Preparation of recombinant human thymic stromal lymphopoietin protein

Human thymic stromal lymphopoietin (hTSLP) protein plays a central role in inflammation. Characterizing properties of hTSLP requires a recombinant overexpression system that produces correctly folded, active hTSLP. In this report, an efficient overexpression system for the production of hTSLP was developed. We constructed expression plasmids of the full-length hTslp gene with or without the signal peptide and transformed the plasmids into Escherichia coli. The design of the recombinant proteins included an N-terminal His-tag, which facilitated purification. An affinity gradient elution method was used to improve recovery and concentration levels of denatured hTSLP, with 90% purity observed following affinity chromatography. Refolding of the denatured hTSLP was tested using four different protein refolding approaches. The optimal refolding conditions involved stepwise buffer exchanges to reduce the urea concentration from 4 to 0?M in 50?mM Tris (pH 8.0), 1?mM EDTA, 50?mM NaCl, 10% glycerol, 400?mM L-Arg, 0.2?mM oxidized glutathione, and 2?mM reduced glutathione. The activity of the refolded recombinant hTSLP protein was measured by an ELISA assay. Interestingly, the presence of N-terminal signal peptide inhibited the overexpression of hTSLP in E. coli. The amount of recombinant hTSLP protein purified reached a level of 2.52?×?10^?3?mg/L.     

NMR structures of salt-refolded forms of the 434-repressor DNA-binding domain in 6 M urea

The N-terminal 63-residue fragment of the phage 434-repressor, 434(1-63), has a well-defined globular fold in H(2)O solution, and is unfolded in 6 M urea at pH 7.5. In this study, 434(1-63) has been refolded by adding either 1.7 M NaCl or 0.47 M NaTFA to the solution in 6 M urea, and the NMR structures of both refolded forms have been determined. The two refolded forms have similar free energies of unfolding and are approximately 16 kJ/mol less stable than the protein in H(2)O solution. 434(1-63) refolded with NaCl exhibits NMR chemical shifts very similar to and a three-dimensional structure nearly identical to those of 434(1-63) in H(2)O solution. The protein refolded with NaTFA also has a similar global fold, but it shows local differences near Phe44, of which two different orientations of the aromatic ring are compatible with the experimental data. This local conformational polymorphism attracted our interest because hydrophobic contacts between two subdomains of residues 1-36 and 45-63 are mediated by the Phe44 side chain. Anion binding experiments suggest that this polymorphism is caused by binding of TFA(-) anions to a cluster of positively charged Arg and Lys residues located in the loop connecting the two subdomains, with apparent binding constants for TFA(-) (K(app)) on the order of 30 mM(-1).     

Expression, purification, and in vitro refolding of a humanized single-chain Fv antibody against human CTLA4 (CD152)

A human-derived single-chain Fv (scFv) antibody fragment specific against human CTLA4 (CD152) was produced at high level in Escherichia coli. The scFv gene was cloned from a phagemid to the expression vector pQE30 with a N-terminal 6His tag fused in-frame, and expressed as a 29 kDa protein in E. coli as inclusion bodies. The inclusion body of scFv was isolated from E. coli lysate, solubilized in 8M urea with 10mM dithiothreitol, and purified by ion-exchange chromatography. Method for in vitro refolding of the scFv was established. The effects of refolding buffer composition, protein concentration and temperature on the refolding yield were investigated. The protein was renatured finally by dialyzing against 3mM GSH, 1mM GSSG, 150 mM NaCl, 1M urea, and 50 mM Tris-Cl (pH 8.0) for 48 h at 4 degrees C, and then dialyzed against phosphate-buffered saline (pH 7.4) to remove remaining denaturant. This refolding protocol generated up to a 70% yield of soluble protein. Soluble scFv was characterized for its specific antigen-binding activity by indirect cellular ELISA. The refolded scFv was functionally active and was able to bind specifically to CTLA4 (CD152). The epitopes recognized by refolded anti-CTLA4 scFv do not coincide with those epitopes recognized by CD80/CD86.     

