Production of single chain fragment variable (scFv) antibodies in Escherichia coli using the LEX™ bioreactor

doi:10.1016/j.jbiotec.2012.07.011

Journal of Biotechnology

Volume 163, Issue 2, 20 January 2013, Pages 105-111

https://doi.org/10.1016/j.jbiotec.2012.07.011 Get rights and content

Abstract

For proteome research, antibodies against a growing number of antigens must be generated and characterized. The high throughput generation of antibody fragments, using in vitro selection, requires bacterial expression of antibody fragments. This created a need to establish an expression method to improve the parallel production of many antibody fragments. In this study, we describe the development of a high throughput bacterial production method for single chain fragment variables (scFvs) using shaking flasks or the LEX™ bioreactor. We compared the influence of a set of production parameters on Escherichia coli production of four different scFv. The results led to an optimized protocol for the parallel production of multiple antibody fragments.

Highlights

► First production of antibody fragments (scFv) in the LEX bioreactor. ► High scale production of different scFv in Escherichia coli. ► Simple, robust and parallelized scFv production. ► Higher production yields compared to shake flasks.

Introduction

Recombinant antibody fragments are essential tools for research, diagnostics and therapy (Schütte et al., 2009, Dübel et al., 2010, Thie et al., 2011). Therefore, specific antibody formats such as single chain fragment variables (scFvs) or single chain fragment antigen bindings (scFabs) can be isolated using in vitro selection systems like phage display (McCafferty et al., 1990, Breitling et al., 1991, Hust et al., 2007a, Hust et al., 2007b, Schirrmann and Hust, 2010) and can be produced with bacterial expression systems (Hust et al., 2007a). In addition, those selected antibody formats can be converted into full size IgG or scFv-Fc fusion proteins (Boel et al., 2000, Schirrmann and Büssow, 2010). An increasing number of recombinant antibodies is generated and evaluated (Colwill and Gräslund, 2011). The phage display technique provides a platform for high-throughput generation of antibody fragments (Hallborn and Carlsson, 2002, Konthur et al., 2005, Buckler et al., 2008). Especially for proteome research, antibodies against a multitude of antigens must be generated and characterized. This fact is important, since the human genome contains ∼20,000 genes, which can encode at least 100,000 different protein products as potential targets for research (Taussig et al., 2007). Therefore, a production pipeline capable of dealing with the increasing needs for parallelized production in Escherichia coli is necessary. However, the expression of antibody fragments needs to be carefully optimized in respect of expression temperature, aeration, cultivation media, media supplements or the E. coli strain (Schirrmann et al., 2008, Hust et al., 2009).

In this study, we compare the LEX™ bubble column bioreactor to shaking flask production for the parallel generation of multiple scFv antibody fragments in E. coli.

Section snippets

Bacterial strains and genotypes

The following E. coli bacterial strains were used: XL1 Blue MRF′, JM109 and BL21 (Stratagene, Amsterdam, Table 1). Standard procedures were carried out according to Sambrook and Russell (2001).

Antibody fragments

The following single chain fragment variable (scFv) antibodies were used for comparison of different parameters for large scale production: D1.3 (murine anti-hen egg-white lysozyme) (Kirsch et al., 2005), HT186-D11 (human anti-human MUC1) (Thie et al., 2011), SH313-B5 (human anti-CD30) (Wezler et al., 2012

Production of scFv in shaking flask

As starting point, the scFv production using the E. coli expression vector pOPE101-XP (Hust et al., 2007b) was optimized. Several parameters were tested in small scale (100 mL) shaking flasks. These parameters included the expression temperature (18 °C, 26 °C or 30 °C), media (2× YT, phosphate-buffered 2× YT, LB, phosphate-buffered LB or TB), media supplements (sucrose, sorbitol or betaine), IPTG concentration (ranging from 25 μM up to 1 mM), production time and different bacterial strains (XL1-Blue

