Expression of human epidermal growth factor using Ssp DnaB mini-intein as fusion partner in Escherichia coli BL21(DE3)

The human epidermal growth factor (hEGF) is widely used clinically as a wound healer, as it has a vital role in stimulating cell proliferation, differentiation, and migration. Consequently, the large-scale production of recombinant hEGF in E. coli has been developed to meet the high demand for hEGF clinically. However, intracellular proteins in E. coli, especially small proteins like hEGF, are degraded by proteases. To overcome this issue, hEGF was fused with CBD-Ssp DnaB to construct a fusion protein CBD-Ssp DnaB-hEGF. This study was conducted to obtain refolded hEGF from the inclusion bodies of CBD-Ssp DnaB-hEGF. The experiment was carried out using E. coli BL21(DE3) containing plasmid pD861-CBD-Ssp DnaB-hEGF. The CBD-Ssp DnaB-hEGF gene was constructed by fusing CBDSsp DnaB and hEGF gene and then was optimized. The method was started with E. coli transformation, CBD-Ssp DnaB-hEGF expression, inclusion bodies solubilization, refolding, and simultaneous cleavage to release hEGF. The CBD-Ssp DnaB-hEGF was expressed as inclusion bodies, which can then be purified by washing with Triton X-100 and 1 M urea. The inclusion bodies were solubilized in 8 M urea, the solubilized CBD-Ssp DnaB-hEGF was reformed by dialysis, and then hEGF was spliced by shifting the pH from 8.5 to 6.0 to yield a concentration of 0.163 mg/ml. Therefore, we concluded that hEGF was obtained from the solubilized CBD-Ssp DnaB-hEGF from inclusion bodies produced by E. coli BL21(DE3).


INTRODUCTION
Human epidermal growth factor (hEGF) is a protein with 53 amino acid residues with 6.2 kDa of molecular weight. This polypeptide has three intramolecular disulfide bonds and heat resistance properties (Eissazadeh et al., 2017;Sriwidodo et al., 2019;Tang et al., 2016). hEGF plays a role in stimulating cell proliferation, differentiation, and migration in the wound-healing process. This has led to the high demand for hEGF in the clinical field, thus encouraging efforts to increase hEGF production through recombinant DNA technology (Ma et al., 2016;Zheng et al., 2016).
E. coli was one of the most widely used hosts for recombinant protein expression (Hayat et al., 2018). It has a high rate of protein expression and rapid growth, is inexpensive, and produces culture in the high-density cell (Kaur et al., 2018;Kim et al., 2017;. There were previous studies on the extracellular expression of hEGF in E. coli using various signal peptides (Indriyani et al., 2019;Maksum et al., 2017;Melati et al., 2019;. Nevertheless, the extracellular expression system has a lower expression rate than the intracellular system (Su et al., 2006). To overcome this problem, the intracellular expression approach was currently used to increase hEGF expression. Intracellular protein expression in E. coli especially small protein such as hEGF mostly did not work well because the small protein was easily degraded by proteases from the host cell. This causes a significant loss in yields and made it difficult in the purification process. Alternatively, the protein was fused with a fusion partner so the target protein could avoid degradation by proteases (Sun et al., 2005;Zhang et al., 2015). Several fusion tags were commonly used in fusion protein expression systems such as glutathione S-transferase, small ubiquitin-related modifier, thioredoxin, and maltose-binding protein. However, there were some difficulties in fusion tag removal such as unspecific cleavage site, the steric hindrance at the cleavage site, and the expensive cost of specific protease (Costa et al., 2014;Zheng et al., 2016).
The intein is a protein segment that can cleavage itself from protein precursors where it fuses two segments of flanking proteins or amino acids (Li, 2011). The intein-mediated expression system has been developed as a powerful tool for protein expression that utilized its ability to self-cleavage. Target protein synthesized in the fusion of intein form which has been genetically designed to do controllable cleavage from peptide bond at either N-or C-terminal of the intein. After protein cleavage by intein, the target protein was produced with the desired terminal amino acid (Jiang et al., 2015). Ssp DnaB miniintein derived from Synechocystis sp. was used in the commercial plasmid, such as IMPACT TM system (Ding et al., 2003). The release of the target protein can be induced simply by changing pH, giving the advantage of cleaving peptide bonds without using specific protease (Sun et al., 2005).
The expression of recombinant protein in E. coli frequently leads to the formation of inclusion bodies. Inclusion bodies are protein aggregate that consists of partially folded and misfolded protein that interacts via hydrophobic interaction. Their formation is often undesirable because they generally lack biological activity (Gomes et al., 2016;Silaban et al., 2019;Singh and Panda, 2005). This approach can give the advantage by refolding the inactive protein in inclusion bodies into its active form (Upadhyay et al., 2016). Moreover, inclusion bodies are produced in high amount and easy to separate because they were denser than the other cellular components (Singh and Panda, 2005;Singh et al., 2015). To recover active protein from inclusion bodies, inclusion bodies were solubilized in presence of chaotropic agents such as urea and guanidine chloride and afterward solubilized protein was refolded by decreasing the concentration of chaotropic agents (Singh et al., 2015;Vallejo and Rinas, 2004).
In this report, we construct fusion protein CBD-Ssp DnaB-hEGF by fused hEGF gene with the CBD-Ssp DnaB gene in pD861 plasmid. E. coli BL21(DE3) was transformed with recombinant plasmid. CBD-Ssp DnaB-hEGF was successfully expressed as inclusion bodies. Inclusion bodies were isolated, solubilized, and refolded. Ssp DnaB mini-intein splicing activity could be achieved by gradually shifting pH from 8.5 to 6.0. After Ssp DnaB mini-intein cleavage, hEGF protein was obtained with a concentration of 0.163 mg/ml. The result demonstrated that hEGF protein could be recovered by inclusion bodies obtained from intracellular expression in E. coli.