Production and purification of refolded recombinant Plasmodium falciparum beta-ketoacyl-ACP reductase from inclusion bodies

A recombinant form of Plasmodium falciparum beta-ketoacyl-ACP reductase (PfFabG) was overexpressed in Escherichia coli BL-21 codon plus (DE3). The resulting insoluble inclusion bodies were separated from cellular debris by extensive washing with buffer containing 0.05% Tween 20 and solubilized by homogenization with 8 M urea. Attempts to refold PfFabG from solubilized inclusion bodies employing Rotofor (separation based on different pIs of proteins in a mixture) followed by Ni(2+) or cation exchange chromatography were not successful either by bringing down the urea concentration instantaneously, stepwise, or by dialysis. Denatured PfFabG was therefore initially purified by cation exchange chromatography and was then correctly refolded at a final concentration of 100-200 microg/ml in a 20 mM Na-acetate buffer, pH 5.3, with 300 mM NaCl, 10% glycerol, and 0.05% Tween 20. The protein was found to be properly folded only in the presence of the cofactor NADPH and salt at a concentration 300 mM by drop dilution method at 2-8 degrees C for 12 h. The purified final product was >98% pure by denaturing gel electrophoresis. The purified protein was biologically active in a standard enzymatic assay using acetoacetyl-CoA as a substrate. The enzyme was found to be stable up to fourth day of purification and glycerol was found to stabilize enzyme activity for several weeks, during storage. This effort paves the way for elucidation of the structure-function correlations for PfFabG as well as exploration of the enzyme for developing inhibitors against it for combating malaria.     

Striated fibers of the rho form of Mycoplasma: in vitro reassembly, composition, and structure.

The rho-form of Mycoplasma contains a striated, axial fiber and associated terminal structure. The presence of this organelle was correlated with the synthesis of two proteins, A and B, of molecular weights of approximately 85,000 and 26,000, respectively, each accounting for about 10% of the total cell protein. Their amino acid compositions showed them to have distinct polypeptide chains. After osmotic lysis of rho-form cells the organelles disappeared; protein A accompanied the membrane fraction, whereas protein B was partly released in soluble form. After lysis by Nonidet P-40 in a medium composed of 4 M glycerol, 50 mM phosphate, and 10 mM MgSO4 at pH 6 (GPM-6), the organelles were preserved and released with ultrastructure unchanged. Protein A was recovered in the soluble fraction and protein B in the particulate (crude fiber) fraction. Treatment of the crude fiber fraction with 0.5 M NaCl in GPM-6 or with a solution containing 4 M glycerol, 10 mM morpholinoethanesulfonate, and 1 mM ethylenediaminetetraacetate at pH 7.0 caused the fibers to disassemble into subunits. By subsequent changes in the ionic conditions and temperature it was possible to cause the subunits to reassemble into ordered aggregates having the same ultrastructure as the native rho-fibers. The optimum temperature for reassembly in the presence of 4 M glycerol was 37 C, the optimum pH was 6.5 to 7.0, and the presence of Mg-2+, replaceable by Ca-2+, SR-2+, or Ba-2+, was essential. Protein B was the only protein detected in the purified, reconsituted fibers.     

Effects of solutes on solubilization and refolding of proteins from inclusion bodies with high hydrostatic pressure

High hydrostatic pressure (HHP)-mediated solubilization and refolding of five inclusion bodies (IBs) produced from bacteria, three gram-negative binding proteins (GNBP1, GNBP2, and GNBP3) from Drosophila, and two phosphatases from human were investigated in combination of a redox-shuffling agent (2 mM DTT and 6 mM GSSG) and various additives. HHP (200 MPa) combined with the redox-shuffling agent resulted in solubilization yields of approximately 42%-58% from 1 mg/mL of IBs. Addition of urea (1 and 2 M), 2.5 M glycerol, L-arginine (0.5 M), Tween 20 (0.1 mM), or Triton X-100 (0.5 mM) significantly enhanced the solubilization yield for all proteins. However, urea, glycerol, and nonionic surfactants populated more soluble oligomeric species than monomeric species, whereas arginine dominantly induced functional monomeric species (approximately 70%-100%) to achieve refolding yields of approximately 55%-78% from IBs (1 mg/mL). Our results suggest that the combination of HHP with arginine is most effective in enhancing the refolding yield by preventing aggregation of partially folded intermediates populated during the refolding. Using the refolded proteins, the binding specificity of GNBP2 and GNBP3 was newly identified the same as with that of GNBP1, and the enzymatic activities of the two phosphatases facilitates their further characterization.     

