Native-like intermediate on the folding pathway of Escherichia coli succinyl-CoA synthetase

The transition between the native and denatured states of the tetrameric succinyl-CoA synthetase from Escherichia coli has been investigated by circular dichroism, fluorescence spectroscopy, cross-linking by glutaraldehyde and activity measurements. At pH 7.4 and 25 degrees C, both denaturation of succinyl-CoA synthetase by guanidine hydrochloride and refolding of the denatured enzyme have been characterized as reversible reactions. In the presence of its substrate ATP, the denatured enzyme could be successfully reconstituted into the active enzyme with a yield of 71-100%. Kinetically, reacquisition of secondary structure by the denatured enzyme was rapid and occurred within 1 min after refolding was initiated. On the other hand, its reactivation was a slow process which continued up to 25 min before 90% of the native activity could be restored. Both secondary and quaternary structures of the enzyme, reconstituted in the absence of ATP, were indistinguishable from those of the native enzyme but the renatured protein was catalytically inactive. This observation indicates the presence of catalytically inactive tetramer as an intermediate in the reconstitution process. The reconstituted protein could be reactivated by ATP even 10 min after the reacquisition of the native secondary structure by the refolding protein. However, reactivation of the protein by ATP 60 min after the regain of secondary structure was significantly less, suggesting that rapid refolding and reassociation of the monomers into a native-like tetramer and reactivation of the tetramer are sequential events; the latter involving slow and small conformational rearrangements in the refolded enzyme that are likely to be associated with phosphorylation.     

Phospholipid assisted folding of a denatured heme protein: effect of phosphatidylethanolamine

The role of the aminophospholipid, phosphatidylethanolamine (PE), has been well established to act as a non-protein molecular chaperone in the folding and assembly of polytopic membrane proteins. However, such studies with soluble proteins have not been done so far and in particular with the heme proteins. We have used the heme enzyme, horseradish peroxidase (HRP), as the model heme protein and studied the effect of different phospholipids on its refolding from denatured state. Dimyristoylphosphatidylethanolamine (DMPE), a bilayer-forming PE, was able to increase the reactivation yield of denatured HRP upon 30min refolding at 25 degrees C. However, dioleoylphosphatidylethanolamine (DOPE), containing one double bond in the fatty acid chains, which does not favour bilayer organization, did not support proper refolding. The phospholipids with N-methylated head groups, phosphatidylcholines, e.g., DMPC and DOPC showed differential effects when DMPC remained mostly non-supportive while DOPC on the contrary led to inhibition of the refolding of the denatured heme enzyme. Fluorescence spectroscopic studies also indicated changes in the microenvironments of the heme moiety and the single tryptophan residue of HRP in presence of the aminophospholipid.     

Aggregation of creatine kinase during refolding and chaperonin-mediated folding of creatine kinase

The course of refolding and reactivation of urea-denatured creatine kinase (ATP; creatine N-phosphotransferase, EC 2.7.3.2) has been studied in the absence and presence of molecular chaperonin GroEL. The enzyme was denatured in Tris--HCl buffer containing 6 M urea for 1 h. In the refolding studies, the denatured enzyme was diluted 60-fold into the same buffer containing GroEL or not for activity, turbidity, fluorescence measurements and polyacrylamide gel electrophoresis. The results show that the reactivation process is dependent of creatine kinase concentration in the concentration range 2.5--4 microM. The levels of activity recovery decrease with increasing enzyme concentration because of the formation of wrong aggregates. The molecular chaperonin GroEL can bind the refolding intermediate of creatine kinase and thus prevent the formation of wrong aggregates. This intermediate is an inactive dimeric form that is in a conformation resembling the 'molten globule' state.     

Further characterization of the reassembly of creatine kinase and effect of substrate

