A novel expression system of domain I of human beta2 glycoprotein I in Escherichia coli
© Ioannou et al; licensee BioMed Central Ltd. 2006
Received: 22 July 2005
Accepted: 10 February 2006
Published: 10 February 2006
The antiphospholipid syndrome (APS), characterised by recurrent miscarriage and thrombosis, is a significant cause of morbidity and mortality. Domain I (DI) of human beta 2 glycoprotein I (β2GPI) is thought to contain crucial antibody binding epitopes for antiphospholipid antibodies (aPL), which are critical to the pathogenesis of APS. Expressing this protein in bacteria could facilitate studies investigating how this molecule interacts with aPL.
Using a computer programme called Juniper, sequentially overlapping primers were designed to be used in a recursive polymerase chain reaction (PCR) to produce a synthetic DI gene. Specifically Juniper incorporates 'major' codons preferred by bacteria altering 41 codons out of 61. This was cloned into the expression plasmid pET(26b) and expressed in BL21(DE3) Escherichia coli (E. coli). By virtue of a pelB leader sequence, periplasmic localisation of DI aided disulphide bond formation and toxicity was addressed by tightly regulating expression through the high stringency T7lac promoter.
Purified, soluble his-tagged DI in yields of 750 μg/L bacterial culture was obtained and confirmed on Western blot. Expression using the native human cDNA sequence of DI in the same construct under identical conditions yielded significantly less DI compared to the recombinant optimised sequence. This constitutes the first description of prokaryotic expression of soluble DI of β2GPI. Binding to murine monoclonal antibodies that recognise conformationally restricted epitopes on the surface of DI and pathogenic human monoclonal IgG aPL was confirmed by direct and indirect immunoassay. Recombinant DI also bound a series of 21 polyclonal IgG samples derived from patients with APS.
By producing a synthetic gene globally optimised for expression in E. coli, tightly regulating expression and utilising periplasmic product translocation, efficient, soluble E. coli expression of the eukaryotic protein DI of β2GPI is possible. This novel platform of expression utilising pan-gene prokaryote codon optimisation for DI production will aid future antigenic studies. Furthermore if DI or peptide derivatives of DI are eventually used in the therapeutic setting either as toleragen or as a competitive inhibitor of pathogenic aPL, then an E. coli production system may aid cost-effective production.
The APS is a multi-system autoimmune disease characterised by vascular thrombosis and/or recurrent pregnancy loss in patients who test positive for either aPL or lupus anticoagulant . APS carries a significant burden of morbidity and mortality  with long-term anticoagulation being the only treatment with any proven benefit in reducing recurrent thrombosis . Since anticoagulation carries an inherent risk of bleeding it is desirable to develop alternative treatments that target aPL directly. Patients with APS generally have high levels of serum IgG aPL. Monoclonal and polyclonal IgG aPL have been shown to be pathogenic and promote thrombosis in vivo [4, 5]. aPL from patients with thrombosis only bind phospholipids (PL) in the presence of protein co-factors, of which β2GPI has been studied the most extensively. Understanding how these pathogenic aPL interact with β2GPI at the molecular level could ultimately facilitate the development of targeted therapies.
β2GPI contains five homologous domains  and anchors to PL via domain V . Using domain deletion studies, in which β2GPI was produced with one or more domains deleted, it was shown that the amino terminal domain (DI) is particularly important for aPL binding, suggesting that crucial aPL binding epitopes are contained within this domain [8–10]. A compound based on DI is currently being studied for possible use as a toleragen to treat APS patients by inducing anergy in B lymphocytes that produce aPL . Clearly an efficient method of DI production could enhance the scope for investigating the antigenic role of DI and facilitate epitope mapping studies. An expression system in E. coli could offer such a tool. Furthermore if DI is ultimately used therapeutically, prokaryotic expression lends itself to large-scale cost-efficient production.