Isolation of alpha-connectin, an elastic protein, from rabbit skeletal muscle

alpha-Connectin (also called titin 1) has been isolated from rabbit back muscle. Myofibrils were well washed with 5 mM NaHCO3 and then extracted with 0.2 M sodium phosphate, pH 7.0. The extract was dialyzed against 0.1 M potassium phosphate, pH 7.0, to sediment myosin. The supernatant, adjusted to 0.18 M potassium phosphate, pH 7.0, and 4 M urea, was subjected to DEAE Toyopearl column chromatography. beta-Connectin was eluted in the flow-through fraction and alpha-connectin was eluted at around 0.1 M NaCl, when a 0 to 0.25 M NaCl gradient was applied. The separated alpha-connectin was dialyzed against 0.2 M potassium phosphate, pH 7.0. The resultant alpha-connectin showed the same mobility as that in an SDS extract of rabbit back muscle on SDS gel electrophoresis using 1.8% polyacrylamide gels. A monoclonal antibody against chicken breast muscle beta-connectin reacted with the alpha-connectin isolated from rabbit back muscle.     

An efficient refolding method for the preparation of recombinant human prethrombin-2 and characterization of the recombinant-derived alpha-thrombin

Human recombinant prethrombin-2 was produced in Escherichia coli. The expressed prethrombin-2 formed intracellular inclusion bodies from which the protein was refolded by a simple one-step dilution process in buffer consisting of 50 mM Tris-HCl, containing 20 mM CaCl(2), 500 mM NaCl, 1 mM EDTA, 600 mM arginine, 1 mM cysteine, 0.1 mM cystine, 10% (v/v) glycerol, and 0.2% (w/v) Brij-58 at pH 8.5. After refolding, prethrombin-2 was purified by hirudin-based COOH-terminal peptide affinity chromatography, and then activated with Echis carinatus snake venom prothrombin activator (ecarin). The activated protein, alpha-thrombin, was then tested for several activities including activity toward chromogenic substrate, release of fibrinopeptide A from fibrinogen, activation of protein C, and thrombin-activatable fibrinolysis inhibitor, reactivity with antithrombin, clotting activity, and platelet aggregation. The kinetic data showed no differences in activity between our recombinant alpha-thrombin and plasma-derived alpha-thrombin. The yield of refolded recombinant human prethrombin-2 was about 4-7% of the starting amount of solubilized protein. In addition, the final yield of purified refolded protein was 0.5-1%, and about 1 mg of recombinant prethrombin-2 could be isolated from 1 liter of E. coli cell culture.     

Production of recombinant human serum albumin from Saccharomyces cerevisiae

Human serum albumin has been constitutively expressed in a Saccharomyces cerevisiae brewing yeast. After cell growth and disruption the product was associated with the insoluble fraction and represented approximately 1% of total cell protein. After the cell debris was extensively washed, the albumin was solubilized with 8 M urea and 28 mM 2-mercaptoethanol in 50 mM sodium carbonate buffer, pH 10. The denatured albumin was refolded by dialysis and further purified by anion exchange and gel filtration chromatography. Losses of renatured material could be reduced, or higher protein concentrations used during refolding, if the denatured product was purified by cation-exchange chromatography in urea prior to refolding. Apart from an additional N-terminal N-acetyl methionine, the refolded product proved identical to human serum albumin derived from plasma when compared by a variety of physical, chemical, and biological analytical methods.     

On-column refolding of an insoluble histidine tag recombinant exopolyphosphatase from Trypanosoma brucei overexpressed in Escherichia coli

An exopolyphosphatase gene has been cloned by polymerase chain reaction (PCR) from Trypanosoma brucei and the corresponding protein overexpressed as a recombinant His-tag (histidine tag) exopolyphosphatase in E. coli in order to characterize its biochemical activity and to produce antibody to determine its localization. Because overexpression of this protein in bacteria resulted in the formation of inactive inclusion bodies, these structures were first solubilized in denaturant condition (6 M urea). Secondly, after a capture step using immobilized metal affinity chromatography (IMAC), a gradual refolding of the protein was performed on-column from 6 M to 0 M urea in the presence of 1% Triton X-100. Triton X-100 was used to abolish protein aggregation during the refolding step. The purified enzyme was active, demonstrating that at least part of the proteins was properly refolded.     

Overexpression, purification and characterization of the catalytic component of Oplophorus luciferase in the deep-sea shrimp, Oplophorus gracilirostris

The luciferase secreted by the deep-sea shrimp Oplophorus consists of 19 and 35kDa proteins. The 19-kDa protein (19kOLase), the catalytic component of luminescence reaction, was expressed in Escherichia coli using the cold-shock inducted expression system. 19kOLase, expressed as inclusion bodies, was solubilized with 6M urea and purified by urea-nickel chelate affinity chromatography. The yield of 19kOLase was 16 mg from 400 ml of cultured cells. 19kOLase in 6M urea could be refolded rapidly by dilution with 50mM Tris-HCl (pH 7.8)-10mM EDTA, and the refolded protein showed luminescence activity. The luminescence properties of refolded 19kOLase were characterized, in comparison with native Oplophorus luciferase. Luminescence intensity with bisdeoxycoelenterazine as a substrate was stimulated in the presence of organic solvents. The 19kOLase is a thermolabile protein and is 98 % inhibited by 1muM Cu2+. The cysteine residue of 19kOLase is not essential for catalysis of the luminescence reaction.     