Upon exposure to 8 M urea, creatine kinase from rabbit muscle exhibited a rapid increase in intrinsic fluorescence and a rapid decrease in fluorescence polarization. Polarization changes were complete after 5 min, while fluorescence changes continued for at least 15 min. Fluorescence polarization changes accompanying reassembly were complex, and appeared to involve a concentration dependent reaction. Enzyme sampled at intervals during denaturation exhibited refolding kinetics displaying two first-order rate constants, the first dependent and the second independent of the duration of exposure to urea. There was evidence for an additional renaturation step, occurring within the mixing phase of the denatured protein with solvent. Reactivation kinetics and yield of reactivated enzyme exhibited a dependency upon length of exposure to denaturant. The exposure of renaturing creatine kinase to trypsin was shown to prevent further reactivation, and provided use of a method to determine reactivation rates at discrete intervals after initiation of reassembly. The presence of 2 mM MgADP during reactivation enhanced the rate of reactivation immediately after initiation of reactivation. Reactivation was not accelerated if nucleotide substrate was added after reactivation was initiated nor did nucleotide substrate increase the overall reactivation yield. The presence of MgADP also enhanced the rate of refolding at an early stage as judged by changes in intrinsic fluorescence and resistance to tryptic hydrolysis. While in addition to MgADP, creatine phosphate accelerated resistance by refolding creatine kinase to trypsin, according to the other criteria measured, the phosphagen substrates did not promote reactivation or renaturation. The unfolding-refolding studies and role of substrate in reassembly were consistent with a mechanism involving at least two steps, possibly involving cis-trans isomerization of proline. These data also supported the suggestion that the formation of the nucleotide binding region is an early event in the refolding of creatine kinase in vitro.     

Cis-trans isomerization is rate-determining in the reactivation of denatured human carbonic anhydrase II as evidenced by proline isomerase.

The refolding of human carbonic anhydrase II is a sequential process. The slowest step involved is the recovery of enzymic activity (t1/2 = 9 min). Kinetic data from 'double-jump' measurements indicate that proline isomerization might be rate determining in the reactivation of the denatured enzyme. Proof of this is provided by the effect of proline isomerase on the reactivation kinetics: the presence of isomerase during reactivation lowers the half-time of the reaction to 4 min, and inhibition of proline isomerase completely abolishes this kinetic effect. A similar acceleration of the refolding process by proline isomerase is also observed for bovine carbonic anhydrase II, in contrast to what has previously been reported. In human carbonic anhydrase II there are two cis-peptidyl-Pro bonds at Pro30 and Pro202. Two asparagine single mutants (P30N and P202N) and a glycine double mutant (P30G/P202G) were constructed to investigate the role of these prolines in the rate limitation of the reactivation process. Both in the presence and absence of PPIase the P202N mutant behaved exactly like the unmutated enzyme. Thus, cis-trans isomerization of the Pro202 cis-peptidyl bond is not rate determining in the reactivation process. The mutations at position 30 led to such extensive destabilization of the protein that the refolding reaction could not be studied.     

Polyols induce ATP-independent folding of GroEL-bound bacterial glutamine synthetase.

We have previously assessed the GroE chaperonin requirements for folding of bacterial glutamine synthetase (GS) and established that, at 37 degrees C in 50 mM Tris buffer, ATP binding to the GroEL-GS complex is mandatory for the release and reactivation of dodecameric enzyme. However, we demonstrate here that the addition of 1-4 M glycerol to GroEL-GS complexes resulted in release and reactivation of GS in the absence of nucleotide. Furthermore, the kinetics of refolding and refolding yields of this glycerol-induced refolding were similar to those observed with ATP. Other polyols such as sucrose, 1,2-propanediol, or 1,3-propanediol also facilitated nucleotide-independent refolding of GS from chaperonin complex. The observed phenomenon cannot be attributed to the viscosity or molecular crowding effects because solutions of dextran or Ficoll with the same viscosity as 4 M glycerol failed to reactivate GroEL-bound GS. Like glycerol, other osmolytes such as betaine and sarcosine or high salt (500 mM NaCl) facilitated spontaneous folding of GS. However, no reactivation of GroEL-bound GS was observed with these additives. The presence of glycerol affected binding of fluorescent probe 1,8-anilinonaphthalene to GroEL, suggesting that glycerol may alter the chaperonin structure. Our data suggest that low-molecular-weight polyols affect both GroEL and bound GS monomers to reduce their binding affinity. This results in an increased partitioning of GS toward active, assembly-competent states.     

