Introduction

Proteins or peptides required for structural and biochemical study are usually produced via bacterial expression systems when posttranslational modifications are not required (1,2). In many cases, however, poor expression, low yield, instability, and difficulty in purification are complications (2,3). Furthermore, production of heterogeneous proteins in Escherichia coli often results in the accumulation of targets in insoluble aggregates, forming inclusion bodies. Recovery of the biological activity of the target protein by an in vitro refolding process is a difficult task (4).

Protein fusion technology is widely used to improve expression, purification, and solubility of recombinant protein expressed in E. coli. Several proteins such as maltose-binding protein (MBP), thioredoxin (TRX), transcription pausing factor L (NusA), thiol-disulfide oxidoreductase, glutathione S-transferase (GST), and others have been reported to enhance the solubility of fused protein (2,5,6). However, no system or tool respresents a complete solution, and improvement is always necessary. For example, among these proteins, MBP was found more effective in solubilizing fused proteins than GST and TRX, but it was unable to promote proper folding of the fused proteins, and in many cases, it could not enhance expression of difficult-to-express proteins (6,7,8).

Recent reports have described the use of some thermostable proteins as fusion partners to improve the stability and purification of targets fused to them (9). In this context, we examined several E. coli thermostable proteins that might be capable of enhancing solubility of target proteins or peptides in order to find improved fusion partners. By identifying an appropriate protein for use as a fusion partner, target protein and peptides that form inclusion bodies in the conventional expression system were successfully expressed and purified without the need for solubilization and recovery.

Materials and Methods

Identification of Thermostable Proteins from E. coli

Whole cell lysate was prepared from E. coli DH5α cells grown at 37°C overnight by 4 to 5 passages through a French pressure cell at 18,000 lb/in² in lysis buffer A (20 mM Tris-Cl, pH 8.0, 10 mM NaCl, 0.1 mM phenylmethane-sulphonylfluoride, 0.1 mM EDTA, and 1 mM 2-mercaptoethanol) and centrifugation at 47,000× g for 1 h. The soluble fraction (cell extract) was boiled for 10 min, centrifuged at 16,500× g for 30 min, and the soluble fraction was again recovered. Proteins in the fraction were resolved by native 10% polyacrylamide gel electrophoresis (PAGE). Then, the gel portion containing proteins was excised and denatured by boiling in the presence of 0.1% 2-mercaptoethanol for 10 min, followed by sodium dodecyl sulfate PAGE (SDS-PAGE). The selected protein spots were analyzed by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF; ProteomeTech. Inc., Seoul, Korea) and identified by National Center for Biotechnology Information (NCBI) BLAST search on the basis of probability.

Cloning of Thermostable Proteins

Genes encoding the identified thermostable proteins were PCR-amplified using E. coli DH5α genomic DNA as a template and cloned in NdeI and BamHI sites (or NdeI/HindIII for FK-506-binding protein, FKBP) of the pET21b expression system (Novagen, Madison, WI, USA). The primers used for cloning DnaK, trigger factor (TF), MBP, D-ribose periplasmic protein (RbsB), putative EscN protein (EscN), GroES, FKBP, Adenylate kinase (AKN), and NusA are shown in Table 1.

Table 1. Primers Used for PCR (Click to enlarge)

Expression and Heat Stability of the Prospective Thermostable Cloned Genes

Expression of proteins in E. coli BL21 (DE3) pLysS transformed with each plasmid was induced with 0.4 mM isopropyl-β-D-thiogalactopyranoside (IPTG) at an A₆₀₀ of ∼0.6, after which the cells were cultured at 37°C for an additional 4 h. Cells were harvested, lysed in the lysis buffer A, disrupted by a brief sonication, and centrifuged at 16,500× g for 1 h. SDS-PAGE analysis of the supernatants before and after boiling for 10 min was carried out to examine the expression and thermostability of the proteins.

Construction of Thermostable Protein Ubiquitin Target Fusion Plasmids

Amyloid β_(1–42) [Aβ_(1–42)] was PCR-amplified with appropriate primers (labeled Aβ₄₂ in Table 1) using the Aβ_(1–42) gene as a template and cloned in the BamHI and XhoI sites of pET28b (Novagen). Ubiquitin (kindly provided by Rohan T. Baker, Australian National University) was cloned upstream of the Aβ_(1–42) gene using NdeI/BamHI sites. Then, the ubiquitin-Aβ_(1–42) fusion was PCR-amplified (primers: Ub-Aβ42 in Table 1) after removal of the BamHI site by site-directed mutagenesis [primers: Ub-Aβ42 (BamHI) in Table 1] and cloned into the plasmid containing the thermostable protein to be tested. This step used the BamHI and XhoI sites, placing the ubiquitin-Aβ_(1–42) fusion downstream of the thermostable protein. The ubiquitin-Aβ_(1–42) construct was prepared by insertion of a termination codon two amino acid residues from the end of the Aβ_(1–42) sequence (primers: Ub-Aβ40 in Table 1). The ubiquitin-Aβ_(1–42) [a reverse form of Aβ_(1–42)] fusion was synthesized in a multistep process using the primers shown in Table 1 [Ub-Aβ42 forward and Aβ_(42-1)], after which it was subcloned after the TF fusion protein. Humanin peptide (10,11) was synthesized using the primers listed in Table 1 and subcloned as described above. To construct reverse-caspase-2 (12), the small and large subunit of the protein were PCR-amplified separately (primers: Casp-2 SS and LS in Table 1). The small subunit of the full-length caspase-2 was cloned in pET21b downstream of the large subunit. Finally, the whole structure was subcloned into the BamHI/XhoI sites of a newly constructed TF-ubiquitin plasmid. The plasmid was constructed by PCR amplification with appropriate primers (listed as reverse-caspase-2 in Table 1) and then cloned into the NdeI/BamHI sites of the TF-ubiquitin pET21b expression system, after removing the BamHI site located between TF and ubiquitin.