Discussion

ScFv antibody fragments are a proven alternative to full size IgG for applications were there is no need for the effector function of the immunoglobulins, such as toxin neutralization (Chahboun et al., 2011) or antibody–drug conjugates (Kellner et al., 2011). Moreover, scFv fragments are now validated as valuable tools for basic research (Colwill and Gräslund, 2011). In the latter application, the production of antibody fragments in high throughput using E. coli cultures would allow a rapid and

Acknowledgments

We gratefully acknowledge the financial support by SFB 578 of the Deutsche Forschungsgemeinschaft “Vom Gen zum Produkt” (Teilprojekt A6), the EU FP7 collaborative projects “Affinity Proteome” (contract 222635) and “Affinomics” (contract 241481), and mAb-factory GmbH. We thank Martin Becker for careful corrections on the manuscript.

References (37)

E. Boel et al.
Functional human monoclonal antibodies of all isotypes constructed from phage display library-derived single-chain Fv antibody fragments
Journal of Immunological Methods
(2000)
F. Breitling et al.
A surface expression vector for antibody screening
Gene
(1991)
D.R. Buckler et al.
Screening isolates from antibody phage-display libraries
Drug Discovery Today
(2008)
S. Dübel et al.
Generating recombinant antibodies to the complete human proteome
Trends in Biotechnology
(2010)
S. Goletz et al.
Selection of large diversities of antiidiotypic antibody fragments by phage display
Journal of Molecular Biology
(2002)
S. Gräslund et al.
The use of systematic N- and C-terminal deletions to promote production and structural studies of recombinant proteins
Protein Expression and Purification
(2008)
M. Hust et al.
A human scFv antibody generation pipeline for proteome research
Journal of Biotechnology
(2011)
M. Hust et al.
Improved microtitre plate production of single chain Fv fragments in Escherichia coli
Nature Biotechnology
(2009)
C. Kellner et al.
Human kappa light chain targeted Pseudomonas exotoxin A—identifying human antibodies and Fab fragments with favorable characteristics for antibody-drug conjugate development
Journal of Immunological Methods
(2011)
S.M. Kipriyanov et al.
High level production of soluble single chain antibodies in small-scale Escherichia coli cultures
Journal of Immunological Methods
(1997)

M. Kirsch et al.

Parameters affecting the display of antibodies on phage

Journal of Immunological Methods

(2005)

Z. Konthur et al.

Perspectives for systematic in vitro antibody generation

Gene

(2005)

M. Vedadi et al.

Genome-scale protein expression and structural biology of Plasmodium falciparum and related Apicomplexan organisms

Molecular and Biochemical Parasitology

(2007)

J.C. Bardwell et al.

A pathway for disulfide bond formation in vivo

Proceedings of the National Academy of Sciences of the United States of America

(1993)

S. Chahboun et al.

Isolation of a nanomolar scFv inhibiting the endopeptidase activity of botulinum toxin A, by single-round panning of an immune phage-displayed library of macaque origin

BMC Biotechnology

(2011)

K. Colwill et al.

A roadmap to generate renewable protein binders to the human proteome

Nature Methods

(2011)

S. Dübel et al.

Regulated secretion and purification of recombinant antibodies in E. coli

Cell Biophysics

(1992)

S. Gottesman et al.

Protein quality control: triage by chaperones and proteases

Genes and Development

(1997)

Cited by (22)