E. coli is the most frequently used prokaryotic expression system for production of heterologous proteins due to its efficiency, cost-effectiveness and potential for high-level production [12, 13]. However, various properties of different genes, their transcribed mRNAs and protein products may preclude efficient eukaryotic protein expression in bacteria. Current expression methods for DI use baculovirus containing a cloned cDNA sequence of the DI gene to infect Spodoptera frigiperda insect cells . This method of expression is relatively expensive, laborious and less amenable to being scaled-up in comparison to prokaryotic expression systems. Other than the formation of two disulphide bonds, no other post-translational modifications such as glycosylation are necessary for biologically active DI expression. Hence an established E. coli expression system could hold several significant advantages over the currently available system of production.
There are no published reports of DI expression in E. coli. Previous attempts to achieve this may have been unsuccessful for a number of reasons. We hypothesise that these reasons are related to E. coli codon bias against eukaryotic codons, potential toxicity of DI to E. coli and the relative difficulty of disulphide bond formation in the reducing environment of bacterial cytoplasm. In this paper we have addressed these problems and now report the first system for expressing DI in E. coli. We have demonstrated the binding properties of the expressed protein to murine monoclonal anti-antibodies that recognise conformational epitopes on the surface of DI, to pathogenic human monoclonal IgG aPL and to polyclonal IgG derived from APS patients.
Results and Discussion
Pan gene codon optimisation and tight regulation of expression are critical in ensuring efficient production of DI by E. coli
Recombinant periplasmic his-tagged DI is conformationally correct and may be purified using nickel chromatography
Human IgG aPL binds to expressed DI
Clearly recombinant DI expressed by E. coli binds IgG from patients with APS. The fact that DI binds IS4 better than whole β2GPI in the fluid phase implies that recombinant DI may be used as a competitive inhibitor of the pathogenic aPL/β2GPI interaction and thus act as a potential therapeutic agent. The ability to produce such an agent by expression in E. coli would be advantageous. Given that DI is produced in bacteria however sufficient measures should be undertaken to ensure expressed samples are free from lipopolysaccharide.
We have shown that, by using a stepwise strategy to address specific problems relating to codon optimisation, periplasmic protein translocation and tight regulation of expression, efficient bacterial expression of DI of human β2GPI is possible. Recombinant DI was conformationally correct and bound human monoclonal aPL in both solid and fluid phase assays. DI also bound a series of IgG samples derived from patients with APS. This is the first description of DI production using an E. coli expression system. Ease and efficiency of expression will be utilised to study different epitopes on DI and investigate the binding of variants of DI to polyclonal IgG affinity purified aPL derived from patients with APS. Furthermore if DI or a peptide derived from DI is eventually used in the therapeutic setting  an E. coli system of production would be likely to facilitate production and cost.
Finally, we submit that the simple design and production of a synthetic gene globally optimised for expression in E. coli using Juniper and one-step recursive PCR, as illustrated in this paper, is an important technique that can be applied to other eukaryotic proteins, particularly if the cDNA sequences of these genes have clusters of codons used infrequently by E. coli.
Oligonucleotide primers were synthesised by Thermobiosciences (Germany) and used without further purification. Restriction enzymes NcoI, XhoI and BglII and T4 DNA ligase were purchased from Promega, (Southampton, UK). Plasmids pET-26b(+) and BL21(DE3) E. coli cells were purchased from Novagen (Nottingham, UK). DH5α E. coli cells were supplied by Gibco (Paisley, UK). Automated sequencing was carried out by MWG-Biotech (Ebensburg, Germany).
Three human monoclonal antibodies (mAb) were produced in Chinese Hamster Ovary (CHO) cells, an in-vitro expression system, which has been described in detail in previous papers [25, 26]. Affinity purification of the antibodies from these cells was carried out by Chemicon Europe Ltd, Southampton, UK. Murine monoclonal anti-human DI antibodies mAb-16 and 6C4C10 were kind gifts from Dr M Linnik and Dr M Iverson, La Jolla Pharmaceuticals (LJP), California, USA.
