Perturbation analysis in the N-terminal domain of hPGK1: a biophysical, structural and statistical mechanical study

doi:10.21203/rs.3.rs-1945879/v1

Download PDF

Article

Perturbation analysis in the N-terminal domain of hPGK1: a biophysical, structural and statistical mechanical study

https://doi.org/10.21203/rs.3.rs-1945879/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Phosphoglycerate kinase has been a model for the stability, folding cooperativity and catalysis of a two-domain protein. The human isoform 1 (hPGK1) is associated with cancer development and rare genetic diseases that affect several of its features. To investigate how mutations affect hPGK1 folding landscape and interaction networks, we have introduced mutations at a buried site in the N-terminal domain (F25 mutants) that either created cavities (F25L, F25V, F25A), enhanced conformational entropy (F25G) or introduced structural strain (F25W) and evaluated their effects using biophysical experimental and theoretical methods. All F25 mutants folded well, but showed reduced unfolding cooperativity, kinetic stability and altered activation energetics according to the results from thermal and chemical denaturation analyses. These alterations correlated well with the structural perturbation caused by mutations in the N-terminal domain and the destabilization caused in the interdomain interface as revealed by H/D exchange under native conditions. Importantly, experimental and theoretical analyses showed that these effects are significant even when the perturbation is mild and local. Our approach will be useful to establish the molecular basis of hPGK1 genotype-phenotype correlations due to phosphorylation events and single amino acid substitutions associated with disease.

Protein folding

Structural Perturbation

Disease

Phosphoglycerate kinase (PGK) is a glycolytic enzyme that catalyzes the reversible phosphotransfer reaction between ATP and 3-phosphoglycerate (3-PG) to form ADP and 1,3 bisphophoglycerate (1,3-BPG) in the presence of Mg^{2+ 1–4}. Its main activity is to generate ATP in glycolysis, and this activity is highly conserved among all living organisms ¹. Structurally, PGK is typically a monomeric protein of about 45 kDa with two well-defined domains: an N-terminal domain (NTD) that binds 3-PG/1,3-BPG (usually the first 200 amino acids), a C-terminal domain (CTD) (typically the last 200 amino acids) that binds the nucleotide and both domains are linked by a 10–15 amino acids long α-helix (Fig. 1) ^3,5. Binding of Mg²⁺ triggers the transition between the inactive open conformation to a closed conformation that leads to catalysis⁴ (Fig. 1). Importantly, PGKs from different organisms have been crystalized in different ligation states (only for the human isoform 1, hPGK1, we retrieved 16 different structures by February 2022 from the Protein Data Bank; Table S1), and metal binding may trigger long-range cooperative communication between domains (Fig. 1). These facts make PGK a highly suitable model to investigate allosteric communication upon ligand binding (or mutations)^4,6. Actually, due to the reversible nature of its unfolding (at least by chemical denaturants), PGK has been the subject of several key studies on folding thermodynamics, kinetics and cooperativity for a two-domain protein in vitro, inside living cells and even in multicellular organisms ^7–10. We must note that this type of studies carried out at high resolution typically use small and/or single domain proteins ^11–13. Thus, this type of investigation using a two-domain and disease-associated protein can provide novel insight into the communication of mutational and ligand binding effect across the structure of a middle size and more structurally and functionally complex protein with potential biomedical implications.

Particularly in the last decade, research hPGK1 has attracted additional attention for several reasons: 1) hPGK1 is multifunctional (a so-called moonlighting protein). As we have previously mentioned, it is essential for glycolysis, but it also phosphorylates other proteins ¹⁴, it is involved in DNA replication and repair ^14–16, protein:protein interactions ^14,17,18, it activates L-nucleoside analogues used in anticancer and antiviral activities ^19–22 and plays a key role in autophagy ¹⁴. 2) hPGK1 can operate in different cellular compartments. Obviously, due to its role in glycolysis, it is found at high levels in the cytosol. However, due to its multifunctionality, it can also localize in the nucleus ¹⁴, mitochondria ²³ and extracellularly ²⁴. These findings suggest that non-native states in the energy landscape of hPGK1 may play important roles in the transport to different compartments. For instance, to be imported to mitochondria, proteins must be unfolded by the mitochondrial import machinery or even presented to this machinery in a partially unfolded state by Hsp70 chaperones ²⁵. In the particular case of hPGK1, the interaction with Hsc70 and Hsp70 chaperones inside the cell is likely ¹⁸, thus suggesting a role of partially (un)folded states of hPGK1 in mitochondrial import of this enzyme (this import is associated with cancer metabolism and tumorigenesis) ¹⁴. 3) Alterations in hPGK1 activity and intra- or subcellular location are associated with several diseases, including cancer ^26,27, hemolytic anemia ³, myopathy ³ and neurological dysfunction (possibly including amyotrophic lateral sclerosis) ^3,24. In most cases, a deficiency in hPGK1 associated with disease is concomitant with a modification of the protein sequence by mutations ^2,27−30. These mutations often reduce catalytic efficiency, protein stability and unfolding cooperativity, and increase aggregation propensity ^3,27,29,30. In several cases, a single mutation manifests this loss of function through several of these detrimental mechanisms, and mutations associated with cancer development ²⁷ or genetic hPGK1 deficiency (OMIM 300653) ^2,3,28,29 behave similarly regarding these mechanisms.

Therefore, it is plausible that missense mutations may propagate their destabilizing effect across the structure of hPGK1 affecting different functional features such as activity, stability, or folding cooperativity ^13,31. In this work, we analyzed the wild-type (WT) protein as well as five mutants aimed to progressively change the magnitude of the local destabilizing effect at a single site. To do so, we mutated a fully buried and large hydrophobic residue (F25; see Fig. 1 and Table S1) to the smaller residues (L, V, A and G) expected to create cavities of different size in the hydrophobic core of the NTD, increase conformational entropy (mutation to G) as well as a mutation that increase the size of the side-chain (mutation to W) expected to generate structural strain. Although these mutations are not associated with disease, they are particularly suitable for this type of study (particularly cavity-creating mutations) because their effects on protein stability and energy landscapes have been extensively characterized by experiments and computation using typically smaller and simpler proteins ^31–36. Our approach uses a combination of experimental biophysical techniques including optical spectroscopy, calorimetry, thermal and chemical denaturation and hydrogen/deuterium exchange monitored by mass spectrometry (HDX-MS) with statistical mechanical calculations.