Soybean disease resistance protein RHG1-LRR domain expressed, purified and refolded from Escherichia coli inclusion bodies: preparation for a functional analysis

Introduction and expression of foreign genes in bacteria often results accumulation of the foreign protein(s) in inclusion bodies (IBs). The subsequent processes of refolding are slow, difficult and often fail to yield significant amounts of folded protein. RHG1 encoded by rhg1 was a soybean (Glycine max L. Merr.) transmembrane receptor-like kinase (EC 2.7.11.1) with an extracellular leucine-rich repeat domain. The LRR of RHG1 was believed to be involved in elicitor recognition and interaction with other plant proteins. The aim, here, was to express the LRR domain in Escherichia coli (RHG1-LRR) and produce refolded protein. Urea titration experiments showed that the IBs formed in E. coli by the extracellular domain of the RHG1 protein could be solubilized at different urea concentrations. The RHG1 proteins were eluted with 1.0-7.0M urea in 0.5M increments. Purified RHG1 protein obtained from the 1.5 and 7.0M elutions was analyzed for secondary structure through circular dichroism (CD) spectroscopy. Considerable secondary structure could be seen in the former, whereas the latter yielded CD curves characteristic of denatured proteins. Both elutions were subjected to refolding by slowly removing urea in the presence of arginine and reduced/oxidized glutathione. Detectable amounts of refolded protein could not be recovered from the 7.0M urea sample, whereas refolding from the 1.5M urea sample yielded 0.2mg/ml protein. The 7.0M treatment resulted in the formation of a homogenous denatured state with no apparent secondary structure. Refolding from this fully denatured state may confer kinetic and/or thermodynamic constraints on the refolding process, whereas the kinetic and/or thermodynamic barriers to attain the folded conformation appeared to be lesser, when refolding from a partially folded state.     

Heparin-copper biaffinity chromatography of fibroblast growth factors

A novel method is described to separate and identify the various forms of fibroblast growth factor (FGF) based on their differential affinities for both heparin and copper. FGFs were extracted from bovine hypothalamus and purified by batchwise adsorption to heparin-Sepharose. The partially purified FGFs were then applied to an affinity column prepared by mixing equal portions of heparin-Sepharose and copper-Sepharose. The column was rinsed consecutively with the following four reagents: (i) 2 M NaCl, (ii) 0.6 M NaCl, (iii) 0.6 M NaCl plus 10 mM imidazole, and (iv) 0.6 m NaCl. FGFs were then eluted with a linear NaCl/imidazole gradient (from 0.6 m NaCl without imidazole to 2 M NaCl plus 10 mM imidazole). Fractions eluted from the column were analyzed by sodium dodecyl sulfate-gel electrophoresis with silver staining and electrophoretic immunoblot using site-specific antibodies against basic and acidic FGF. The results demonstrate that it is possible to resolve from hypothalamus at least two basic FGF species (with Mr values of 19,000 and 18,000) and three acidic FGF species (with Mr values of 18,000, 16,400, and 15,600). These findings indicate that heparin-copper biaffinity chromatography may have wide applicability in the study of the structure and activity of FGFs.     

Effect of hyperosmotic solutions on salt excretion and thirst in rats

We investigated urinary changes and thirst induced by infusion of hyperosmotic solutions in freely moving rats. Intracarotid infusions of 0.3 M NaCl (4 ml/20 min, split between both internal carotid arteries) caused a larger increase in excretion of Na(+) and K(+) than intravenous infusions, indicating that cephalic sensors were involved in the response to intracarotid infusions. Intravenous and intracarotid infusions of hyperosmotic glycerol or urea (300 mM in 150 mM NaCl) had little or no effect, suggesting the sensors were outside the blood-brain barrier (BBB). Intracarotid infusion of hypertonic mannitol (300 mM in 150 mM NaCl) was more effective than intravenous infusion, suggesting that cell volume rather than Na(+) concentration of the blood was critical. Similarly, intracarotid infusion (2 ml/20 min, split between both sides), but not intravenous infusion of hypertonic NaCl or mannitol caused thirst. Hyperosmotic glycerol, infused intravenously or into the carotid arteries, did not cause thirst. We conclude that both thirst and electrolyte excretion depend on a cell volume sensor that is located in the head, but outside the BBB.     