Production of a construct containing a synthetic gene encoding the DI sequence
A synthetic gene was designed to encode for DI of human β2GPI, an N-terminal OmpA leader sequence and BglII/NcoI flanking restriction sites. This gene was designed by the computer programme Juniper, which designed six 60 mer overlapping oligonucleotide primers based on the published amino acid sequence of DI . The gene encoding for DI was synthesised using these primers by recursive PCR  (figure 1). Twenty pmol of each outer primer and two pmol of each internal primer were used in a reaction containing 2 U of Vent DNA polymerase (New England Biolabs (NEB), Hertfordshire, UK) and 25 mM of 2'-deoxynucleoside 5'-triphosphates (dNTPs') in 100 μl of the 10× supplied buffer and ddH20. PCR was performed under the following conditions: 95°C for 8 min, 30 cycles of 94°C for 2 min, 57°C for 1 min, 72°C for 1 min with a final extension step of 72°C for 10 min. This gene was then ligated into the expression plasmid and sequenced to exclude PCR errors.
The pET system originally described by Studier and Moffat  was used to express recombinant DI. A variety of constructs were used to optimise expression. The final cloned expression construct encoded for a pelB leader sequence followed by DI and a C-terminal his6-tag cloned into pET-26b(+). Both the leader and the his6-tag are present within the pET-26b(+) plasmid so that the OmpA leader sequence that had originally been created by recursive PCR (figure 1) was redundant and removed at this stage. The target gene was under the control of the high stringency phage T7lac promoter  and a strong T7 translation initiation site.
The final recombinant expression vector pET26b(+), containing the DI gene and a kanamycin resistance gene, was transferred to the expression strain BL21(DE3). Single colonies were picked from the transformants, and 5-ml cultures prepared in 50-ml falcons (overnight with shaking, 30°C). Fresh 500-ml cultures in 2.5-litre flasks were then set up using the overnight culture to inoculate the medium to an optical density (OD)600 of 0.1. Cultures were induced with a range of IPTG (0.1 mM, 0.4 mM or 1 mM) at an OD600 of 0.6 and allowed to grow for a further 4 hours (30°C, shaking). 'Terrific' broth containing 60 μg/ml of kanamycin was used to prepare the overnight cultures with the addition of 1% glucose for the expression cultures. OD600 was recorded at periodic intervals before and after induction as a measure of bacterial growth in the culture.
Preparation of periplasmic fraction and purification
The method for periplasmic extraction of protein from E. coli is based on previously published methods by our group for the expression of periplasmic Fabs of human autoantibodies in W3110 E. coli . Briefly, cultures were spun (4000 g, 20 min, 4°C) following 4 hours induction and supernatant was saved (at 4°C) for the detection of any leaked DI protein from the cells. The cell pellet was then exposed to osmotic shock by suspension in ice-cold water (30 ml of dH20 per litre of culture), stirred for 30 min at 4°C, and spun (8000 g, 20 min, 4°C). This supernatant constitutes the periplasmic fraction containing recombinant DI, an aliquot of which was stored at -20°C for subsequent SDS-PAGE and Western blot analysis.
For purification periplasmic extract was loaded on a column containing nickel charged resin, Novagen (Nottingham, UK). DI bound to the column by virtue of the C-terminal his6-tag and was eluted with 300 mM imidazole, (2 M NaCl, 80 mM Tris-HCl, pH 7.9). Recombinant his6-tagged DI was then dialysed extensively against PBS-10% glycerol overnight at 4°C using dialysis visking tubing, MWCO-3500, Medicell (London, UK). The purity of the eluted DI was assessed on SDS-PAGE 15% gels.
Affinity purification of IgG aPL
Polyclonal IgG was purified from 21 patients satisfying the preliminary diagnostic criteria for APS . IgG was also purified from 9 normal controls and 14 patients with SLE (but no APS) as disease controls. 19/21 patients with APS, 13/14 patients with SLE and 7/10 healthy controls were female. The mean ages of the subjects in the three groups were comparable (APS – 44.0 years, SD 7.0; SLE 37.2 years, SD 10.0; healthy controls 33.2 years, SD 9.0). Ethical approval for the study was granted by University College London Hospital Research Ethics Committee. Protein G beads (Amersham, Bucks, UK) were prepared by washing with PBS. 1 ml serum was mixed with 2 mls 0.02 M sodium phosphate (pH 7) binding buffer and incubated at room temperature (RT) for two hours with 0.5 ml prepared Protein G beads. This mixture was then spun (200 g, 5 min, 4°C) and the beads washed a further 3 times with binding buffer. Elution of IgG from the beads was performed by mixing the beads with 2 mls 0.1 M glycine (pH 2.7) for 1 min. This was spun (200 g, 5 min, 4°C) and the supernatant stored as the IgG fraction at -20°C. The amount of IgG was quantified using a direct IgG ELISA described in a previous paper .
