Human aldose reductase unfolds through an intermediate

Background: Human aldose reductase (hAR) is the first and rate-limiting enzyme of the polyol pathway. For the development of secondary complications of diabetes in chronic hyperglycemic conditions, one of the critical factors is the increased flux of glucose through the polyol pathway. Due to this clinical implication, hAR attracted considerable attention from the drug discovery perspective. In spite of extensive characterization in the context of biochemical and structural aspects, we know very little about the unfolding behavior of hAR. This study reports equilibrium unfolding studies of hAR. Methods: We carried out thermal denaturation and chemical-induced equilibrium unfolding studies of hAR monitored by circular dichroism and fluorescence spectroscopy. Results: Thermal denaturation studies presented a classical picture of two-state unfolding from native to the denatured state. The data was used to derive thermodynamic parameters and study the thermostability of hAR. Chemical induced equilibrium unfolding studies led us to discover an intermediate state, which gets populated at 3.5-4.0 M and 0.7-2.0 M of urea and GuHCl, respectively. Thermodynamic parameters derived from chemical-induced unfolding are in agreement with those obtained from thermal denaturation of hAR. Conclusion: This study revealed that aldose reductase unfolds from native to the unfolded state via an intermediate. Assessment of the thermodynamic stability of native, intermediate, and unfolded states shows that significant energy barriers separate these states, which ensures the cooperativity of unfolding. As hAR functions in cells that are under osmotic and oxidative stress, these in vitro findings may have implications for its native conformation under the physiological state.


Amendments from Version 1
We are thankful to the reviewers for their valuable comments. In response to the comments, we have changed the manuscript. The followings are significant changes from the previous version of the manuscript. 1. We have included additional references and have stressed the relevance of the study. 2. Figures have been revised as per suggestion by the reviewer(s). 3. The word 'thermal unfolding' has been replaced by 'thermal denaturation.' 4. We have gone through the manuscript carefully and have made necessary changes regarding linguistic issues.

Introduction
Human aldose reductase (hAR) (EC 1.1.1.21) is an NADPHdependent oxidoreductase that belongs to the superfamily of aldo-keto reductases 1 . hAR is the first and rate-limiting enzyme of the polyol pathway and converts glucose to sorbitol 2 . The polyol pathway is up-regulated under hyperglycemic conditions, and a significant proportion of glucose gets fluxed through this pathway, which leads to the accumulation of sorbitol, consumption of NADPH, and redox imbalance of NADPH/NADP + ratio. All these factors have been linked with various tissue-based pathologies associated with secondary complications of diabetes mellitus 3 . Due to its clinical importance from the perspective of developing potent inhibitors to prevent or delay the onset of secondary diabetic complications, hAR has been widely studied in almost every aspect 4 .
Extensive information is available in the literature about the structure and function of hAR, mainly related to active site of hAR from high-resolution crystal structures with a number of potential inhibitors 5 , flexibility in the hAR binding site pocket 6 and the thermodynamics of closing/opening of the specificity pocket within binding site pocket of hAR 7 . There is little investigation related to the folding/unfolding mechanism of hAR.
Protein folding is a fundamental process in all living systems. The linear sequence of amino acids encodes the information required for a polypeptide to fold. Despite extensive research and progress made over the years to understand this fundamental process, a complete understanding of protein folding mechanism(s) and the folding code remains elusive 8 . Various models have been proposed to explain the mechanism underlying protein folding reaction from time to time 9 . 'New view folding model' based on energy landscape theory (ELT) and 'defined pathways model' based on the concept of foldons are current alternatives to explain the protein folding mechanism. 10 .
Among several hallmarks of protein folding are its spontaneity and cooperativity, along with fast reaction timescale. Usually, a small free energy change is required to shift the equilibrium between the folded and unfolded state ensemble. Complex cellular solvation environment plays a vital role in regulating the equilibrium between folded and unfolded ensemble in vivo, which are separated by small free energy barrier (~5 -10 Kcal mol-1) 11 . Under physiological conditions, protein structure fluctuates among different native conformations separated by close free energy barriers 12 . Since hAR activity leads to sorbitol accumulation, leading to osmotic stress, it seems to function under stress conditions which might perturb its native conformation ensemble. Here we report on thermal denaturation and chemical induced unfolding studies of hAR. Thermal denaturation revealed a simple two-state transition, whereas chemical induced unfolding led us to discover an intermediate state during hAR unfolding.