Refolding and reactivation of calf intestinal alkaline phosphatase with excess magnesium ions

It is well known that Mg(2+) is an essential component in many biological processes. This research investigated the courses of both the reactivation and the refolding in the absence and presence of Mg(2+) ions. Calf intestinal alkaline phosphatase (CIP) was extensively denatured in 3 M guanidine hydrochloride (GdnHCl) solution for 2 h. Under suitable renaturation conditions, about 60-70% of the activity was recovered in the absence and presence of different magnesium ion concentrations. The refolding processes followed two-phase courses, whereas the reactivation processes were monophasic after dilution in proper solutions with or without Mg(2+). The magnesium ions affected both the reactivation and the refolding courses of unfolded CIP. A comparison of rate constants for the refolding of unfolded CIP with those for recovery of enzyme activity at different Mg(2+) concentrations showed that they were not synchronized. The activity recovery was speeded up due to the presence of Mg(2+) ions; while the refolding course of unfolded CIP was somewhat inhibited by the excess Mg(2+).     

Artificial chaperone-assisted refolding of chemically denatured alpha-amylase

It is now well established that alpha-cyclodextrin (alpha-CD) is a valuable folding agent in refolding processes of several denatured enzyme solutions. The refolding of Gu-HCl denatured alpha-amylase in the dilution-additive mode revealed that alpha-CD enhanced the refolding yield by 20-30% depending upon alpha-CD concentration. However, the refolding efficiency of the Gu-HCl denatured alpha-amylase through the artificial chaperone-assisted method indicated that alpha-CD enhanced the activity recovery of denatured alpha-amylase by almost 50% and also increased the reactivation rate constant relative to the unassisted control sample. The higher refolding efficiency should be due to different mechanism played by alpha-CD in this technique. In addition, our data indicated that higher refolding yields are obtained when the residual Gu-HCl concentration is low in the refolding environment and when the capture agent is removed not in a stepwise manner from the protein-detergent complexes in the stripping step of the whole process. Collectively, the results of this investigation expand the range of procedural variations used to refold different denatured proteins through artificial chaperone-assisted method.     

Monoclonal antibodies assisting refolding of firefly luciferase

The reactivation efficiency in the refolding of denatured luciferase in the presence and the absence of monoclonal antibodies (mAbs) has been studied. Luciferase could be partially reactivated when the protein was denatured in high concentrations of guanidium chloride (GdmCl; >4.5 M) and the refolding was carried out in very low protein concentrations. The refolding yield was, however, significantly lower when it was performed on luciferase that had been denatured with lower concentrations of GdmCl. The efficiency of refolding decreases when the formation of aggregates increases. Three of the five luciferase mAbs tested (4G3, N2E3, S2G10) dramatically increased the yield of reactivation and simultaneously eliminated the formation of aggregates. It is proposed that these mAbs assisted the refolding of luciferase by binding to the exposed hydrophobic surface of the refolding intermediate, thus preventing it from aggregating. The epitopes interacting with these refolding-assisting mAbs are all located in the A-subdomain of the N-terminal region of luciferase. These results have also shed light on the structural features of the intermediate and its interface involved in protein aggregate formation, contributing to the understanding of the protein folding mechanism.     

Changes at the N-terminus of human brain creatine kinase during a transition between inactive folding intermediate and active enzyme.

CK-STAR, a monoclonal antibody against human brain creatine kinase (CK), can be shown by chemical cleavage mapping and peptide synthesis to recognize an epitope at the free N-terminus of the enzyme. The epitope could be largely reproduced by a synthetic peptide based on the first 18 amino acids and could be partly formed by the first 11 amino acids. The antibody did not bind to native CK, but it did bind to CK in various partially denatured forms and to an enzymically inactive intermediate in the refolding process. Competitive binding studies have shown that the N-terminal conformations of both the refolding intermediate and the free peptide resemble that of CK partially denatured by attachment to plastic. The results suggest that the final stages of CK refolding and reactivation involve a structural change at the N-terminus or its interaction with some other part of the CK molecule, thus masking the CK-STAR epitope.     

Chaperone activity and prodan binding at the self-associating domain of erythroid spectrin

