyloid fibrils in vitro.     

Analysis of posttranslational modifications of proteins by tandem mass spectrometry

Protein activity and turnover is tightly and dynamically regulated in living cells. Whereas the three-dimensional protein structure is predominantly determined by the amino acid sequence, posttranslational modification (PTM) of proteins modulates their molecular function and the spatial-temporal distribution in cells and tissues. Most PTMs can be detected by protein and peptide analysis by mass spectrometry (MS), either as a mass increment or a mass deficit relative to the nascent unmodified protein. Tandem mass spectrometry (MS/MS) provides a series of analytical features that are highly useful for the characterization of modified proteins via amino acid sequencing and specific detection of posttranslationally modified amino acid residues. Large-scale, quantitative analysis of proteins by MS/MS is beginning to reveal novel patterns and functions of PTMs in cellular signaling networks and biomolecular structures.     

The amino acid sequence of the sex steroid-binding protein of rabbit serum

The amino acid sequence of the sex steroid-binding protein (SBP or SHBG) of rabbit serum, specific for binding testosterone and 5 alpha-dihydrotestosterone, was determined using a complementary combination of mass spectrometric and Edman degradation techniques. The monomeric unit of the homodimeric protein is a single chain glycopeptide of 367 amino acid residues, with N-linked oligosaccharide side chains at Asn-345 and Asn-361 and disulfide bonds connecting Cys-158 to Cys-182 and Cys-327 to Cys-355. The polypeptide molecular weight of the monomer calculated from the sequence is 39,769. The molecular weight of the homodimer including 9% carbohydrate is 87,404. The sequence contains a relatively hydrophobic segment between Trp-241 and Leu-282, which includes many leucine residues in an alternating pattern. An amino acid sequence repeat is also located within that segment. Both of these patterns are present in human SBP and in the androgen-binding protein of rat epididymis. The sequence data indicate that the previously reported microheterogeneity of rabbit SBP in sodium dodecyl sulfate-polyacrylamide gel electrophoresis reflects variants generated by differential glycosylation of the monomer rather than different gene products. Seventy-nine percent of the amino acids of rabbit SBP are identical to those of human SBP; rabbit SBP thus joins human SBP and rat androgen-binding protein in one gene family that is distinct from the steroid hormone receptor superfamily. It appears that the problem of binding sex steroid hormones has been solved independently in two different gene families that contain completely different steroid-binding domains. Since the nonhomologous steroid-binding domains of both families of proteins recognize essentially the same steroid structure, it will be interesting to determine the structural basis of the two different protein designs that lead to similar steroid-binding specificity.     

Sequence of the D-aspartyl/L-isoaspartyl protein methyltransferase from human erythrocytes. Common sequence motifs for protein, DNA, RNA, and small molecule S-adenosylmethionine-dependent methyltransferases

A widely distributed protein methyltransferase catalyzes the transfer of a methyl group from S-adenosyl-methionine to the free carboxyl groups of D-aspartyl and/or L-isoaspartyl derivatives of L-aspartyl and L-asparaginyl residues. This enzyme has been postulated to function in the repair or the catabolism of age-damaged proteins. We present here the complete amino acid sequence of the more basic isozyme I of this enzyme from human erythrocytes. The sequence was determined by Edman degradation and mass spectral analysis of overlapping trypsin, Staphylococcus aureus V8 protease, Pseudomonas fragi endoproteinase Asp-N, cyanogen bromide, and hydroxylamine-generated fragments. The NH2-terminus is modified by acetylation and the protein contains 226 amino acids for a calculated molecular weight of 24,575. This value is in good agreement with the molecular weight determined for the purified protein by polyacrylamide gel electrophoresis in the presence of dodecyl sulfate and by gel filtration chromatography under nondenaturing conditions. The identification of 2 different amino acid residues at both positions 22 and 119 may indicate the presence of allelic variants or of two or more closely related structural genes. Finally, comparison of this sequence with those of methyltransferases for RNA, DNA, and small molecules, as well as other S-adenosylmethionine-utilizing enzymes, shows that many of these proteins share elements of three regions of sequence similarity and may be structurally or evolutionarily related.     

Nucleotide sequence of the phoS gene, the structural gene for the phosphate-binding protein of Escherichia coli.

phoS is the structural gene for the phosphate-binding protein, which is localized in periplasm and involved in active transport of phosphate in Escherichia coli. It is also a negative regulatory gene for the pho regulon, and the gene expression is inducible by phosphate starvation. The complete nucleotide sequence of the phoS gene was determined by the method of Maxam and Gilbert (A. M. Maxam and W. Gilbert, Methods Enzymol. 65:499-560, 1980). The amino acid sequences at the amino termini of the pre-PhoS and PhoS proteins and at the carboxy terminus of the PhoS protein were determined by using the purified proteins. Furthermore, the amino acid sequence of enzymatically digested peptide fragments of the PhoS protein was determined. The combined data established the nucleotide sequence of the coding region and the amino acid sequence of the pre-PhoS and the PhoS proteins. The pre-PhoS protein contains an extension of peptide composed of 25 amino acid residues at the amino terminus of the PhoS protein, which has the general characteristics of a signal peptide. The mature PhoS protein is composed of 321 amino acid residues, with a calculated molecular weight of 34,422, and lacks the disulfide bond and methionine. The regulatory region of phoS contains a characteristic Shine-Dalgarno sequence at an appropriate position preceding the translational initiation site, as well as three possible Pribnow boxes and one -35 sequence. the nucleotide sequence of the regulatory region of phoS was compared with those of phoA and phoE, the genes constituting the pho regulon.