Optimized extraction of a single-chain variable fragment of antibody by using aqueous micellar two-phase systems
2015, Protein Expression and Purification
Citation Excerpt :
Due to the advantageous features above mentioned, antibody fragment pipeline is now expanding [4,5], thus, its industrial production and purification is highly demanded [6]. Up to date, the production of recombinant antibodies fragments has reached high titers [7–9], which imply that the downstream unit operations delimit the cost of the whole process [7,10–12]. For example, it is known that 30% of the total purification cost corresponds to the Protein A resin, with an estimated value of 4–5 millions dollars in a scale of 10.000 L [11,13].
In this work, the purification of a single-chain variable fragment (scFv) of an antibody by using liquid–liquid extraction in aqueous micellar two-phase systems was optimized by means of central composite design. Protein partitioning assays were performed by using the selected system composition in previous works: Triton X-114 at 4% wt/wt, yeast fermentation supernatant at 60% wt/wt, McIlvaine buffer pH 7.00. The other system component concentrations, Cibacron Blue F3GA (CB), Fabsorbent™ F1P HF (HF) and NaCl, were selected as independent variables. ScFv recovery percentage (%R) and purification factor (PF) were selected as the responses. According to the optimization process both, scFv recovery percentage and purification factor were favored with the addition of HF and NaCl in a range of concentrations around the central point of the second central composite design (HF 0.0120% w/w, CB 0.0200% w/w, NaCl 0.200% w/w). These experimental conditions allowed the concentration and pre-purification of scFv in the micelle-rich bottom phase of the systems with a recovery percentage superior to 88% and a purification factor of approximately 3.5. These results improved the previously presented works and demonstrated the convenience of using aqueous micellar two-phase systems as a first step in the purification of scFv molecules.
Site-specific immobilization of recombinant antibody fragments through material-binding peptides for the sensitive detection of antigens in enzyme immunoassays
2014, Biochimica et Biophysica Acta - Proteins and Proteomics
Citation Excerpt :
The development of phage display systems as well as technologies for recombinant protein production have made it possible to isolate valuable recombinant antibody fragments such as scFv, Fab and VHH [20–29]. Recombinant antibody fragments are expected to be a capture ligand alternative to conventional whole Abs in immunodiagnostics and affinity separation, due to the following reasons: 1) the recombinant antibody fragments can be economically produced by recombinant microorganisms such as Escherichia coli and yeast due to a structure that is simpler than that of conventional whole Abs [30–39]; 2) smaller molecular sizes of recombinant antibody fragments result in an increase in immobilization density, so the binding capacity of antibody-immobilized solid supports will also be increased [40–44]; and, 3) the genetic fusion of various affinity peptide sequences to their N or C-termini will result in the achievement of single-step purification as well as site-specific immobilization onto the solid supports via the introduction of affinity peptide tags [40–44]. Yoshimoto et al. reported the monitoring of adsorption-induced inactivation of antibody fragments, Fab′ immobilized on a gold surface which is used as a sensor chip surface of surface plasmon resonance (SPR) sensor.
The immobilization of an antibody is one of the key technologies that are used to enhance the sensitivity and efficiency of the detection of target molecules in immunodiagnosis and immunoseparation. Recombinant antibody fragments such as VHH, scFv and Fabs produced by microorganisms are the next generation of ligand antibodies as an alternative to conventional whole Abs due to a smaller size and the possibility of site-directed immobilization with uniform orientation and higher antigen-binding activity in the adsorptive state. For the achievement of site-directed immobilization, affinity peptides for a certain ligand molecule or solid support must be introduced to the recombinant antibody fragments. In this mini-review, immobilization technologies for the whole antibodies (whole Abs) and recombinant antibody fragments onto the surfaces of plastics are introduced. In particular, the focus here is on immobilization technologies of recombinant antibody fragments utilizing affinity peptide tags, which possesses strong binding affinity towards the ligand molecules. Furthermore, I introduced the material-binding peptides that are capable of direct recognition of the target materials. Preparation and immobilization strategies for recombinant antibody fragments linked to material-binding peptides (polystyrene-binding peptides (PS-tags) and poly (methyl methacrylate)-binding peptide (PMMA-tag)) are the focus here, and are based on the enhancement of sensitivity and a reduction in the production costs of ligand antibodies. This article is part of a Special Issue entitled: Recent advances in molecular engineering of antibody.