Functional recombinant rabbit muscle phosphoglucomutase from Escherichia coli

The gene coding for phosphoglucomutase (PGM) from Oryctolagus cuniculus (rabbit) has been expressed in Escherichia coli under a T7 expression system with a His-tag. About half of the expressed PGM protein was present in inclusion bodies, but this protein was inactive when solubilized. The protein in the soluble cell fraction was isolated and purified in one step on a Ni-NTA column. The eluate from this column was adjusted to 95% saturated ammonium sulfate, and the resulting protein precipitate was resuspended in sodium phosphate buffer and dialyzed against 2.5 M ammonium sulfate. The final yield of protein was about 10 mg per liter of LB medium. The protein was judged to be greater than 90% pure on the basis of gel electrophoresis and activity measurements (128 U per milligram). Our motivation for developing this bacterial production system for PGM has been to prepare sufficient quantities of stable-isotope-labeled protein for experiments that utilize recently developed NMR technologies suitable for proteins the size of PGM (61.6 kDa). Preliminary NMR studies indicate that the current level of purity is adequate for this work. The construct described here was designed to incorporate an N-terminal His-tag for ease of isolation. Although PGM is a metalloprotein, the His-tag does not appear to interfere with activity. The presence of the His-tag should not pose a problem for proposed (31)P NMR investigations of the protein and its complexes in aqueous solution or incorporated into reverse micelles. However, we plan to design a cleavable His-tag for later (1)H, (13)C, (15)N studies of the active site, which includes essential histidine residues.     

Conformational stability of 17 beta-hydroxysteroid dehydrogenase from the fungus Cochliobolus lunatus

The functional activities of proteins are closely related to their molecular structure and understanding their structure-function relationships remains one of the intriguing problems of molecular biology. We investigated structural changes in 17beta-hydroxysteroid dehydrogenase from the fungus Cochliobolus lunatus (17beta-HSDcl) induced by pH, temperature, salt, urea, guanidine hydrochloride, and coenzyme NADPH binding. At 25 degrees C and within the relatively narrow pH range of 7.0-9.0, 17beta-HSDcl exists in its native conformation as a dimer. This native conformation is thermally stable up to 40 degrees C in this pH range. At 25 degrees C and pH 2.0 in the presence of 150-300 mM NaCl, 17beta-HSDcl forms soluble aggregates enriched in alpha-helical and beta-sheet structures. At higher temperatures and NaCl concentrations, these soluble aggregates start to precipitate. The denaturants urea and guanidine hydrochloride unfold 17beta-HSDcl at concentrations of 1.2 and 0.4 M, respectively. Binding of the coenzyme NADPH to 17beta-HSDcl causes local structural changes that do not significantly affect the thermal stability of this protein.     

Expression of Torpedo californica creatine kinase in Escherichia coli and purification from inclusion bodies

The pET17 expression vector was used to express creatine kinase from the electric organ of Torpedo californica as inclusion bodies in Escherichia coli BL21(DE3) cells. The insoluble aggregate was dissolved in 8M urea and, following extraction with Triton X-100, the enzyme was refolded by dialysis against Tris buffer (pH 8.0) containing 0.2M NaCl. After two buffer changes, chromatography on Blue Sepharose was used as a final step in the purification procedure. Approximately 54mg active protein was recovered from a 1L culture and the refolded enzyme had a specific activity of 75U/mg. The molecular mass of the purified protein was consistent with that predicted from the amino acid sequence and the CD spectrum of the refolded enzyme was essentially identical to that of creatine kinase from human muscle (HMCK). The K(m) values of ATP and ADP were also similar to those of HMCK, while the K(m) values for both phosphocreatine and creatine were approximately 5-10-fold higher. The purification described here is in marked contrast with earlier attempts at purification of this isozyme where, in a process yielding less than 1mg/L culture, enzyme with a specific activity of ca. 5U/mg was obtained.     