Materials
All chemicals were reagent grade and purchased from Sigma-Aldrich.

Protein purification
The hAR cDNA cloned into expression vector pET-15b (Novagen) was a kind gift from Dr. Alberto Podjarny (Department of Integrated Structural Biology, Institut de Génétique et de Biologie Moléculaire et Cellulaire, CNRS, INSERM, France). The plasmid, coding for a hexahistidine-tagged protein, was expressed into E. coli strain BL21 (DE3) (Novagen). The cells with recombinant plasmid were grown with 100 μM ampicillin at 37°C to an OD 600 nm value of 0.7, and protein expression was induced by adding 1 mM IPTG. Cells were grown for a further 3 hours at 37°C. All further operations were carried out at 4°C unless otherwise stated. Cells were centrifuged, resuspended, and lysed by sonication. A Ni-NTA affinity column (GE Healthcare) was used for protein purification. The material used for the stationary phase for the column was Ni-Sepharose, and the flow rate of the column was adjusted to 0.5 ml min -1 . Imidazole and other salts were removed by repeated dialysis in 50 mM potassium phosphate, pH-7 buffer containing 50 mM NaCl. Protein concentration was estimated using the molar extinction coefficient and absorbance reading at 280 nm. The histidine tag from recombinant protein was removed by thrombin (4 units of thrombin per mg of recombinant protein at room temperature for 3 hours). The cleaved protein was passed through the Ni-NTA column, and purified protein without tag was collected as flow-through. Enzyme activity was checked as per standard assay 13 . We analyzed the homogeneity and molecular weight of hAR with and without histidine tags under denaturing conditions on 15% SDS-PAGE. Purified hAR was stored at -20°C for further studies.

Thermal denaturation
Thermal denaturation was carried out at a final concentration of 2.8 μM protein in 50 mM potassium phosphate buffer, pH 7.0, containing 50 mM NaCl and 0.1mM TCEP. The transition between 20-70 °C was followed using a far-UV circular dichroism (CD) signal at 222 nm by using a 0.1 cm path-length cuvette at a sampling rate of 1.0 °C min -1 in a Jasco J-810 spectropolarimeter. After subtracting buffer blank, we have reported the change in ellipticity (millidegree) at 222 nm.

Chemical induced unfolding
To follow chemical induced unfolding, we prepared samples with 1.4 μM protein concentration in phosphate buffer (described earlier), containing different concentrations of GuHCl/ urea. Samples were incubated for 12 hours to reach equilibrium at 25°C, after which no change in signal occurred either in fluorescence or CD spectra. Trp fluorescence (excitation at 295 nm and emission recorded between 300 nm to 400 nm) and ANS fluorescence (excitation at 370 nm and emission recorded between 400 nm to 600 nm) measurements were performed using a Hitachi F-7000 fluorescence spectrophotometer for GuHCl samples and Jasco J-815 spectropolarimeter for urea samples. Far-UV CD measurements were performed using Jasco J-810 spectropolarimeter for GuHCl samples and Jasco J-815 spectropolarimeter for urea samples. Quartz cuvette of 1 cm and 0.5 cm path-length were used for fluorescence and CD measurements, respectively. All measurements were done at 25 °C. We have reported spectra as ellipticity (millidegree) after baseline correction.

Heat capacity change (ΔC p ) calculations
We have calculated the value of ΔC p for the unfolding of hAR from the change in accessible surface area (ΔASA) according to Equation 1 14 ProtSA webserver was used to calculate the change in accessible surface area from native to the unfolded conformation of hAR 15 .

Data analysis
GraphPad Prism version 7.04 for Windows (GraphPad Software, La Jolla, California) was used for the analysis of thermal denaturation and chemical-induced unfolding data based on two-state and three-state models respectively, as described in following sections.

Thermal denaturation
The Least-square analysis was used to fit the data to Equation 2 16 .
Where A n and A u are native and unfolded state baseline intercepts, respectively, and b N and b U are native and unfolded baseline slopes, respectively. ΔH m is enthalpy change at melting temperature (T g ). T is absolute temperature, and R is the gas constant.