Aqueous micellar systems containing Triton X-114 and Pichia pastoris fermentation supernatant: A novel alternative for single chain-antibody fragment purification
2014, Separation and Purification Technology
Citation Excerpt :
In fact, thousands of examples of scFv and derivatives are currently in development or are undergoing clinical testing [2,3]. At present, most of scFv molecules are produced by microorganisms, such as Pichia pastoris, and purified directly from the fermented broth [4–6]. However, the efficient purification of the target molecule requires several unit operations (e.g. centrifugation, filtration, precipitation and different chromatographic techniques) due to the complexity of culture medium (high viscosity, high density, mixture of biomolecules) [7].
The main goal of this work was to evaluate the feasibility of using aqueous micellar two-phase systems (AMTPS) of Triton X-114 to purify a single-chain antibody fragment (scFv) directly from yeast fermentation supernatant. The coexistence curves of aqueous micellar two-phase systems, highly loaded (up to 90% wt/wt) with the biological feedstock, were determined. Besides, the effect of several additives such as inorganic salts and affinity ligands on the phase separation behavior was investigated. The obtained coexistence curves demonstrated that the assayed surfactant/yeast broth solutions were able to separate into two phases at temperatures lower than 24 °C. This information was then utilized to assay scFv partitioning and purification. In general, proteins present in the yeast broth, including the scFv, partitioned to the top, micelle-depleted phase due to its hydrophilic character. When affinity ligands were used, an opposite behavior was observed for scFv due to the uneven partitioning of ligands toward the micellar-rich phase. The best purification performances were attained for the system consisting of 4% wt/wt of Triton X-114, 60% wt/wt of yeast fermentation supernatant and the synthetic ligand Fabsorbent™ F1P HF, with a recovery percentage of 88% and a purification factor of 2. These results demonstrate the potential applicability of these systems for designing early steps for scFv purification directly from yeast broth. New perspectives are now open for the use of this methodology for recombinant antibody fragments purification.
Single-chain antibody fragments: Purification methodologies
2013, Process Biochemistry
Citation Excerpt :
The affinity stretch or tags can be short polypeptide sequences or whole proteins co-expressed as fusion partners with the target proteins [40]. At present, immobilized metal affinity chromatography (IMAC), which employs chelating metal ions as ligand, is one of the most extended methodology used to purify recombinant antibodies [41,42]. Several vector systems, which express proteins containing a stretch of 5–6 histidines in the final product, are commercially available [40].
At present, single-chain variable fragments (scFv) of antibodies are considered one of the most important tools in human therapies. Wide applications of antibodies are being exploited in different medical, pharmaceutical and research areas. These molecules maintain the same binding functionality that full length antibodies but possess several advantageous features as quickness to penetrate the tissues, easy manipulation, fast elimination of their immunocomplex and the possibility of being produced in simple expression systems like bacteria and yeast. The increasing demand in antibody based methodologies is driving advances in the production and purification of genetically engineered antibodies and antibody fragments. While advances in expression systems allow the production of high titers of antibodies, there exist some limits imposed by the downstream methodologies which are not efficient enough to ensure their industrialization.
The main aim of this review is to highlight the principal characteristics of single-chain variable fragments of antibodies addressing advances and perspectives on scFv purification.
Construction of an scFv library by enzymatic assembly of V<inf>L</inf> and V<inf>H</inf> genes
2013, Journal of Immunological Methods
Citation Excerpt :
The scFv is suitable for generating new recombinant antibodies from a library since it is easily displayed on phage (McCafferty et al., 1990; Marks et al., 1991). ScFvs can be readily expressed in Escherichia coli, which grow rapidly in simple and inexpensive media for the generation of considerable amounts of protein (Hust et al., 2009; Jeong et al., 2011; Miethe et al., 2013). To construct scFvs, each V-domain coding region of an antibody is amplified by PCR, usually from B cell cDNA, and the VL and VH domains are joined together by a flexible linker.