Expression of recombinant CBD-Ssp DnaB-hEGF fusion protein
To express CBD-Ssp DnaB-hEGF, CaCl 2 competent E. coli BL21(DE3) cells were transformed with pD861-CBD-Ssp DnaB-hEGF plasmid using the heat shock method. Transformed cells were grown in LB agar plates supplemented with 50 μg/ ml kanamycin at 37°C for 16-18 hours. Single colonies were replicated and the selected replica was grown in 5 ml LB medium supplemented with 50 μg/ml kanamycin at 37°C for 16-18 hours. A total of 1 ml of starter culture was transferred into 100 ml LB medium and incubated at 37°C with shaking at 200 rpm until OD 600 0.6 was reached and then was induced by adding L-rhamnose until the final concentration of 4 mM and incubated for 5 hours. The culture was centrifuged at 6,000 g for 20 minutes at 4°C. The cell pellet was stored at −20°C for further analysis.

Inclusion bodies preparation
Cell pellet (approximately 2 g wet weight) were resuspended in 1:4 ratio with lysis buffer (20 mM Tris-Cl, 1 mM EDTA, and pH 8.5); then, the suspension was sonicated for 1 minute (1 second on and 1 second off pulse). Sonication was performed on ice for 10 cycles with 1 minute gap for each cycle. The lysed cell suspension was centrifuged at 12,000 g for 20 minutes at 4°C. The pellet obtained was washed twice using washing buffer I (20 mM Tris-Cl, 1 mM EDTA, 1% Triton X-100, 1 M urea, and pH 8.5), once using washing buffer II (20 mM Tris-Cl, 1 mM EDTA, 1% Triton X-100, and pH 8.5), and twice using washing buffer III (20 mM Tris-Cl and pH 8.5). Purified inclusion bodies were resuspended in 20 mM Tris-Cl and pH 8.5 and used for solubilization.

Optimization of urea concentration for solubilization of inclusion bodies
Purified inclusion bodies' suspension was solubilized with 20 mM Tris-Cl pH 8.5 containing different urea concentrations to optimize solubilization conditions for inclusion bodies of CBD-Ssp DnaB-hEGF fusion protein. Inclusion bodies' suspension was solubilized with 1:10 solubilizing buffer (20 mM Tris-Cl and pH 8.5) with different concentrations of urea (0-8 M urea). The mixture was incubated for 2 hours at room temperature with shaking at 150 rpm and after that centrifuged at 13,000 rpm for 20 minutes at 4°C. Solubilized inclusion bodies were characterized by sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) and analyzed by ImageJ software.

Construction CBD-Ssp DnaB-hEGF gene in recombinant plasmid
CBD-Ssp DnaB-hEGF gene was constructed from the sequence of the hEGF (GeneBank GQ214314.1) and CBD-Ssp DnaB (GeneBank HG792069.1). The first methionine from hEGF was eliminated and fused to asparagine at the end of the C-terminal of CBD-Ssp DnaB (Fig. 1). Asparagine residue on C-terminal of Ssp DnaB mini-intein was directly fused to alanine residue on N-terminal of hEGF. The expression rate of heterologous protein on E. coli was closely related to the codon usage bias phenomenon (Gomes et al., 2016). Bias codon could be avoided by optimizing the CBD-Ssp DnaB-hEGF gene to have high CAI value and GC content in the range of 30%-70% (Parret et al., 2016;Silaban et al., 2014;. CBD-Ssp DnaB-hEGF gene was optimized by using E. coli preference codon to achieve better protein expression. The optimized CBD-Ssp DnaB-hEGF gene had a length of 843 bp. CAI value of gene was 1.00 with GC content of 51.89%. CBD-Ssp DnaB-hEGF gene was inserted into plasmid pD861 and synthesized by ATUM (Fig. 2). The promoter was a key component in protein expression because it plays a role in regulating the expression of the gene of interest. L-rhamnose promoters provide more tight regulation which will prevent leakage of protein (Kaur et al., 2018). Therefore, it is assumed to be able to increase protein acquisition after the addition of an inducer. The use of plasmid with high copy numbers and a strong ribosomal binding site can also increase the rate of protein expression (Hu et al., 2012;Marschall et al., 2016).