Direct binding ELISA of aPL binding to purified recombinant his6-tagged DI
Nickel chelate-coated microwell plates, VH Bio (Gateshead, UK) were coated with 50 μl of recombinant purified his6-tagged DI and diluted to a concentration of 50 μg/ml using PBS. One half of the plate (the test wells) was coated with DI and the other half with PBS (the control wells). Plates were incubated at RT for 2 hours and were then washed three times with PBS, blocked with 100 μl of 0.25% gelatin, Sigma (Poole, UK) in PBS and incubated for a further 1 hour at RT. After washing the plates three times with PBS, 50 μl of a monoclonal human IgG aPL (IS4VH/IS4VL, IS4VH/B3VL or IS4VH/UK-4VL) derived from CHO cell culture supernatant  in sample, enzyme and conjugate dilution (SEC) buffer (100 mM Tris-HCl (pH 7), 100 mM NaCl, 0.02% Tween-20 and 0.2% BSA) was added at concentrations ranging from 9 ng/ml to 80 ng/ml. Serial dilutions of mAb were loaded such that for each dilution loaded in a test well, there was a corresponding control well loaded with the same dilution. For polyclonal IgG aPL purified from APS patients, SLE and normal controls, 50 μl of 20 μg/ml of IgG in SEC buffer was added to each well. Binding of human antibodies to the plate was detected by adding goat anti-human IgG alkaline phosphatase conjugate, Sigma (Poole, UK) diluted 1:2000 in SEC. After incubation at 37°C for 1 hour, bound antibody was detected by addition of alkaline phosphatase chromogenic substrate. The OD was measured in a Genios microplate autoreader (Tecan, Reading, UK). A net OD was calculated for each well to take into account background (OD test well – OD control well). Results of polyclonal IgG were expressed as a percentage binding of a standard APS patient sample known to bind DI, whole β2GPI and CL.
Competitive inhibition ELISA
A direct binding ELISA carried out as above was used to determine the concentration of native affinity purified antibody IS4VH/IS4VL required to achieve ~50% maximum binding. This concentration was 200 ng/ml. DI and β2GPI as test inhibitors were diluted in PBS at concentrations ranging from 0 (i.e. no inhibitor) to 30 μM. Affinity-purified IS4VH/IS4VL was then added to each concentration of inhibitor to achieve 200 ng/ml final concentration of antibody for each sample. The samples were incubated at RT for 2 hours and then tested for binding to DI on an ELISA plate as described earlier for the direct ELISA above. The per cent inhibition for each concentration of inhibitor was determined from the following formula:
% inhibition=(A0-A/A0) × 100, where A is the OD from the well containing the inhibitor (corrected for background) and A0 is the OD from the well containing no inhibitor (corrected for background).
beta 2 glycoprotein I
- E. coli :
systemic lupus erythematosus.
We would like to thanks Drs M Linnik and M Iverson of (LJP, CA, USA) for the donation of murine monoclonal anti-DI antibodies and helpful advice regarding biological properties of DI. We are also indebted to Dr Pojen Chen (UCLA, CA, USA) for the donation of IS4 and Drs Miratul Muqit and Barry Ripley for help regarding methodology of some experiments. This work was funded by the Arthritis Research Campaign, UK.