Calculation of ΔG values for the transition region
Signal for native (Y N = A n + b N × T) and unfolded baseline (Y U = A u + b U × T) for every point in transition region was calculated from Equation 2. If Y is signal for a particular point in transition region, then the fraction of unfolded protein (F u ) at this point is given by Equation 3 .
The equilibrium constant (k) was calculated from the relative population of species using Equation 4.
ΔG was calculated as a function of temperature using Equation 5 .
Thermal stability curve The thermal stability curve of hAR was constructed based on Equation 6-8 17 .
( ) Where ΔH T and ΔS T are enthalpy change and entropy change, respectively, at temperature T with reference to T g . T h , T s and T g are the temperatures at which ΔH, ΔS, and ΔG are zero, respectively. ΔG s is the stabilization free energy of the native state relative to the unfolded state.

Chemical induced unfolding
The least-square analysis was used to fit the data to Equation 9 16 .

Results
Thermal denaturation monitored by far-UV CD Change in ellipticity at 222 nm fitted well based on a two-state model ( Figure 1A). This analysis gave values for ΔH g and T g , which along with the ΔC p value calculated from Equation 1 were used for non-linear regression of the transition region (±5 kJ mol -1 ) to Equation 8 ( Figure 1C). Values of ΔH and ΔS were calculated over an extended range of temperatures by using Equation 7 and Equation 8, respectively ( Figure 1D). The thermal stability curve is an extrapolation of transition region, assuming constant ΔC p during the unfolding transition ( Figure 1E). The relationship between T s , T h and ΔG (T s -T h = ΔG s /ΔC p ) is presented in Figure 1F. Thermodynamic parameters obtained from the analysis of thermal denaturation data are listed in Table 1. All raw data are available as Underlying data 18 . Fluorescence emission profiles of hAR equilibrated with different concentrations of denaturants are presented in Figure 2 ( Figure 2A and Figure 3A for urea and GuHCl, respectively). A plot of λ max against denaturant concentration indicated a cooperative transition from native to unfolded state ( Figure 2B and     Figure 4C1 and 4C2 for GuHCl and urea, respectively). Thermodynamic parameters derived from far-UV CD data are listed in Table 1.

Discussion
The In all three probes used in studying unfolding, the value of ΔG (N-I) obtained is ~30 kJ mol -1 and ~15 kJ mol -1 for ureaand GuHCl-induced unfolding, respectively while a ΔG s of ~70 kJ mol -1 is almost same for both denaturants (Table 1)

9.
However, intrinsic fluorescence intensity is easily affected by numerous environmental factors and is not as reliable as the emission maximum wavelength. Yes, there is an obvious increase in ANS fluorescence. Previously Turoverov and his co-authors have shown that the appearance of off-pathway aggregates, in some cases, can enhance ANS fluorescence. It is necessary to avoid the misleading interpretation by identifying the potential intermediate(s) via more methods.
It is unclear whether all the solid lines are fitted data or not. Moreover, it is better to provide the residuals to facilitate the readers to evaluate whether the fitting is appropriate.
In Figure 3, the ANS profiles seem to comprise more than one state and are likely to be the sum of at least two states centered at around 1 and 2 M GdnHCl? CD signals are recommended to report in molar ellipticities. The quality of the far-UV CD spectra is not enough. High concentration of denaturants may affect the signals at low wavelengths, generally below 210 nm. However, the low signal-to-noise seems to be affected by other factors. The authors can try to adjust the measuring parameters such as the slit width and number of repetitions. Furthermore, the authors mentioned that all chemicals were Sigma reagent grade. To my experience, GdnHCl with an ultrapure grade will produce much better signals that the lower grades. Note that the impurities or contaminated solvents will greatly affect spectroscopic measurements. Before spectroscopic measurements, the samples should be filtered and degassed. Detailed guidelines of spectroscopic measurements can be found in the paper by Kelly et al. Figures 2 and 3 are quite different. Are these data normalized by the same method? Furthermore, the spectra of N in panels E-2 should refer to the state in the absence of denaturant. Why the spectrum of N in Figure 2E-2 is different from that in Figure 3E-2? To me, the spectrum of N in Figure 2E-

Is the work clearly and accurately presented and does it cite the current literature? Partly
Is the study design appropriate and is the work technically sound? Yes

If applicable, is the statistical analysis and its interpretation appropriate? Yes
Are all the source data underlying the results available to ensure full reproducibility? Partly

Are the conclusions drawn adequately supported by the results? Yes
No competing interests were disclosed.

Competing Interests:
I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias You can publish traditional articles, null/negative results, case reports, data notes and more The peer review process is transparent and collaborative Your article is indexed in PubMed after passing peer review Dedicated customer support at every stage For pre-submission enquiries, contact research@f1000.com