2.1. F25 mutations do not impair protein expression of hPGK1 as a folded protein

Upon expression in E.coli and purification with two chromatographic steps, IMAC and SEC (Figure 2A) we obtained hPGK1 of a very high purity and also confirmed that all variants showed similar hydrodynamic behavior to that of the WT protein (i.e. monomeric; the elution volume was very similar to that of ovalbumin, with a MW of 44 kDa). Since hPGK1 is often purified with certain small molecules bound to the protein (possibly nucleotides or small polynucleotides, based on the UV-visible absorption and CD spectra and consistent with the ability of the enzyme to interact with such type of molecules, e.g. ADP and ATP), we therefore included a step of precipitation with streptomycin between the IMAC and the SEC. The final protein samples contained virtually no additional signals in the near-UV range (Figure 2B).

Further conformational analyses were carried out using CD to test the effect of mutations on the overall conformation of the protein. Far-UV CD spectra revealed the typical spectra for a protein rich in α-helix such as hPGK1 (Figure 2C) and closely resembled results from previous studies ^28,37. We must note that the Far-UV CD intensity was slightly lower for the mutant F25G, suggesting that this mutation could slightly distort the overall secondary structure of the protein. Mutants F25G and F25W also reduced the strongest signal in the Near-UV CD spectra (at about 290 nm; consistent with the dichroic signal from Trp residues) suggesting local distortions of the tertiary structure (Figure 2D).

Overall, these analyses support that the F25 mutants largely maintained the overall structure of the WT protein.

2.2. Thermal stability of hPGK1 variants

Thermal denaturation of hPGK1 has been described well by using a simple two-state irreversible denaturation model ²⁸. To initially assess the effect of mutations on hPGK1 thermal stability, we acquired the CD signals spectra varying the temperature from 20 to 75 ^oC. Both Far- and Near-UV CD monitored thermal denaturation experiments provided similar results (Figure 3). The apparent T_m decreased according to the size of cavity (from 2-3 ^oC in F25L to 6-8 ^oC in F25G, compared with those values of WT), whereas F25W (that may introduce certain conformational strain) showed an apparent T_m very similar to that of F25A.

To provide deeper knowledge of the effects of F25 mutants on the kinetic stability (i.e. the rate constant for irreversible denaturation, k and activation energy, E_a) of hPGK1, we applied DSC using previously described protocols for this protein ³⁰. The results from these analyses are summarized in Figure 4 and Tables 1-2. In agreement with thermal denaturation monitored by CD, we observed a progressive decrease in T_m in the cavity-making mutants (F25G decreases the T_m by ~7 ^oC). The mutant F25W showed again a similar behavior to that of F25A, with a decrease of ~4.5 ^oC. The destabilization induced in the thermal denaturation process by each individual mutation using different probes (DSC, Far-UV and Near-UV CD spectroscopies) are very close at similar scan rates (Figure 3 and S1-S2).

It is also interesting to note that the decrease in thermal stability correlates strongly with the E_a values for irreversible denaturation and the T_m values as determined by DSC (Figures 4B-D and Tables 1-2). An overall linear fitting of E_avs. T_m provides a value for the change in activation heat capacity (ΔC_p^≠) of 10.6±0.5 kcal·mol^-1 in a very good agreement with our previous analyses that yielded a value of 9.1±0.8 kcal·mol^-1 ²⁸ (Figure 4D). Notably, the mutation F25G decreased the E_a value by 75 kcal·mol^-1 (basically to a half of that for the WT protein).

These results have remarkable implications for the kinetic stability of hPGK1 variants in vitro, since the decrease in T_m values correlated with the decrease in E_a values, and thus, the extrapolated values of the kinetic rate constants at 37 ^oC (and therefore, the associated half-lives for denaturation) vary strongly among variants. We observed a gradual decrease of the half-life for irreversible denaturation correlating with the size of the cavity, from a small ~15-fold in F25L to a large ~6000-fold in F25G. Again, F25A and F25W showed similarly low kinetic stabilities (about 600-fold lower kinetic stability than that of the WT protein; extrapolated from Arrhenius plots to 37 ^oC) (Table 1).

It is worth noting that previous studies have shown that mutations associated with hPGK1 deficiency (a rare X-linked genetic disorder ^3,29) also showed different effects on kinetic stability when extrapolated to 37 ^oC, up to 5 orders of magnitude lower than that of the WT protein in some cases. Interestingly, mutations found in COSMIC database also showed some signature of hPGK1 to be extremely sensitive to mutations regarding stability associated against thermal and chemical denaturation ²⁷.

To further investigate the correlation found between the changes in T_m and E_a values, we have carried out DSC experiments in the presence of urea (Table 2 and Figure S1-S2). These analyses allow to extract the kinetic m^≠ value for irreversible denaturation from detailed DSC analyses at different scan rates and urea concentrations ^28,30,38. Comparing this value with the theoretical m value derived from the protein size (i.e. in number of amino acids), we can provide an estimate of the degree of the native hPGK1 structure which is denatured in the transition state for irreversible thermal denaturation (Table 3). Actually, these analyses revealed that, in contrast to hPGK1-deficiency causing mutants characterized so far ²⁸, only the F25G showed a small decrease in value for the m^≠ (Table 3). Thus, the Hammond effect (i.e. that the kinetic stability decreases as the transition state becomes more native-like) proposed for the hPGK1-deficiency causing mutants characterized in ²⁸ does not seem to follow the same behavior that those of the F25 mutants characterized in the present work (Table 3).

2.3. Urea denaturation of F25 mutants reveals effects on both unfolding cooperativity and resistance to the urea-induced unfolding

An interesting feature of PGKs is that their chemical denaturation is generally reversible (e.g. using urea or guanidium chloride) ^8,28,39. Therefore, this type of analysis is amenable for comparing the effects of mutations on thermodynamic stability and unfolding cooperativity. Urea-denaturation of WT hPGK1 resembles quite well a two-state folder ^27,28 although the equilibrium m value is somewhat lower than the expected for a protein of this size. Importantly, urea-induced denaturation of 14 mutants, including those found in cancer cell lines and hPGK1-deficient patients, shows reduced unfolding cooperativity (in some cases the m_urea value was half of that of the WT protein). Interestingly, only in a few cases changes in the C_m values were observed ^27–29.