Spectrin, the major constituent protein of the erythrocyte membrane skeleton, exhibits chaperone activity by preventing the irreversible aggregation of insulin at 25 degrees C and that of alcohol dehydrogenase at 50 degrees C. The dimeric spectrin and the two subunits, alpha-spectrin and beta-spectrin prevent such aggregation appreciably better, 70% in presence of dimeric spectrin at an insulin:spectrin ratio of 1:1, than that in presence of the tetramer of 25%. Our results also show that spectrin binds to denatured enzymes alpha-glucosidase and alkaline phosphatase during refolding and the reactivation yields are increased in the presence of the spectrin derivatives when compared with those refolded in their absence. The unique hydrophobic binding site on spectrin for the fluorescence probe, 6-propionyl-2-(dimethylamino)naphthalene (Prodan) has been established to localize at the self-associating domain with the binding stoichiometry of one Prodan/both dimeric and tetrameric spectrin. The other fluorescence probe, 1-anilinonaphthalene-8-sulfonic acid, does not show such specificity for spectrin, and the binding stoichiometry is between 3 and 5 1-anilinonaphthalene-8-sulfonic acid/dimeric and tetrameric spectrin, respectively. Regions in alpha- and beta-spectrins have been found to have sequence homology with known chaperone proteins. More than 50% similarities in alpha-spectrin near the N terminus with human Hsp90 and in beta-spectrin near the C terminus with human Hsp90 and Escherichia coli DnaJ have been found, indicating a potential chaperone-like sequence to be present near the self-associating domain that is formed by portions of alpha-spectrin near the N terminus and the beta-spectrin near the C terminus. There are other patches of sequences also in both the spectrin polypeptides, at the other termini as well as in the middle of the rod domain having significant homology with well known chaperone proteins.     

A stable cold folding intermediate of rabbit muscle D-glyceraldehyde 3-phosphate dehydrogenase.

With decreasing temperature the reactivation yield of denatured D-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) upon dilution increases but the reactivation rate decreases. Neither reactivation nor aggregation during refolding can be detected at 4 degrees C in 48 h, and at 3 degrees C even in 6 days. However, the reactivation takes place once the temperature is raised with little decrease of the yield after incubation for 6 days at 3 degrees C. A cold folding intermediate forms in a burst phase of refolding at 4 degrees C as shown by a fast change of the intrinsic fluorescence followed by further conformational adjustment to a stable state in about 1 h. The stable folding intermediate has been characterized to be a dimer of partially folded GAPDH subunit with secondary structure between that of the native and denatured enzymes, a hydrophobic cluster not found in either the native or the denatured state, and an active site similar to but different from that of the native state. Chaperonin 60 (GroEL) binds with all intermediates formed at 4 degrees C, but the intermediates formed at the early folding stage reactivate with higher yield than those formed after conformational adjustment when dissociated from GroEL in the presence of ATP and further folded and assembled into the native tetramer.     

Stabilization of heat-induced changes in plant peroxidase preparations by ClpX, a bacterial heat shock protein

Peroxidases (PODs) are known to be quite stable at elevated temperatures. Moreover, partially denatured peroxidases are able to regain their catalytic activity during incubation at room temperature. In this paper, we describe the effects of some heat shock proteins on the self-reactivation of plant peroxidase preparations. Horseradish and artichoke peroxidases (HRP and ARP, respectively) were first heated (at 60 °C or 90 °C), then incubated at a slightly elevated temperature (30 °C). The heat-treatment resulted in a considerable loss of activity of both enzymes but the subsequent incubation allowed their reactivation. However, no reactivation could be detected when incubation was carried out in the presence of the molecular chaperone ClpX. Other chaperones that were tested (DnaK, DnaJ and GrpE) did not show the inhibitory effect. Electrophoretic analyses further indicated that the heat-treated horseradish peroxidase, but not the native enzyme, binds to ClpX eliminating the possibility of undesirable protein refolding that would result in aggregation.     

ClpB cooperates with DnaK, DnaJ, and GrpE in suppressing protein aggregation. A novel multi-chaperone system from Escherichia coli.

ClpB is a heat-shock protein from Escherichia coli with an unknown function. We studied a possible molecular chaperone activity of ClpB in vitro. Firefly luciferase was denatured in urea and then diluted into the refolding buffer (in the presence of 5 mM ATP and 0.1 mg/ml bovine serum albumin). Spontaneous reactivation of luciferase was very weak (less than 0.02% of the native activity) because of extensive aggregation. Conventional chaperone systems (GroEL/GroES and DnaK/DnaJ/GrpE) or ClpB alone did not reactivate luciferase under those conditions. However, ClpB together with DnaK/DnaJ/GrpE greatly enhanced the luciferase activity regain (up to 57% of native activity) by suppressing luciferase aggregation. This coordinated function of ClpB and DnaK/DnaJ/GrpE required ATP hydrolysis, although the ClpB ATPase was not activated by native or denatured luciferase. When the chaperones were added to the luciferase refolding solutions after 5-25 min of refolding, ClpB and DnaK/DnaJ/GrpE recovered the luciferase activity from preformed aggregates. Thus, we have identified a novel multi-chaperone system from E. coli, which is analogous to the Hsp104/Ssa1/Ydj1 system from yeast. ClpB is the only known bacterial Hsp100 protein capable of cooperating with other heat-shock proteins in suppressing and reversing protein aggregation.     