Expression of recombinant CBD-Ssp DnaB-hEGF and inclusion bodies preparation
Recombinant CBD-Ssp DnaB-hEGF was successfully expressed as 31.2 kDa protein in E. coli. CBD-Ssp DnaB-hEGF was majorly expressed in inclusion bodies form compared to soluble form (Fig. 3a). Induction at mid-log phase (OD 600 0.6) provides a high expression rate due to E. coli rapid growth which leads to a higher protein expression rate (Fazaeli et al., 2019;Kaur et al., 2018). In a previous study, the L-rhamnose promoter was successfully expressing recombinant protein extracellularly using signal peptide (Indriyani et al., 2019;. However, a recent study showed that the L-rhamnose promoter   can also be utilized for intracellular expression of CBD-Ssp DnaB-hEGF as well. Inclusion bodies were washed by detergent to remove contaminants such as cell membrane and cellular protein following aggregate with inclusion bodies. Triton-X100 was used to wash cell membranes and also the addition of a low concentration of urea could help to remove cellular protein (Qi et al., 2015;Singh et al., 2015). Purified inclusion bodies were obtained after several washing steps (Fig. 3b) and it was used for solubilization and refolding.

Optimization of urea concentration for solubilizing inclusion bodies
Different concentration of urea (0-8 M) was used to solubilize CBD-Ssp DnaB-hEGF from inclusion bodies. Urea would help to disrupt hydrophobic interaction between proteins in inclusion bodies resulting in solubilized CBD-Ssp DnaB-hEGF (Patra et al., 2000;Singh and Panda, 2005). CBD-Ssp DnaB-hEGF was slightly solubilized from inclusion bodies in the presence of 2 M urea and conversely showed better solubility in the presence of 6-8 M urea. CBD-Ssp DnaB-hEGF solubility tended to increase when the concentration of urea was increased (Fig. 4a). The highest amount of solubilized CBD-Ssp DnaB-hEGF was achieved by using a solubilization buffer containing 8 M urea (Fig.  4b). This result showed that the solubilization buffer containing 8 M urea could give better solubilization of CBD-Ssp DnaB-hEGF from inclusion bodies. The use of higher concentration of urea will give more solubilized protein that would be in denatured state, so refolding was necessary to convert protein into its native conformation by removal of denaturant through dialysis (Yamaguchi et al., 2013).

Refolding and simultaneous cleavage of solubilized CBD-Ssp DnaB-hEGF
Solubilized CBD-Ssp DnaB-hEGF was refolded by lowering the urea concentration using a stepwise dialysis method. The dialysis method could prevent protein loss because of the uncontrolled cleavage of Ssp DnaB mini-intein (Zhang et al., 2015). The concentration of urea was lowered from 8 to 2 M then  until approximately reaching 0 M. No aggregate was observed after dialysis; this showed that the refolding process was successful. Solubilized CBD-Ssp DnaB-hEGF started to refold as the urea concentration was decreasing (Singh et al., 2015;Yamaguchi and Miyazaki, 2014). On the other side, a proper disulfide bond was formed from reduced disulfide bond through disulfide shuffling while stepwise dialysis was performed (Yamaguchi et al., 2013). Even stepwise dialysis was taking more time and "slow," this method could give good results. There was no aggregation observed after dialysis was performed.
CBD-Ssp DnaB-hEGF shows Ssp DnaB splicing activity which is characterized by a band at 25 kDa; this band belonged to CBD-Ssp DnaB (Fig. 5a). The pH condition for splicing was investigated (pH 8.0-6.0) to ensure the pH optimum for splicing. Protein splicing occurred even at pH 8.0 and so did at the lower pH until pH 6.0 (Fig. 5b). This showed that gradually lowering pH until 6.0 could induce Ssp DnaB splicing activity. hEGF was released from CBD-Ssp DnaB-hEGF, which showed by a band at 6.2 kDa (Fig. 5c). Splicing activity of Ssp DnaB intein involved charge relay between amino acid residues and was ended by asparagine cyclization at C-terminal of intein (Ding et al., 2003;Topilina and Mills, 2014). Uncontrolled splicing activity was caused by Ssp DnaB intein which was sensitive to pH changes (Chen et al., 2002) and the cleavage was observed even at pH 8.0. By gradually decreasing pH from 8.0 to 6.0, hEGF was obtained. hEGF was obtained with a concentration of 0.163 mg/ml which was quantified by the densitometry method.

CONCLUSION
In summary, CBD-Ssp DnaB-hEGF was successfully expressed as inclusion bodies in E. coli BL21(DE3). Refolded hEGF was recovered from inclusion bodies after isolation, solubilization, and refolding steps with a concentration of 0.163 mg/ml. The solubilization and refolding method presented in this study could be used for the recovery of refolded hEGF from CBD-Ssp DnaB-hEGF inclusion bodies. The result shows a great opportunity to produce large amounts of hEGF from bacterial inclusion bodies. This finding leads to the optimization and development method in the future for maximizing the recovery of refolded hEGF. So, it could be applicable for a larger scale, especially at the industrial scale.