To compare the behavior of F25 mutants vs. those naturally-occurring and disease-associated, we carried out similar urea-induced unfolding experiments (Figure 5). As previously described, denaturation of hPGK1 provides a value for the unfolding free energy of about 8 kcal·mol^-1, based on the linear-extrapolation method (Table 4). The F25 mutants showed changes in urea-induced unfolding compared to the WT protein. The mutants F25L and F25V, as well as the strain-inducing F25W, caused a significant decrease in the unfolding cooperativity and C_m values (note that in this case F25W resembles F25V). Some mutants decrease the m_urea value by ≤ 50% of the value of the WT protein (Table 4). The most destabilizing variants were F25A (for which we could obtain reasonable and physically meaningful values) and F25G (that showed a denaturation profile similar to that of F25A but we could not get reliable estimations of C_m or m_urea, see Table 4). Therefore, for F25G and F25A, the unfolding cooperativity was severely reduced. Overall, these results suggest that increasing the cavity size, or introducing conformational strain, at least at the F25 site in hPGK1, increase the population of non-native state conformations, possibly even under native conditions. We further test this hypothesis by carrying out additional experimental and computational analyses under native conditions.

2.4. Proteolysis suggests a mild increase in the population of partially folded states in the hPGK1 mutants under native conditions

Proteolysis has been shown to provide information on the population of partially (un)folded, proteolytic sensitive substates in the native state ensemble of hPGK1 ^28,29. At low protease concentrations, the overall proteolytic rate for the WT protein is dependent on the protease concentration (Figure 6 and S4) ³⁰ and thus, effects on this rate constant reflect to some extent the increased population of partially unfolded (i.e. cleavable) substates under native conditions. Interestingly, mutants F25L, F25V and F25A showed modest effects on proteolysis (1.3-1.8-fold increased sensitivity; Figure 6 and S4) while mutations F25G and F25W showed more clear effects (1.7-2.6-fold and 3.9-5.3-fold increase, respectively). Interestingly, the latter mutants showed overall proteolytic rates rather insensitive to protease concentration (Figure 6 and S4) suggesting that in these two variants the unfolding rate to the cleavable state may dominate the overall proteolytic rate.

2.5. Perturbations in the native ensemble of F25 mutants by HDX-MS

To provide experimental information on the effects of F25 mutations on the local stability of the native state ensemble, we carried out HDX-MS experiments. Analysis of the WT protein in a ligand-free conformation (Figure 7) provided some interesting results. First, the apparent local stability based on the overall exchange rate for diverse protein segments is highly different (Figure 7A). Second, the exchange rates are complex, and phenomenologically described in many cases by a double exponential function with a burst-phase (Figure 7B and Table 5). This suggests that different conformational substates are populated in the native state ensemble.

Similar experiments were carried out with the F25 mutants (Figures S5-S9). F25 mutations selectively affected the stability of certain regions in the protein (Figures S5-S9). Mutants F25L and F25V showed mild stability decreases in regions close to F25, while these effects were more extensive due to the mutation F25A (Figure 8). Mutants F25G and F25W showed the strongest effects in stability and were propagated to regions further than the mutated sites, particularly relevant in the domain-domain interface (Figure 8 and S10-11).

The kinetic stability (as determined by DSC; Tables 1-3) and unfolding cooperativity (as determined by isothermal urea denaturation; Table 4) are likely associated with the interaction surface between the N- and C-terminal domains (shown in Figures S10-S11). We next analyzed the effects of F25 mutations on this interface from HDX-MS data (Figure S11). These analyses showed that even the mutation F25V begins to perturb this interface, and mutants F25G and F25W have the largest effect, affecting 30-40% of the residues forming this interface (Figure S11). These results strongly suggest that the effects of F25 mutations on conformational stability and unfolding cooperativity can be, at least, partially explained by the structural destabilization of the domain:domain interface.

2.6. Statistical mechanical analysis of the effects of F25 mutations

To understand the extent to which mutations affect the different regions of the protein, we introduced mutations in silico (using PyMol) and studied three representative F25 mutants (F25V, F25A and F25G) via the WSME model. In each of the cases, no further modulation of structure was introduced as these are primarily truncating mutations. The mutated structures were fed into the WSME model and the free-energy profiles were predicted without changing any model parameters. The resulting ranking of destabilization matched the expectations from experiments, i.e. F25G > F25A > F25V (Figure 9A) and also indicated that the folded state of the NTD and CTD is uncoupled to some extent in the native state ensemble (intermediates I₁ and I₂, Figure 9B). In addition, we calculated the positive coupling free energies () for these variants, a parameter that report on the degree of relative thermodynamic coupling between residues that are folded ⁴⁰. The effect of each of the mutations (mut) vs. the WT protein results in the mean difference in coupling free energies or perturbation, defined as , (plotted as a function of the residue number; Figure 9C). These results show a similar ranking of the mutations (F25G > F25A > F25V) according to their negative impact on thermodynamic coupling of folded regions in the native state ensemble.

As expected from the decoupled nature of the two domains, the perturbations are mostly localized to the NTD mirroring experimental observations. The mean perturbation magnitude (), i.e. the average of the values in Figure 9C corresponding to the first 200 residues (N-terminal domain), follows experimental HDX patterns with the F25G being the most perturbed (Figure 9D, top panel). The corresponding standard deviation in perturbation (), that measures the extent of distribution, followed a similar trend (Figure 9D, bottom panel), highlighting that the origins of the complex HDX-MS kinetics are a manifestation of population redistribution in the conformational landscape. The perturbation when mapped onto the structure (Figure 9E) vividly reveals the extent to which a single mutation affects neighbouring residues and the extent of coupling between them.

It is interesting that our local stability measurements using HDX-MS (Figure 8) and statistical mechanical models (Figure 9) show that destabilization caused by F25 mutations mostly restrict to NTD and increases with the predicted size of the cavity (F→L < F→V < F→A) and conformational entropy (F→A < F→G). These analyses indicate that gradual local destabilization (due to alterations of the interaction networks) is connected with the effects on the global stability (i.e. reversible by urea or kinetically-controlled by temperature) and unfolding cooperativity (Figures 4-5 and Tables 1, 2 and 4). Since hPGK1 WT resembles a two-state (un)-folder by urea, our results support that it displays certain heterogeneity of substates populated in the native state ensemble. It is likely that F25 mutation enhance such heterogeneity towards partially unfolded substates, thus accelerating HDX and reducing the apparent unfolding cooperativity. This link between local destabilization of the NTD and global denaturation effects (i.e. thermal and chemical denaturation) could be also partially explained by the effects of the F25 mutations on the domain:domain interface (Figure S11).

