The chaperonin GroEL binds to non-native substrate proteins via hydrophobic interactions, preventing their aggregation, which is minimized at low temperatures. In the present study, we investigated the refolding of urea-denatured rhodanese at low temperatures, in the presence of ox-GroEL (oxidized GroEL), which contains increased exposed hydrophobic surfaces and retains its ability to hydrolyse ATP. We found that ox-GroEL could efficiently bind the urea-unfolded rhodanese at 4°C, without requiring excess amount of chaperonin relative to normal GroEL (i.e. non-oxidized). The release/reactivation of rhodanese from GroEL was minimal at 4°C, but was found to be optimal between 22 and 37°C. It was found that the loss of the ATPase activity of ox-GroEL at 4°C prevented the release of rhodanese from the GroEL–rhodanese complex. Thus ox-GroEL has the potential to efficiently trap recombinant or non-native proteins at 4°C and release them at higher temperatures under appropriate conditions.
- ATP hydrolysis
- Escherichia coli
GroEL is a protein from Escherichia coli whose synthesis is stimulated under stressful conditions, including oxidative stress. Numerous studies have shown that the chaperonin GroEL binds to the non-native state of substrate polypeptides and hence prevents their aggregation in vivo [1,2]. GroEL has a central cavity that is lined with hydrophobic residues that serve to bind substrate proteins, preventing their aggregation. The protein refolding/release is then triggered by the binding of the co-chaperonin, GroES to the GroEL–substrate protein complex, and the hydrolysis of ATP by GroEL . Ox-GroEL (oxidized GroEL) with low concentrations of H2O2, such as those of physiological significance, was shown previously to have an increased exposure of hydrophobic surfaces and retain its ability to hydrolyse ATP . Ox-GroEL would, therefore, be expected to efficiently bind to non-native proteins and support their refolding. The extent of the protein refolding associated aggregation is reduced by achieving the refolding step at low temperature. Thus we evaluated the efficiency of the ox-GroEL-assisted refolding at low (4°C) temperature as opposed to the same process at higher temperatures. We found that ox-GroEL could efficiently bind the urea-unfolded rhodanese forming a stable complex at low temperature. The release of rhodanese from ox-GroEL, however, could not be observed unless the temperature was increased. It was found that the loss of the ATPase activity of ox-GroEL at low temperature prevented the release of rhodanese from the ox-GroEL–rhodanese complex.
MATERIALS AND METHODS
All the reagents used were of analytical grade. A new bottle of 3% H2O2, purchased from Longs Drugs, was used in each experiment. Rhodanese from bovine liver was prepared as described previously  and stored at −70°C as a crystalline suspension in 1.8 M ammonium sulfate. Rhodanese concentration was determined spectrophotometrically using a molecular mass of 33 kDa  and an absorption coefficient of 57750 M−1·cm−1. The chaperonin GroEL was purified as described  from lysates of E. coli cells bearing the multicopy plasmid pGroESL  that were provided by Carl Frieden. After purification, the chaperonin was dialysed against 50 mM Tris/HCl, pH 7.8 and 1 mM DTT (dithiothreitol). Then, glycerol was added at a final concentration of 10% (v/v) and the protein was rapidly frozen in liquid nitrogen and stored at −70°C. Before its utilization, the chaperonin was dialysed against 50 mM Tris/HCl, pH 7.8 and kept at 4°C. The protomer concentration of GroEL was estimated by its absorbance at 280 nm using an absorption coefficient of 12200 M−1·cm−1  and a molecular mass of 57.259 kDa .
Oxidation of GroEL with H2O2
Ox-GroEL was prepared by incubating GroEL (416 nM, 14-mers) at 37°C with 2 mM H2O2 for 16 h in 50 mM Tris/HCl, pH 7.8. Excess oxidant was removed from the protein by dialysis into 50 mM Tris/HCl, pH 7.8, followed by a Bio-spin column from Bio-Rad.
Complex formation between ox-GroEL and urea-unfolded rhodanese
Rhodanese (108 nM) that had been unfolded with urea as described previously  was diluted into ice-cold (4°C) buffer containing 50 mM Tris/HCl, pH 7.8 and 5 mg/ml dextran in the absence or presence of ox-GroEL (108 nM, 14-mers) in a final volume of 250 μl. After centrifugation, the sample was fractionated into four equal aliquots, which were analysed by SDS/PAGE using the Laemmli system . The gel was stained with Coomassie Blue .