Both the isomerase and chaperone activities of protein disulfide isomerase are required for the reactivation of reduced and denatured acidic phospholipase A2.

The spontaneous reactivation yield of acidic phospholipase A2 (APLA2), a protein containing seven disulfide bonds, after reduction and denaturation in guanidine hydrochloride is very low. Protein disulfide isomerase (PDI) markedly increases the reactivation yield and prevents the aggregation of APLA2 during refolding in a redox buffer containing GSH and GSSG. S-methylated PDI (mPDI), with no isomerase but as nearly full chaperone activity as native PDI, has no effect on either the reactivation or aggregation of APLA2. However, the simultaneous presence of PDI and mPDI in molar ratios to APLA2 of 0.1 and 0.9 respectively fully reactivates the denatured enzyme, as does PDI alone at a ratio of 1. At ratios of 0.1 and 0.15 respectively, they completely suppress APLA2 aggregation, as does PDI alone at a ratio of 0.25. Moreover, delayed addition of PDI to the refolding buffer greatly diminished the reactivation yield of APLA2, but this deteriorating effect can be alleviated markedly by the presence of mPDI in the refolding buffer. Without GSSG, mPDI prevents the aggregation of APLA2 during refolding. It is proposed that the in vitro action of PDI as a foldase consists of both isomerase and chaperone activities, and the latter activity can be fully replaced by mPDI.     

Effects of the chaperonin GroE on the refolding of tryptophanase from Escherichia coli. Refolding is enhanced in the presence of ADP.

The refolding of the tetrameric enzyme tryptophanase was facilitated by the chaperonin GroE. Maximum refolding yield of tryptophanase molecules (about 80%) was attained in the presence of a 15-fold excess of GroE 21-mer over tryptophanase monomer. The GroEL subunit was required for this improvement in refolding yield, whereas the GroES subunit was not. Light scattering experiments of the refolding reaction revealed that GroE bound to tryptophanase folding intermediates and suppressed their aggregation. The presence of ATP was required for the efficient dissociation of tryptophanase from GroEL. However, our experiments indicated that tryptophanase dissociated readily from GroEL in the presence of not only ATP, but also in the presence of non-hydrolyzable ATP analogues such as ATP gamma S (adenosine 5'-O-(3-thiotriphosphate)) and AMP-PNP (adenyl-5'-yl imidodiphosphate) as well. Surprisingly, the release of tryptophanase from GroEL was facilitated in the presence of ADP as well. We concluded that the binding of nucleotides such as ATP and ADP changed the conformation of GroEL and facilitated the dissociation of tryptophanase molecules. The conformation formed in the presence of ADP was distinct from the conformation formed in the presence of ATP, as shown by the selective dissociation of various folding proteins from the two conformations.     

Reactivation of 3-phosphoglycerate kinase from its unfolded proteolytic fragments

Limited trypsinolysis of pig muscle 3-phosphoglycerate kinase yielded a nicked enzyme without loss of catalytic activity [Jiang, S. X. & Vas, M. (1988) FEBS Lett. 231, 151-154]. The reactivation rate of the nicked enzyme after denaturation does not differ substantially from the reactivation rate of the denatured intact enzyme: t 1/2 varies between 70-110 s at 25 degrees C, pH 7.0 in both cases. Thus, the absence of a covalent linkage between the two proteolytic fragments of the enzyme molecule apparently does not affect the refolding. The two proteolytic fragments can be separated by FPLC under denaturing conditions. Fluorescence spectra of the isolated fragments may indicate that the tryptic cleavage site is within the N-terminal domain. Thus, the larger fragment (molecular mass about 30 kDa) probably contains the whole nucleotide-binding C-terminal domain plus a small part of the N-terminal domain. The inactive isolated fragments were used in renaturation experiments to study the reassembly of active 3-phosphoglycerate kinase. Kinetic measurements revealed the presence of a bimolecular rate-limiting step of reactivation. Separate preincubation of the fragments under renaturing conditions did not cause substantial acceleration of reactivation. This implies that assembly of the separate structural units (possibly domains) may limit the reactivation of the intact enzyme.     