Understanding how mutations affect the native state ensemble is essential to obtain insight into mutational effects on protein stability and function as well as their potential pathological consequences ^41,42. This is a challenging task, since proteins display multiple features (stability, enzymatic function, regulation, biomacromolecular interactions, different subcellular locations, etc.) and large genetic diversity exists in human population ^41–43. In this work, we have investigated how mutations targeting a fully buried Phe residue propagate their effects across the structure of the model two-domain hPGK1 enzyme, associated with disease ^3,27,29. In our strategy, we progressively decreased the size of the side-chain (F→L, F→V, F→A), additionally increased the entropy of the side chain (F→G) or potentially introduced conformational strain (F→W). A combination of biophysical studies (thermal and chemical denaturation, proteolysis), HDX-MS and statistical mechanical calculations showed that even small local perturbations (e.g. F→L or F→V) have remarkable effects on the unfolding cooperativity and energetics. Interestingly, the mutation aimed to introduce conformational strain (F→W) is as damaging to conformational stability of hPGK1 as the mutation F→G, with severe effects on thermal/kinetic stability, denaturation energetics, unfolding cooperativity and local stability of the NTD. It is worth nothing that COSMIC mutations in hPGK1 (e.g. V216F and F241S)²⁷ as well as several mutations analyzed from inherited hPGK1 deficiency leading to changes in the size-chain or the conformational entropy of the protein backbone (e.g. L89P and S320N)²⁹ show effects on hPGK1 conformational landscape consistent with those reported here for the F25 mutants. We propose that our approach, combining biophysical analyses, HDX-MS studies and statistical mechanical calculations may allow to obtain high-resolution conformational information on the effects of cancer-associated mutations ²⁷, mutations associated with hPGK1 deficiency^3,29 and cancer-associated post-translational modifications ⁴⁴. We expect that similar analyses on mutations found in COSMIC (https://cancer.sanger.ac.uk/cosmic/gene/analysis?ln=PGK1_ENST00000644362), OMIM (https://omim.org/entry/311800) and gnomAD (https://gnomad.broadinstitute.org/gene/ENSG00000102144) databases will allow us to generate predictive tools to discern between pathogenic and neutral mutations ⁴³.

4.1. Protein expression and purification

hPGK1 WT was chemically synthesized by GenScript (Leiden, The Netherlands) and cloned into pET15b. The cloning strategy introduced in the N-terminal Met1 of the hPGK1-WT the following sequence: MGSSHHHHHHSSGLVPRGSH. Residues underlined constitute the His₆-tag used to purify the proteins by immobilized-metal affinity chromatography (IMAC) and the sequence in bold is the thrombin cleavage site. Mutants were generated by standard site-directed mutagenesis (GenScript, Leiden, The Netherlands). pET15b-hPGK1 plasmids were used to transform BL21(DE3) E.coli cells (Agilent Technologies, Santa Clara, CA, USA) for protein expression. For each variant, 40 mL of Luria-Bertani (LB) medium containing 0.1 mg·mL^− 1 ampicillin (Canvax Biotech, Córdoba, Spain) were inoculated with transformed cells and grown for 16 h at 37 ^oC. These cultures were diluted into 800 mL of LB containing 0.1 mg·mL^− 1 ampicillin, grown at 37 ^oC for 3 h to reach an optical density of about 1.0 and then these were transferred to 25 ^oC and induced with 0.5 mM IPTG (Isopropyl β-d-1-thiogalactopyranoside, Canvax Biotech, Córdoba, Spain). After 6 h, bacteria were harvested by centrifugation and frozen at -80 ^oC overnight. Cells were resuspended in binding buffer, BB (20 mM Na-phosphate, 300 mM NaCl, 50 mM imidazole, pH 7.4) plus 1 mM PMSF (phenylmethylsulfonyl fluoride, Sigma-Aldrich) and sonicated in an ice bath. These extracts were centrifuged (20000 g, 30 min, 4°C) and the supernatants were loaded into IMAC (Ni-Sepharose immobilized-metal affinity chromatography, from Cytiva, Barcelona, Spain) columns, washed with 40 bed volumes of BB and eluted in BB containing a final imidazole concentration of 500 mM. These eluates were treated with streptomycin 10% W/V to remove (poly)-nucleotides that may be bound to the hPGK1. After 30 min of centrifugation at 20000 g and 4 ^oC, supernatants were loaded into PD-10 columns (Cytiva, Barcelona, Spain) and buffer exchanged to the storage buffer (20 mM HEPES-NaOH, 200 mM NaCl, pH 7.4). These samples were flash-frozen in liquid N₂ and stored at -80 ^oC.

To remove protein aggregates and increase protein purity, proteins were subjected to size exclusion chromatography on a Superdex 200 16/60 column (GE Healthcare) using 20 mM HEPES-NaOH, 200 mM NaCl, pH 7.4 as mobile phase at 1.5 mL·min^− 1 flow rate. Monomeric fractions were concentrated and stored at -80 ºC after flash-freezing in liquid N₂.

UV-visible absorption spectra were recorded at 10–20 µM protein concentration in a Cary Eclipse 50 spectrophotometer using 1 cm pathlength cuvettes, using an extinction coefficient of e₂₈₀ = 27960 M^− 1·cm^− 1 except for F25W for which we used e₂₈₀ = 33460 M^− 1·cm^− 1.

4.2. Circular dichroism (CD) spectroscopy

CD spectra were registered in a Jasco J-710 spectropolarimeter with a Peltier element for temperature control. Before the acquisition of Far-UV and Near-UV/visible CD spectra, protein samples were buffer exchanged to 20 mM K-phosphate pH 7.4 using PD-10 columns (Cytiva, Barcelona, Spain). Spectra in the Far-UV range were acquired using 6–7 µM of protein in a 1 mm path-length quartz cuvette. These spectra were registered in the 200–260 nm range, at 25 ^oC, with a scan rate of 100 nm·min^− 1, a bandwidth of 1 nm and a time response of 2 s. Six spectra of each sample were acquired and averaged, and the corresponding blank (containing the same buffer but no protein) was acquired under the same conditions at the beginning and the end of each experimental series to assess for baseline drifts during the day, were averaged and subtracted from the spectra of protein samples. Similar conditions were used to acquire Near-UV/visible CD spectra with some modifications. Protein concentration for these experiments was 15 µM and 10 spectra were acquired and averaged, the wavelength range was 270–330 nm and the path-length of the cuvette was 5 mm. To calculate mean molar residue ellipticities [Q]_MRW, Eq. 1 was used:

${\left[{\Theta }\right]}_{MRW}=\frac{MRW·{\Theta }}{10·l·c}$(Eq. 1)

where MRW is the molecular weight of each hPGK1 variant divided by 416 (the residues of native hPGK1 minus 1), Q is the observed ellipticity of the sample in mdeg, l is the pathlength of the cuvette in cm and c the protein concentration in mg·mL^− 1.