- Wilson WA, Gharavi AE, Koike T, Lockshin MD, Branch DW, Piette JC, Brey R, Derksen R, Harris EN, Hughes GR, Triplett DA, Khamashta MA: International consensus statement on preliminary classification criteria for definite antiphospholipid syndrome: report of an international workshop. Arthritis Rheum. 1999, 42 (7): 1309-1311. 10.1002/1529-0131(199907)42:7<1309::AID-ANR1>3.0.CO;2-F.View ArticleGoogle Scholar
- Cervera R, Piette JC, Font J, Khamashta MA, Shoenfeld Y, Camps MT, Jacobsen S, Lakos G, Tincani A, Kontopoulou-Griva I, Galeazzi M, Meroni PL, Derksen RH, de Groot PG, Gromnica-Ihle E, Baleva M, Mosca M, Bombardieri S, Houssiau F, Gris JC, Quere I, Hachulla E, Vasconcelos C, Roch B, Fernandez-Nebro A, Boffa MC, Hughes GR, Ingelmo M: Antiphospholipid syndrome: clinical and immunologic manifestations and patterns of disease expression in a cohort of 1,000 patients. Arthritis Rheum. 2002, 46 (4): 1019-1027. 10.1002/art.10187.View ArticleGoogle Scholar
- Khamashta MA, Cuadrado MJ, Mujic F, Taub NA, Hunt BJ, Hughes GR: The management of thrombosis in the antiphospholipid-antibody syndrome. N Engl J Med. 1995, 332 (15): 993-997. 10.1056/NEJM199504133321504.View ArticleGoogle Scholar
- Pierangeli SS, Harris EN: In vivo models of thrombosis for the antiphospholipid syndrome. Lupus. 1996, 5 (5): 451-455.Google Scholar
- Pierangeli SS, Liu X, Espinola R, Olee T, Zhu M, Harris NE, Chen PP: Functional analyses of patient-derived IgG monoclonal anticardiolipin antibodies using in vivo thrombosis and in vivo microcirculation models. Thromb Haemost. 2000, 84 (3): 388-395.Google Scholar
- Steinkasserer A, Estaller C, Weiss EH, Sim RB, Day AJ: Complete nucleotide and deduced amino acid sequence of human beta 2- glycoprotein I. Biochem J. 1991, 277 (Pt 2): 387-391.View ArticleGoogle Scholar
- Hunt J, Krilis S: The fifth domain of beta 2-glycoprotein I contains a phospholipid binding site (Cys281-Cys288) and a region recognized by anticardiolipin antibodies. J Immunol. 1994, 152 (2): 653-659.Google Scholar
- Iverson GM, Victoria EJ, Marquis DM: Anti-beta2 glycoprotein I (beta2GPI) autoantibodies recognize an epitope on the first domain of beta2GPI. Proc Natl Acad Sci U S A. 1998, 95 (26): 15542-15546. 10.1073/pnas.95.26.15542.View ArticleGoogle Scholar
- Reddel SW, Wang YX, Sheng YH, Krilis SA: Epitope studies with anti-beta 2-glycoprotein I antibodies from autoantibody and immunized sources. J Autoimmun. 2000, 15 (2): 91-96. 10.1006/jaut.2000.0427.View ArticleGoogle Scholar
- Giles IP, Isenberg DA, Latchman DS, Rahman A: How do antiphospholipid antibodies bind beta2-glycoprotein I?. Arthritis Rheum. 2003, 48 (8): 2111-2121. 10.1002/art.11101.View ArticleGoogle Scholar
- Horizon A, Weisman MH, Wallace DJ, Merrill JT, Linnik MD, Cockerill KA, Furie RA: Results of a Randomised, Placebo Controlled, Double-blind Phase 1/2 Clinical Trial (RCT) to Assess the Safety and Tolerability of LJP 1082 in Patients with Antiphospholipid Syndrome. Arthritis Rheum. 2003, 48 (9 (Supp)): S364-Google Scholar
- Hannig G, Makrides SC: Strategies for optimizing heterologous protein expression in Escherichia coli. Trends Biotechnol. 1998, 16 (2): 54-60. 10.1016/S0167-7799(97)01155-4.View ArticleGoogle Scholar
- Sorensen HP, Mortensen KK: Soluble expression of recombinant proteins in the cytoplasm of Escherichia coli. Microb Cell Fact. 2005, 4 (1): 1-10.1186/1475-2859-4-1.View ArticleGoogle Scholar
- Igarashi M, Matsuura E, Igarashi Y, Nagae H, Ichikawa K, Triplett DA, Koike T: Human beta2-glycoprotein I as an anticardiolipin cofactor determined using mutants expressed by a baculovirus system. Blood. 1996, 87 (8): 3262-3270.Google Scholar