Phosphofructokinase. II. Role of ligands in pH-dependent structural changes of the rabbit muscle enzyme.

The effect of ligands, including substrates and allosteric effectors, on the pH-dependent inactivation and reactivation of rabbit muscle phosphofructokinase has been examined in terms of the mechanism proposed previously (Bock, P.E. and Fireden, C. (1976) J. Biol. Chem. 251, 5630-5636). It is concluded thatt many ligands exert their effect by binding preferentially to either protonated or unprotonated forms of the enzyme and thus shifting an apparent pK for the inactivation or reactivation process. ATP and fructose 6-phosphate influence the apparent pK to different extents and in different directions, with ATP binding preferentially to the protonated forms and fructose 6-phosphate to the unprotonated forms. Enzyme inactivated by ATP can be reactivated by the addition of fructose 6-phosphate. The experiments indicate that inactivation and reactivation in the presence of these ligands can occur by kinetically different pathways as has been found for these processes in the absence of ligands. The results are discussed in relation to what might be expected for ligand binding properties of the enzyme as a function of pH, temperature, and enzyme concentration. The effect of ATP and MgATP is complex, perhaps representing more than one site of binding. Citrate appears to bind preferentially to protonated forms of the enzyme while fructose 1,6-bisphosphate and AMP bind preferentially to the unprotonated forms. ADP, K+, and NH4+ appear to have little or no preference in binding to different enzyme forms.     

Activation of L-lysine epsilon-dehydrogenase from Agrobacterium tumefaciens by several amino acids and monocarboxylates

The activation of lysine epsilon-dehydrogenase [EC 1.4.1.] by L-lysine was dependent on lysine concentration and was accompanied by association of the dimeric enzymes to a tetramer. The lysine concentration required for the half-maximal activation was 0.28 mM, which was lower than the Km value for L-lysine. In addition to L-lysine, several compounds, which were neither substrates nor inhibitors, activated the enzyme. The compounds which activated the enzyme have common structural characteristics: they have both a carboxyl group and a hydrophobic side chain. These activators also induced the association of the enzyme. The activation of the enzyme occurred well over the pH range 5.0 to 7.5, and the maximal activation was obtained by preincubation for 5 min at 30 degrees C and pH 7.4, when 5 mM L-lysine or 6-aminocaproate was used as an activator. NADH binding experiments indicated that about 2 mol of NADH bind to 1 mol of the tetrameric enzyme: the dimeric enzyme has one catalytic site. Binding experiments with n-[1-14C]heptanoate and L-[U-14C]lysine showed that approximately 2 mol of ligands bind to 1 mol of the dimeric enzyme and L-lysine could not bind to the catalytic site of the enzyme in the absence of NAD+. These results indicate the presence of one catalytic site and two activator binding binding sites in the dimeric enzyme.     

Beta-cyclodextrin-bonded silica assists alkaline phosphatase and carbonic anhydrase refolding in a solid phase assisted refolding approach

The processes of protein refolding by artificial chaperones suffer from tedious steps of purifications which will finally affect the production costs. Replacement of the soluble stripping agent with immobilized beta-cyclodextrin or beta-cyclodextrin polymer beads might elevate some of these problems. Regarding this fact, we synthesized and evaluated various cyclodextrin-bonded silica particles to evaluate the refolding yields of denatured alkaline phosphatase and carbonic anhydrase. Our results indicated that refolding of denatured alkaline phosphatase raised from 30%, in the absence of chaperone, to about 65% in the presence of 70 mg/ml of the beta-cyclodextrin-bonded silica gel and to 74% in the concomitant presence of the new stripping agent and MgSO_4, a yield near to stripping by soluble beta-cyclodextrin. The refolding yield of carbonic anhydrase in the presence of beta-CD-bounded silica gel resin was significantly lower than the value obtained in the presence of soluble beta-CD (76% vs 54%). These data indicate that refolding of proteins by the silica gel immobilized beta-CD resin can be achieved though with lower yields. Regarding the high cost of downstream purification steps associated with soluble beta-CD, application of insoluble stripping agent might provide an alternative approach to cut down the industrial costs.