Refolding of urea-unfolded rhodanese
Rhodanese (9 μM) was unfolded in 200 mM sodium phosphate buffer, pH 7.4, containing 1 mM 2-mercaptoethanol and 8 M urea for 1 h at 22°C. Then, 3 μl of unfolded rhodanese were diluted into a final volume of 250 μl with a buffer containing ox-GroEL (108 nM, 14-mers) in 50 mM Tris/HCl, pH 7.8, at 4°C. Refolding was initiated by adding the following protein releasing factors: GroES (216 nM, 7-mers), 200 mM 2-mercaptoethanol, 50 mM sodium thiosulfate, 10 mM MgCl2, 10 mM KCl and 2 mM ATP. Samples were incubated at 4°C or higher temperatures for 90 min. To assess the refolding of urea-unfolded rhodanese, the activity of the enzyme was measured by a discontinuous colorimetric method based on the absorbance at 460 nm of the complex formed between ferric ions and the reaction product, thiocyanate . The percentage reactivation was calculated based on the activity of native enzyme that had been subjected to similar refolding conditions.
ATPase activity of GroEL
The ATPase activity of GroEL was determined as described previously . Ox-GroEL (108 nM, 14-mers) was incubated in 50 mM Tris/HCl, pH 7.8, 2 mM MgCl2, and 2 mM KCl at the indicated temperatures. After a 5 min incubation, the mixture was made 1 mM in ATP by the addition of 5 μl of 0.1 M ATP in a final volume of 500 μl, to initiate the hydrolysis reaction. Periodically, 120 μl aliquots were removed and mixed with 3 ml of the 360-3 diagnostic reagent (Sigma Diagnostics), and the absorbance at 340 nm was read at 22°C using a Shimadzu Spectrophotometer. The assay was carried out as described by the manufacturer. This assay is based on the reaction of inorganic phosphorous with ammonium molybdate, in the presence of H2SO4, which results in the production of an unreduced phosphomolybdate complex. The absorbance of this complex at 340 nm is directly proportional to the inorganic phosphorous concentration. The obtained absorbance values were compared with those of the 360-5 calcium/phosphorous standards (Sigma Diagnostics).
RESULTS AND DISCUSSION
Formation of complex between ox-GroEL and urea-unfolded rhodanese
We evaluated the ability of ox-GroEL to bind the unfolded protein rhodanese at 4°C, as opposed to room temperature (22°C), by directly examining the formation of the ox-GroEL–rhodanese complexes using differential sedimentation . These conditions allowed sedimentation of ox-GroEL, without significant sedimentation of rhodanese due to the large difference in the sedimentation values (S) of the proteins: 23S and 3S for GroEL and rhodanese respectively [11,16]. After centrifugation, samples were fractionated into four 50 μl aliquots and these were subjected to SDS/PAGE analysis. Figure 1 shows the binding of urea-unfolded rhodanese to ox-GroEL at 22°C (left-hand panel) and 4°C (right-hand panel). Lanes 1–4 in both panels correspond to the fractions collected from top to bottom. As shown in Figure 1, ox-GroEL sedimented completely and was found mostly in the bottom fraction (upper band in lane 4 of both panels). Rhodanese was found to co-sediment with ox-GroEL at 4°C (lower band in lane 4 of right-hand panel). This is confirmed by the relative depletion of rhodanese from fractions 1 to 3 (lanes 1–3 of right-hand panel). These results are expected if there was a strong interaction between ox-GroEL and unfolded rhodanese, resulting in a complex between ox-GroEL and rhodanese at 4°C. However, when unfolded rhodanese was centrifuged in the presence of ox-GroEL at 22°C, analysis of the fractions indicated that unfolded rhodanese did not completely sediment and was found in similar amounts in all fractions (lanes 1–4 of left-hand panel). These results indicate that unfolded rhodanese did not bind efficiently to ox-GroEL upon dilution in the buffer containing the chaperonin. A similar result was obtained when unfolded rhodanese, alone, was centrifuged as above at either temperature (results not shown).
Refolding of urea-unfolded rhodanese in the presence of ox-GroEL
The refolding of urea-unfolded rhodanese by ox-GroEL was investigated at 4°C. A sample of ox-GroEL–rhodanese complex formed at 4°C was supplemented with the co-chaperonin GroES and other protein releasing factors. The release of active rhodanese was monitored as described in the Materials and methods section. At 4°C, a maximum recovery of approximately 16% active rhodanese was obtained after 90 min. Thus, in view of the low recovery of active rhodanese at 4°C, the ox-GroEL-assisted refolding of urea-unfolded rhodanese was investigated at higher temperatures. Samples of ox-GroEL–rhodanese complexes prepared at 4°C and supplemented with the protein releasing factors were incubated at temperatures between 4°C and 37°C for 90 min. Figure 2 shows that a gradual increase was observed in the recovery of active enzyme as the temperature of the refolding mixture was increased. A maximum recovery of 73% activity was obtained at 22°C. Similar recoveries were obtained at temperatures between 22°C and 37°C (results not shown). The slight decrease in the percentage of activity recovered at 37°C is probably due to the intrinsic thermal instability of the refolded enzyme.