- Prodromou C, Pearl LH: Recursive PCR: a novel technique for total gene synthesis. Protein Eng. 1992, 5 (8): 827-829.View ArticleGoogle Scholar
- Kane JF: Effects of rare codon clusters on high-level expression of heterologous proteins in Escherichia coli. Curr Opin Biotechnol. 1995, 6 (5): 494-500. 10.1016/0958-1669(95)80082-4.View ArticleGoogle Scholar
- Hua Z, Wang H, Chen D, Chen Y, Zhu D: Enhancement of expression of human granulocyte-macrophage colony stimulating factor by argU gene product in Escherichia coli. Biochem Mol Biol Int. 1994, 32 (3): 537-543.Google Scholar
- Hu X, Shi Q, Yang T, Jackowski G: Specific replacement of consecutive AGG codons results in high-level expression of human cardiac troponin T in Escherichia coli. Protein Expr Purif. 1996, 7 (3): 289-293. 10.1006/prep.1996.0041.View ArticleGoogle Scholar
- Martin SL, Vrhovski B, Weiss AS: Total synthesis and expression in Escherichia coli of a gene encoding human tropoelastin. Gene. 1995, 154 (2): 159-166. 10.1016/0378-1119(94)00848-M.View ArticleGoogle Scholar
- Dubendorff JW, Studier FW: Creation of a T7 autogene. Cloning and expression of the gene for bacteriophage T7 RNA polymerase under control of its cognate promoter. J Mol Biol. 1991, 219 (1): 61-68. 10.1016/0022-2836(91)90857-3.View ArticleGoogle Scholar
- Grossman TH, Kawasaki ES, Punreddy SR, Osburne MS: Spontaneous cAMP-dependent derepression of gene expression in stationary phase plays a role in recombinant expression instability. Gene. 1998, 209 (1-2): 95-103. 10.1016/S0378-1119(98)00020-1.View ArticleGoogle Scholar
- Iverson GM, Reddel SW, Victoria EJ, Cockerill KA, Wang YX, Marti-Renom MA, Sali A, Marquis DM, Krilis SA, Linnik MD: Use of single point mutations in domain I of ß2-glycoprotein I to determine fine antigenic specificity of antiphospholipid autoantibodies. J Immunol. 2002, 169: 7097-7103.View ArticleGoogle Scholar
- Pierangeli SS, Gharavi AE, Harris EN: Experimental thrombosis and antiphospholipid antibodies: new insights. J Autoimmun. 2000, 15 (2): 241-247. 10.1006/jaut.2000.0420.View ArticleGoogle Scholar
- Lambrianides N, Giles IP, Latchman DS, Isenberg DA, Rahman A: Stable transfection of CHO cells to express whole human IgG antiphospholipid antibodies. Rheumatology. 2004, 43 (S2): 263-Google Scholar
- Giles IP, Haley J, Nagl S, Latchman DS, Chen PP, Chukwuocha RU, Isenberg DA, Rahman A: Relative importance of different human aPL derived heavy and light chains in the binding of aPL to cardiolipin. Mol Immunol. 2003, 40 (1): 49-60. 10.1016/S0161-5890(03)00100-7.View ArticleGoogle Scholar
- Rahman MA, Kettleborough CA, Latchman DS, Isenberg DA: Properties of whole human IgG molecules produced by the expression of cloned anti-DNA antibody cDNA in mammalian cells. J Autoimmun. 1998, 11 (6): 661-669. 10.1006/jaut.1998.0241.View ArticleGoogle Scholar
- Studier FW, Moffatt BA: Use of bacteriophage T7 RNA polymerase to direct selective high-level expression of cloned genes. J Mol Biol. 1986, 189 (1): 113-130. 10.1016/0022-2836(86)90385-2.View ArticleGoogle Scholar
- Kumar S, Kalsi J, Ravirajan CT, Rahman A, Athwal D, Latchman DS, Isenberg DA, Pearl LH: Molecular cloning and expression of the Fabs of human autoantibodies in Escherichia coli. Determination of the heavy or light chain contribution to the anti-DNA/-cardiolipin activity of the Fab. J Biol Chem. 2000, 275 (45): 35129-35136. 10.1074/jbc.M001976200.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.