4.3. Thermal denaturation by CD spectroscopy

For thermal denaturation experiments by CD spectroscopy, we used exactly the same samples used to acquire CD wavelength-dependent spectra. Samples were thermostatized at 20 ^oC for 10 min and then temperature was raised to 75 ^oC at 2 ^oC·min^− 1. Signals were acquired at 222 nm (for UV-CD in 1 mm path-length cuvettes) or 290 nm (for Near-UV CD in 5 mm path-length), using a bandwidth of 2 nm and a time response of 16 s. Raw data were normalized for linear pre- and post-transition baselines and the apparent T_m was determined as the temperature at which half of the unfolded signal was determined by linear extrapolation. Thermal denaturation leads to irreversible protein aggregation (i.e. formation of protein aggregates of large size).

4.4. Differential scanning calorimetry (DSC)

DSC analyses were carried out using a VP-DSC differential scanning calorimeter (Malvern Panalytical, Malvern, UK) with a cell volume of 137 µL. Experiments were carried out in 20 mM HEPES-NaOH, 200 mM NaCl pH 7.4 and using 7–15 µM protein. Temperature range was 5–75 ^oC and scan rates were 1, 2.5 or 4 ^oC·min^− 1. To investigate the effect of urea on hPGK1 denaturation, stock solutions of urea 5 M in 20 mM HEPES-NaOH, 200 mM NaCl pH 7.4 were diluted to the desired final concentration. All urea concentrations were checked by refractive index measurements.

The DSC traces were modeled using a two-state irreversible model ^28,38 (model 1), that describes very well the irreversible denaturation of hPGK1 ²⁸. In this model, denaturation is considered as a kinetically-controlled and irreversible first-order conversion of the native (N) to a final state (F) that cannot fold back:

$N \to F$ (Model 1)

This process is characterized by a strongly temperature-dependent first-order rate constant k. We also assume that the temperature dependence of k follows a slightly modified version of the Arrhenius equation (Eq. 2):

$\text{ln}k=ln{k}_{T*}- \frac{{E}_{a}}{R} (\frac{1}{T}-\frac{1}{T*})$ (Eq. 2)

where T is the absolute temperature, T* is a reference absolute temperature, E_a is the activation energy and R is the ideal gas constant (1.987 cal·mol^-1·K^-1).

The fitting function used to extract relevant energetic parameters: k, E_a, the enthalpy changes between N and F (ΔH) and the T_m (the maximum of the excess heat capacity change vs. T; usually close to the temperature at which k = 1 min^-1) uses two linear pre- and post-transition temperature-dependent baselines smoothly connecting N and F signals. This procedure allow to isolate the calorimetric peak defined by the term $-\varDelta H·\left(\frac{d{X}_{N}}{dT}\right).$ This peak shows the dependence of the excess heat capacity C_p^Exc from $\left(\frac{d{X}_{N}}{dT}\right)$ which is described by Eq. 3 ^28,38,45:

$\frac{{dX}_{N}}{dT}=-\frac{{E}_{a}}{R{T}_{m}^{2}}\text{exp}\left(\frac{{E}_{a}·\varDelta T}{R{T}_{m}^{2}}\right)·exp\left[-exp\left(\frac{{E}_{a}· \varDelta T}{R{T}_{m}^{2}}\right)\right]$ (Eq. 3)

From C_p^Exc vs. T profiles (i.e. after removing the chemical baselines), the value of the temperature-dependent rate constant k can be determined using Eq. 4 ⁴⁶:

$k= \frac{\nu ·{C}_{P}^{Exc}}{{\Delta }H-<H>}$(Eq. 4)

where ν is the scan rate in ^oC·min^− 1, ${C}_{p}^{Exc}$ is the excess heat capacity at a given T, ΔH is enthalpy change of the transition and < H > is the excess enthalpy at a given T. This equation allows to determine the T-dependence of k and thus yields the corresponding Arrhenius plots (Eq. 2).

To determine kinetic m values (m^≠) from DSC studies we followed the procedure described by ³⁸. Briefly, we carried out DSC experiments under the same conditions as described above but adding a final urea concentration in the range of 0.5–1.1 M. Then, m^≠ values were determined using Eq. 5:

${m}^{\ne }= -\frac{{E}_{a}}{{T}_{m}}\left(\frac{d{T}_{m}}{d\left[Urea\right]}\right)-R·{T}_{m}·\left(\frac{dln({E}_{a}/R{T}_{m}^{2}}{d\left[Urea\right]}\right)$ (Eq. 5)

where E_a, T_m, dT_m/d[Urea] can be readily calculated from fittings at different urea concentrations. Thus, for each protein variant, three different values of m^≠, one for each scan rate were analysed, and the data reported are the mean ± s.d. of these three values.

4.5. Isothermal urea denaturation monitored by CD spectroscopy

Urea was prepared at ~ 6 M in 20 mM HEPES-NaOH, 200 mM NaCl, 1 mM 2-mercaptoethanol (b-ME), pH 7.4 and its actual concentration was determined by refractive index measurements. Urea stock solutions were mixed with hPGK1 solutions in 20 mM HEPES-NaOH, 200 mM NaCl, 1 mM b-ME, pH 7.4 to a final concentration of 5 mM (in protein) and 0–5 M (in urea). Samples were incubated for at least 2 h and denaturation experiments were monitored by Far-UV CD spectroscopy similarly to those described in section 2.2. but restricting the acquisition range to 215–260 nm.