Refolding of urea-unfolded rhodanese in the absence of ox-GroEL
To determine if the observed recovery (16%) of active rhodanese, after 90 min at 4°C, may be due, in part, to unassisted refolding of rhodanese, the refolding of rhodanese in the absence of ox-GroEL was examined at 4°C. Figure 3 shows that ~44% of active enzyme was recovered after 90 min. In agreement with previous results , unassisted refolding of rhodanese at low temperature was slow and ~65% active rhodanese was recovered after 3 h with no significant increase of the recovered amount after 24 h. These results indicate that a significantly higher recovery (~44%) would be expected in the presence of ox-GroEL if rhodanese was completely released upon the addition of the releasing factors after 90 min at 4°C. The lower observed recovery (16%) is probably due to the fact that most rhodanese remained bound to ox-GroEL even after addition of the releasing factors.
Effect of the releasing factors on the ox-GroEL–rhodanese complex at 4°C
To confirm that rhodanese remained bound to ox-GroEL after addition of the releasing factors, we used the differential centrifugation assay. Figure 4 shows the binding of urea-unfolded rhodanese to ox-GroEL after addition of the protein releasing factors and incubation for 90 min at 4°C (right-hand panel) or 22°C (left-hand panel). Lanes 1–4 correspond to the fractions collected from top to bottom. As shown in the Figure, ox-GroEL sedimented completely and was found mostly in the bottom fraction (upper band in lane 4 of both panels). Rhodanese was found to co-sediment with ox-GroEL only at 4°C (lower band in lane 4 of right-hand panel), indicating that rhodanese remained bound to ox-GroEL. Similar results were obtained after the ox-GroEL–rhodanese complex was incubated with the releasing factors for 3 or 6 h at 4°C (results not shown). As shown in Figure 4, when unfolded rhodanese and ox-GroEL were centrifuged after addition of the protein releasing factors, and incubation for 90 min at 22°C (left-hand panel), the protein gel analysis showed that unfolded rhodanese was found in similar amounts in all fractions (lanes 1–4 of left-hand panel). These results indicate that rhodanese was released from the ox-GroEL after addition of the protein releasing factors and incubation at 22°C.
Temperature dependence of the ATPase activity of ox-GroEL
To determine whether the low recovery of active rhodanese at 4°C was due to the inability of ox-GroEL to hydrolyse ATP, the ATPase activity of ox-GroEL was measured at different temperatures. Figure 5 shows that at 22°C, the obtained rate of hydrolysis of ATP by ox-GroEL was in agreement with the previously reported value for this form of GroEL . A higher ATPase activity was obtained at 37°C as was reported previously for native GroEL . However, the ability of ox-GroEL to hydrolyse ATP at temperatures lower than 22°C was significantly decreased. At 4°C, the rate of ATP hydrolysis by the chaperonin was approximately 34% of that seen at 22°C. The decrease in the hydrolysis of ATP by ox-GroEL at lower temperatures correlates fairly well with the expected linear decrease in activity of most enzymes at less than optimal temperatures  and is similar to that displayed by native GroEL . It has been clearly demonstrated that ATP hydrolysis is used to drive conformational changes in the GroEL structure that cause the release of the bound polypeptide [3,21]. Therefore the lower recovery of active rhodanese from the ox-GroEL–rhodanese complex observed at low temperature was perhaps due to the inability of ox-GroEL to undergo the conformational changes that are triggered by ATP hydrolysis and that are needed for the discharge of the bound polypeptide. Apparently, the temperature increase resulted in release of the rhodanese from ox-GroEL, in the absence of ATPase activity, because of the temperature-induced conformation of GroEL. The ability of ox-GroEL to form a complex with unfolded rhodanese at low temperatures requiring molar stoichiometric ratios was in contrast with the formation of complexes using native GroEL in that native GroEL must be added in excess .
In conclusion, the results of the present study suggest that ox-GroEL may potentially be used at 4°C to efficiently bind recombinant or non-native proteins that do not spontaneously refold and prevent them from aggregating. Release of correctly folded substrate proteins from the chaperonin would then be performed at higher temperatures, under buffer conditions that are specific for each protein, in order to maximize recovery of active protein. It will be interesting to assess the role of ox-GroEL, in vivo, in organisms that thrive at low temperatures.
Girish Melkani, Gustavo Zardeneta and Jose Mendoza designed the study. Girish Melkani and Robin Sielaff carried out the experiments. Girish Melkani drafted the paper and Gustavo Zardeneta revised it. Jose Mendoza directed the project and revised the paper prior to submission.
This work was supported by the National Institutes of Health [grant number S06 GM59833-01 (to J.A.M.)].
Abbreviations: ox-GroEL, oxidized GroEL
- © The Authors Journal compilation © 2012 Biochemical Society