The single-chain Fv fragment (scFv) is the most frequently used form of recombinant antibody. It is possible to establish clones specific to a certain target by displaying the scFv library on phages followed by biopanning against the target. For the construction of superior scFv libraries, the light-chain variable region (V_L) and the heavy-chain variable region (V_H) fragments should be assembled into the scFv without loss of diversity. We have provided an efficient method for constructing scFvs by enzymatic assembly of the V_L and V_H domains using the concerted action of λ-exonuclease and Bst DNA polymerase. First, we amplified the chicken V_L and V_H fragments using a phosphorylated primer with a 21-nucleotide overlap in the linker region. Then we recessed the overlapping parts of the V_L and V_H fragments with λ-exonuclease, which yielded single-stranded overhangs that specifically annealed between the V_L and V_H fragments; the complete double-stranded scFv was formed using Bst DNA polymerase. Complete scFvs were obtained using this method, whereby a library of scFvs was constructed from the immune library of chicken IgG. The diversity of this scFv library was analyzed by DNA fingerprinting method. The scFvs specific to the antigen could be isolated from this library after 5 rounds of panning.
Encapsulation of proteins in hydrogel carrier systems for controlled drug delivery: Influence of network structure and drug size on release rate
2013, Journal of Biotechnology
Citation Excerpt :
Converted human serum and antibiotic/antimycotic solution were obtained from PAA (Cölbe, Germany). The single chain fragment variable (scFv) anti-MUC1 (HT186D11; Thie et al., 2011) was produced in a large-scale expression system (LEX, Harbinger Biotech, Toronto, Canada) in Escherichia coli (as described in Miethe et al., 2013) and labeled with a FITC antibody labeling kit (Thermo Scientific, Rockford, USA) leading to a FITC-scFv solution (concentration 1.47 mg/ml). The synthesis of hydroxyethyl starch polyethylene glycol methacrylate is a two step procedure with an initial formation of polyethylene glycol methacrylate imidazole carbamate (P(EG)6MACI) according to van Dijk-Wolthuis et al. (1995).
Novel hydrogels based on hydroxyethyl starch modified with polyethylene glycol methacrylate (HES-P(EG)₆MA) were developed as delivery system for the controlled release of proteins. Since the drug release behavior is supposed to be related to the pore structure of the hydrogel network the pore sizes were determined by cryo-SEM, which is a mild technique for imaging on a nanometer scale. The results showed a decreasing pore size and an increase in pore homogeneity with increasing polymer concentration. Furthermore, the mesh sizes of the hydrogels were calculated based on swelling data. Pore and mesh size were significantly different which indicates that both structures are present in the hydrogel. The resulting structural model was correlated with release data for bulk hydrogel cylinders loaded with FITC-dextran and hydrogel microspheres loaded with FITC-IgG and FITC-dextran of different molecular size. The initial release depended much on the relation between hydrodynamic diameter and pore size while the long term release of the incorporated substances was predominantly controlled by degradation of the network of the much smaller meshes.

View all citing articles on Scopus

View full text

Production of single chain fragment variable (scFv) antibodies in Escherichia coli using the LEX™ bioreactor

Abstract

Highlights

Introduction

Section snippets

Bacterial strains and genotypes

Antibody fragments

Production of scFv in shaking flask

Discussion

Acknowledgments

Journal of Immunological Methods

Gene

Drug Discovery Today

Trends in Biotechnology

Journal of Molecular Biology

Protein Expression and Purification

Journal of Biotechnology

Nature Biotechnology

Journal of Immunological Methods

Journal of Immunological Methods

Journal of Immunological Methods

Gene

Molecular and Biochemical Parasitology

A pathway for disulfide bond formation in vivo

Proceedings of the National Academy of Sciences of the United States of America

Isolation of a nanomolar scFv inhibiting the endopeptidase activity of botulinum toxin A, by single-round panning of an immune phage-displayed library of macaque origin

BMC Biotechnology

A roadmap to generate renewable protein binders to the human proteome

Nature Methods

Regulated secretion and purification of recombinant antibodies in E. coli

Cell Biophysics

Protein quality control: triage by chaperones and proteases

Genes and Development