Data at 222 nm were used for fittings assuming a two-state model (apparently valid for WT hPGK1 using different conformational probes) ^{3, 27–29} to provide the values of experimental equilibrium m (m_urea) and C_m (concentration of urea at which half-denaturation occurs) using Eq. 6:

$S= \left[{S}_{N}+{m}_{N}·\left[urea\right]+\left({S}_{U}+{m}_{U}\right)·\left({exp}^{\frac{{m}_{urea}·\left(\left[urea\right]-{C}_{m}\right)}{R·T}}\right)\right]/[1+\left({exp}^{\frac{{m}_{urea}·\left(\left[urea\right]-{C}_{m}\right)}{R·T}}\right)]$ (Eq. 6)

where S is the experimental CD signal as a function of the [urea], S_N and S_U are the fitted CD signals for the Native and Unfolded state baselines at 0 M urea, respectively, m_N and m_U are the slopes of the native and unfolded state baselines, m_urea describes the unfolding cooperativity, R is the ideal gas constant and T is the experimental temperature (298.15 K). Although this model is likely an oversimplification, it provides a proxy for the cooperativity of reversible chemical unfolding of hPGK1 variants.

4.6. Proteolysis by thermolysin

Thermolysin from Geobacillus stearothermophilus (Sigma-Aldrich, Madrid, Spain) was prepared in 20 mM HEPES-NaOH, 200 mM NaCl, 100 mM CaCl₂, pH 7.4 and its concentration was determined spectrophotometrically using ε₂₈₀ = 66086 M^− 1 · cm^− 1 based on the protein primary sequence. Proteolysis was initiated after 5 min incubation at 25 ºC by mixing hPGK1 variants and thermolysin to a final concentrations of 10 µM and 0.5 or 1 µM, respectively. When the initial reaction was started contained 10 mM of CaCl₂. Aliquots were withdrawn at different times (2 min-96 h), mixed with 25 mM EDTA (final concentration), pH 8 and denatured with Laemmli´s buffer (1:1 in V/V) at 95 ºC for 5 min. Samples were resolved in SDS-PAGE gels (12% acrylamide) and analysed by densitometry using the software ImageJ (https://imagej.nih.gov/ij/).

The decay of the full-length protein (I) vs. time (t) was fitted using the following single exponential function to provide the first-order kinetic constant k_obs:

$$I={I}_{0}·{exp}^{-{k}_{obs}·t}$$

The half-life (t_1/2) for proteolysis was determined using: t_1/2 = ln 2 / k_obs. Errors corresponded to those from the fittings.

4.7. Hydrogen/Deuterium Exchange Mass Spectrometry (HDX-MS)

Amide hydrogen/deuterium exchange (HDX) of hPGK1 was studied for the WT and mutant variants F25A, F25G, F25L, F25V, F25W as described previously ⁴⁷ with some modifications. Briefly, 20 µM protein solution was 10 × diluted with a D₂O-based 20 mM HEPES-NaOH, 100 mM NaCl, 0.5 mM TCEP (tris(2-carboxyethyl)phosphine), pD 7.4, to start the exchange reaction. The exchange was terminated after 10 s, 30 s, 2 min, 5 min, 20 min, 1 h, 3 h 20 min and 20 h by mixing (1:1) with 0.5 M Glycine-HCl, pH 2.3 and the samples were flash frozen in liquid N₂. Time points 30 s, 5 min and 3 h 20 min were replicated. Custom-made nepenthesin-2 column was used for online proteolysis and the peptides were trapped and desalted on a SecurityGuard™ pre-column (ULTRA Cartridges UHPLC Fully Porous Polar C18, 2.1 mm, Phenomenex, Torrance, CA, USA). Solvent for this step − 0.4% formic acid (FA) in water - was pumped by 1260 Infinity II Quaternary pump (Agilent Technologies, Waldbronn, Germany) at a flow rate of 150 µl·min^− 1. Subsequent peptide separation was performed on an analytical column (LUNA® Omega Polar C18 Column, 100 Å, 1.6 µm, 100 mm X 1.0 mm, Phenomenex, Torrance, CA, USA) using linear gradient (5–45% B in 6 min) followed by a quick step to 99% B lasting 5 min pumped by 1290 Infinity II LC System (Agilent Technologies, Waldbronn, Germany) at a flow of 40 µl·min^− 1. Solvent A was 0.1% FA / 2% acetonitrile (ACN) in water, B was 0.1% FA / 98% ACN in water. Digestion, desalting, and separation were performed at 0°C and pH 2.3 to minimize deuterium back-exchange. LC system was directly coupled to an ESI source of 15T FT-ICR mass spectrometer (solariX XR, Bruker Daltonics, Bremen, Germany) operating in broad-band MS mode. Fully deuterated control was prepared for each PGK1 variant and was used to correct back-exchange levels as described previously ^47,48. MS data were exported using DataAnalysis 5.3 and processed by in-house developed program Deutex ⁴⁹. Peptide identification was done in a separate data-dependent LC-MS/MS run performed using the identical LC settings but connected to ESI-timsTOF Pro with PASEF. LC-MS/MS data were exported to mgf files using DataAnalysis 5.3 and searched by MASCOT (Matrix Science, London, UK) against a database containing hPGK1 sequences, nepenthesin-2 and cRAP contaminant sequences. Decoy search was enabled with false-discovery ratio < 1% and Ion Score cut-off 15.

4.8. Computational analyses of the F25 mutational effects

The Wako-Saitô-Muñoz-Eaton (WSME) model based perturbation analysis of hPGK1 was performed as discussed before for mutants of the protein NQO1 ⁵⁰. Briefly, an ensemble of 8,398,795 microstates is constructed employing the block-version of the WSME model - bWSME with a most probable block length of 5 consecutive residues - and the PDB structure 2XE7 ^51,52. The microstate energetics include stabilizing contributions from van der Waals interactions (heavy-atom interactions identified with a 5 Å cut-off with a mean-field interaction energy of -78 J mol^− 1 per heavy atom contact), all-to-all electrostatics at pH 7 protonation state, and simplified solvation free energy (defined by the heat capacity change of -0.36 J mol^− 1 K^− 1 per native contact). In addition, an entropic penalty of -14.5 J mol^− 1 K^− 1 per residue is introduced when fixing an unfolded residue to a folded conformation. The balance between the stabilization (free-)energy terms and entropic penalty determines the free energy of every microstate, the statistical weight and hence the associated probability ⁵¹.

Authorship contribution statement

J.L.P-G.: Protein expression and purification, biophysical studies and data analyses;D.S.L.: HDX-MS studies and data analyses;A.N.N.: Statistical mechanical calculations and data analyses; P.V.: HDX-MS studies and data analyses;M.C-M.: Data analyses; P.M.: Supervision of HDX-MS experiments and data analyses; A.L.P.: Conceptualization, biophysical studies and analyses, supervision of the project, funding acquisition and writing the initial draft. All authors prepared the final version of the manuscript.