In vitro splicing of erythropoietin by the Mycobacterium tuberculosis RecA intein without substituting amino acids at the splice junctions

Protein splicing is a self-catalyzed process involving the excision of an intervening polypeptide sequence, the intein, and joining of the flanking polypeptide sequences, the extein, by a peptide bond. We have studied the in vitro splicing of erythropoietin (EPO) using a truncated form of the Mycobacterium tuberculosis RecA mini-intein in which the homing endonuclease domain was replaced with a hexahistidine sequence (His-tag). The intein was inserted adjacent to cysteine residues to assure that the spliced product had the natural amino acid sequence. When expressed in Escherichia coli, intein-containing EPO was found entirely as inclusion bodies but could be refolded in soluble form in the presence of 0.5 M arginine. Protein splicing of the refolded protein could be induced with a reducing agent such as DTT or tris(2-carboxyethyl)phosphine and led to the formation of EPO and mini-intein along with some cleavage products. Protein splicing mediated by the RecA intein requires the presence of a cysteine residue adjacent to the intein insertion site. We compared the efficiencies of protein splicing adjacent to three of the four cysteine residues of EPO (Cys29, Cys33 and Cys161) and found that insertion of intein adjacent to Cys29 allowed far more efficient protein splicing than insertion adjacent to Cys33 or Cys161. For ease of purification, our experiments involved a His-tagged EPO fusion protein and a His-tagged intein and the spliced products (25 kDa EPO and 24 kDa mini-intein) were identified by Western blotting using anti-EPO and anti-His-tag antibodies and by mass spectroscopy. The optimal splicing yield at Cys29 (40%) occurred at pH 7.0 after refolding at 4 degrees C and splicing for 18 h at 25 degrees C in the presence of 1 mM DTT.     

Purification, refolding, and characterization of recombinant LHRH-T multimer

To make the native LHRH immunogenic, a multimer of LHRH interspersed with T non-B peptides (r-LHRH-d2) was expressed as recombinant protein in Escherichia coli. The expression level of the recombinant protein was around 15% of the total cellular protein and it aggregated as inclusion bodies. Inclusion bodies from the bacterial cells were isolated and purified to homogeneity. Instead of high concentrations of chaotropic agents, r-LHRH- d2 was solubilized in 50 mM citrate buffer at pH 3 containing 2 M urea. The protein was refolded by 5-fold dilution (pulsatile) with cold 10 mM citrate buffer at pH 6 in presence of 0.3 M L-arginine. Purification of r-LHRH-d2 was carried out by successive passages on CM-Sepharose column at pH 6.0 which retained extraneous proteins and pH 4.8 at which r-LHRH-d2 bound to the resin. The elution was carried out by using linear salt gradient (0.1-1 M NaCl). The overall yield of the purified r-LHRH-d2 was 40% of the initial inclusion body proteins. The purity and homogeneity were confirmed by a single homogeneous peak on analytical HPLC eluting out at 29.51 min and by single band on SDS-PAGE reactive with polyvalent anti-LHRH antibodies. Mass spectroscopic analysis indicated the protein to be of 16.6 kDa which equals the theoretically expected mass. The N-terminal amino acid analysis of r-LHRH-d2 showed the sequence which corresponded to the designed protein. The CD spectrum of the refolded r-LHRH-d2 showed that the multimer has considerable beta sheet structure like the monomeric LHRH protein.     

Cloning, expression, and refolding of a secretory protein ESAT-6 of Mycobacterium tuberculosis

A DNA encoding the 6-kDa early secretory antigenic target (ESAT-6) of Mycobacterium tuberculosis was inserted into a bacterial expression vector of pQE30 resulting in a 6x His-esat-6 fusion gene construction. This plasmid was transformed into Escherichia coli strain M15 and effectively expressed. The expressed fusion protein was found almost entirely in the insoluble form (inclusion bodies) in cell lysate. The inclusion bodies were solubilized with 8M urea or 6M guanidine-hydrochloride at pH 7.4, and the recombinant protein was purified by Ni-NTA column. The purified fusion protein was refolded by dialysis with a gradient of decreasing concentration of urea or guanidine hydrochloride or by the size exclusion protein refolding system. The yield of refolded protein obtained from urea dialysis was 20 times higher than that from guanidine-hydrochloride. Sixty-six percent of recombinant ESAT-6 was successfully refolded as monomer protein by urea gradient dialysis, while 69% of recombinant ESAT-6 was successfully refolded as monomer protein by using Sephadex G-200 size exclusion column. These results indicate that urea is more suitable than guanidine-hydrochloride in extracting and refolding the protein. Between the urea gradient dialysis and the size exclusion protein refolding system, the yield of the monomer protein was almost the same, but the size exclusion protein refolding system needs less time and reagents.