Data Availability

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request (A.L.P.).

Funding

J.L.P-G. and A.L.P. were supported by the ERDF/Spanish Ministry of Science, Innovation and Universities—State Research Agency (Grant RTI2018-096246-B-I00, to ALP), Consejería de Economía, Conocimiento, Empresas y Universidad, Junta de Andalucía (Grant P18-RT-2413, to A.L.P.) and ERDF/ Counseling of Economic transformation, Industry, Knowledge and Universities (Grant B-BIO-84-UGR20, to A.L.P.) and the Science and Engineering Research Board (SERB, India; Grant MTR/2019/000392, to ANN). Financial support from Horizon 2020 EU_FT-ICR_MS project (731077), EU/MEYS projects BioCeV (CZ.1.05/1.1.00/02.0109) and CIISB LM2018127 are acknowledged.

Declaration of interest

None.

Rojas-Pirela, M. et al. Phosphoglycerate kinase: structural aspects and functions, with special emphasis on the enzyme from Kinetoplastea. Open Biol 10, 200302 (2020).
Chiarelli, L. R. et al. Molecular insights on pathogenic effects of mutations causing phosphoglycerate kinase deficiency. PLoS One 7, e32065 (2012).
Valentini, G., Maggi, M. & Pey, A. Protein Stability, Folding and Misfolding in Human PGK1 Deficiency. Biomolecules 3, 1030–1052 (2013).
Vas, M., Varga, A. & Gráczer, E. Insight into the mechanism of domain movements and their role in enzyme function: example of 3-phosphoglycerate kinase. Curr Protein Pept Sci 11, 118–47 (2010).
Szilágyi, A. N., Ghosh, M., Garman, E. & Vas, M. A 1.8 A resolution structure of pig muscle 3-phosphoglycerate kinase with bound MgADP and 3-phosphoglycerate in open conformation: new insight into the role of the nucleotide in domain closure. J Mol Biol 306, 499–511 (2001).
Marston, J. P. et al. Structural tightening and interdomain communication in the catalytic cycle of phosphoglycerate kinase. J Mol Biol 396, 345–60 (2010).
Freire, E., Murphy, K. P., Sanchez-Ruiz, J. M., Galisteo, M. L. & Privalov, P. L. The molecular basis of cooperativity in protein folding. Thermodynamic dissection of interdomain interactions in phosphoglycerate kinase. Biochemistry 31, 250–6 (1992).
Young, T. A., Skordalakes, E. & Marqusee, S. Comparison of proteolytic susceptibility in phosphoglycerate kinases from yeast and E. coli: modulation of conformational ensembles without altering structure or stability. J Mol Biol 368, 1438–47 (2007).
Feng, R., Gruebele, M. & Davis, C. M. Quantifying protein dynamics and stability in a living organism. Nat Commun 10, 1179 (2019).
Wirth, A. J., Platkov, M. & Gruebele, M. Temporal variation of a protein folding energy landscape in the cell. J Am Chem Soc 135, 19215–21 (2013).
Rajasekaran, N. & Naganathan, A. N. A self-consistent structural perturbation approach for determining the magnitude and extent of allosteric coupling in proteins. Biochem J 474, 2379–2388 (2017).
Rajasekaran, N., Sekhar, A. & Naganathan, A. N. A Universal Pattern in the Percolation and Dissipation of Protein Structural Perturbations. J Phys Chem Lett 8, 4779–4784 (2017).
Pacheco-Garcia, J. L. et al. Structural basis of the pleiotropic and specific phenotypic consequences of missense mutations in the multifunctional NAD(P)H:quinone oxidoreductase 1 and their pharmacological rescue. Redox Biology 46, 46:102112 (2021).
Qian, X. et al. Phosphoglycerate Kinase 1 Phosphorylates Beclin1 to Induce Autophagy. Mol Cell 65, 917–931.e6 (2017).
Ogino, T. et al. Involvement of a cellular glycolytic enzyme, phosphoglycerate kinase, in Sendai virus transcription. J Biol Chem 274, 35999–6008 (1999).
Popanda, O., Fox, G. & Thielmann, H. W. Modulation of DNA polymerases alpha, delta and epsilon by lactate dehydrogenase and 3-phosphoglycerate kinase. Biochim Biophys Acta 1397, 102–17 (1998).
Myre, M. A. & O’Day, D. H. Calmodulin binds to and inhibits the activity of phosphoglycerate kinase. Biochim Biophys Acta 1693, 177–83 (2004).
Guin, D. & Gruebele, M. Chaperones Hsc70 and Hsp70 Bind to the Protein PGK Differently inside Living Cells. J Phys Chem B 124, 3629–3635 (2020).
Gondeau, C. et al. Molecular basis for the lack of enantioselectivity of human 3-phosphoglycerate kinase. Nucleic Acids Res 36, 3620–9 (2008).
Varga, A., Lionne, C. & Roy, B. Intracellular Metabolism of Nucleoside/Nucleotide Analogues: a Bottleneck to Reach Active Drugs on HIV Reverse Transcriptase. Curr Drug Metab 17, 237–52 (2016).
Varga, A. et al. Nucleotide promiscuity of 3-phosphoglycerate kinase is in focus: implications for the design of better anti-HIV analogues. Mol Biosyst 7, 1863–73 (2011).
Varga, A. et al. Interaction of human 3-phosphoglycerate kinase with L-ADP, the mirror image of D-ADP. Biochem Biophys Res Commun 366, 994–1000 (2008).
Li, X. et al. Mitochondria-Translocated PGK1 Functions as a Protein Kinase to Coordinate Glycolysis and the TCA Cycle in Tumorigenesis. Mol Cell 61, 705–719 (2016).
Lin, C. Y. et al. Extracellular Pgk1 enhances neurite outgrowth of motoneurons through Nogo66/NgR-independent targeting of NogoA. Elife 8, e49175 (2019).
Schmidt, O., Pfanner, N. & Meisinger, C. Mitochondrial protein import: From proteomics to functional mechanisms. Nature Reviews Molecular Cell Biology 11, 655–667 (2010).
Li, X. et al. Nuclear PGK1 Alleviates ADP-Dependent Inhibition of CDC7 to Promote DNA Replication. Mol Cell 72, 650–660.e8 (2018).
Fiorillo, A. et al. The phosphoglycerate kinase 1 variants found in carcinoma cells display different catalytic activity and conformational stability compared to the native enzyme. PLoS One 13, e0199191 (2018).
Pey, A. L., Mesa-Torres, N., Chiarelli, L. R. & Valentini, G. Structural and energetic basis of protein kinetic destabilization in human phosphoglycerate kinase 1 deficiency. Biochemistry 52, 1160–1170 (2013).
Pey, A. L., Maggi, M. & Valentini, G. Insights into human phosphoglycerate kinase 1 deficiency as a conformational disease from biochemical, biophysical, and in vitro expression analyses. Journal of Inherited Metabolic Disease 37, 909–916 (2014).
Pey, A. L. The interplay between protein stability and dynamics in conformational diseases: The case of hPGK1 deficiency. Biochimica et Biophysica Acta - Proteins and Proteomics 1834, 2502–11 (2013).
Naganathan, A. N. Modulation of allosteric coupling by mutations: from protein dynamics and packing to altered native ensembles and function. Curr Opin Struct Biol 54, 1–9 (2019).
Kellis, J. T., Nyberg, K. & Fersht, A. R. Energetics of complementary side-chain packing in a protein hydrophobic core. Biochemistry 28, 4914–22 (1989).
Kellis, J. T., Nyberg, K., Sali, D. & Fersht, A. R. Contribution of hydrophobic interactions to protein stability. Nature 333, 784–6 (1988).
Eriksson, A. E. et al. Response of a protein structure to cavity-creating mutations and its relation to the hydrophobic effect. Science 255, 178–83 (1992).
Roche, J. et al. Effect of internal cavities on folding rates and routes revealed by real-time pressure-jump NMR spectroscopy. J Am Chem Soc 135, 14610–8 (2013).
Xue, M. et al. How internal cavities destabilize a protein. Proc Natl Acad Sci U S A 116, 21031–21036 (2019).
Pey, A. L. PH-dependent relationship between thermodynamic and kinetic stability in the denaturation of human phosphoglycerate kinase 1. Biochimie 103, 7–15 (2014).
Rodriguez-Larrea, D., Minning, S., Borchert, T. v & Sanchez-Ruiz, J. M. Role of solvation barriers in protein kinetic stability. J Mol Biol 360, 715–24 (2006).
Szilágyi, A. N. & Vas, M. Sequential domain refolding of pig muscle 3-phosphoglycerate kinase: kinetic analysis of reactivation. Fold Des 3, 565–75 (1998).
Naganathan, A. N. & Kannan, A. A hierarchy of coupling free energies underlie the thermodynamic and functional architecture of protein structures. Current Research in Structural Biology 3, 257–267 (2021).
Høie, M. H., Cagiada, M., Beck Frederiksen, A. H., Stein, A. & Lindorff-Larsen, K. Predicting and interpreting large-scale mutagenesis data using analyses of protein stability and conservation. Cell Reports 38, 110207 (2022).
Shendure, J. & Akey, J. M. The origins, determinants, and consequences of human mutations. Science 349, 1478–83 (2015).
Stein, A., Fowler, D. M., Hartmann-Petersen, R. & Lindorff-Larsen, K. Biophysical and Mechanistic Models for Disease-Causing Protein Variants. Trends Biochem Sci 44, 575–588 (2019).
Duncan, L., Shay, C. & Teng, Y. PGK1: An Essential Player in Modulating Tumor Metabolism. Methods Mol Biol 2343, 57–70 (2022).
Pey, A. L., Salido, E. & Sanchez-Ruiz, J. M. Role of low native state kinetic stability and interaction of partially unfolded states with molecular chaperones in the mitochondrial protein mistargeting associated with primary hyperoxaluria. Amino Acids 41, 1233–1245 (2011).
Sánchez-Ruiz, J. M., López-Lacomba, J. L., Cortijo, M. & Mateo, P. L. Differential scanning calorimetry of the irreversible thermal denaturation of thermolysin. Biochemistry 27, 1648–52 (1988).
Vankova, P., Salido, E., Timson, D. J., Man, P. & Pey, A. L. A Dynamic Core in Human NQO1 Controls the Functional and Stability Effects of Ligand Binding and Their Communication across the Enzyme Dimer. Biomolecules 9, 728 (2019).
Zhang, Z. & Smith, D. L. Determination of amide hydrogen exchange by mass spectrometry: A new tool for protein structure elucidation. Protein Science 2, 522–31 (1993).
Trcka, F. et al. The assembly and intermolecular properties of the Hsp70-Tomm34-Hsp90 molecular chaperone complex. J Biol Chem 289, 9887–901 (2014).
Pacheco-Garcia, J. L. et al. Allosteric Communication in the Multifunctional and Redox NQO1 Protein Studied by Cavity-Making Mutations. Antioxidants 11, 1110 (2022).
Gopi, S., Aranganathan, A. & Naganathan, A. N. Thermodynamics and folding landscapes of large proteins from a statistical mechanical model. Current Research in Structural Biology 1, 6–12 (2019).
Naganathan, A. N., Dani, R., Gopi, S., Aranganathan, A. & Narayan, A. Folding Intermediates, Heterogeneous Native Ensembles and Protein Function. Journal of Molecular Biology 433, 167325 (2021).
Zerrad, L. et al. A Spring-loaded Release Mechanism Regulates Domain Movement and Catalysis in Phosphoglycerate Kinase. Journal of Biological Chemistry 286, 14040–14048 (2011).
Myers, J. K., Pace, C. N. & Scholtz, J. M. Denaturant m values and heat capacity changes: relation to changes in accessible surface areas of protein unfolding. Protein Sci 4, 2138–48 (1995).

Tables 1-5 are available in the supplementary files section.

No competing interests reported.

Download PDF

Editorial decision: Major revision
13 Sep, 2022
Reviewers agreed at journal
29 Aug, 2022
Reviews received at journal
28 Aug, 2022
Reviewers agreed at journal
18 Aug, 2022
Reviewers invited by journal
18 Aug, 2022
Editor assigned by journal
10 Aug, 2022
Editor invited by journal
10 Aug, 2022
Submission checks completed at journal
10 Aug, 2022
First submitted to journal
09 Aug, 2022

You are reading this latest preprint version

Perturbation analysis in the N-terminal domain of hPGK1: a biophysical, structural and statistical mechanical study

Status:

Version 1

Abstract

Figures

1. Introduction

2. Results And Discussion

3. Conclusions

4. Material And Methods

4.1. Protein expression and purification

4.2. Circular dichroism (CD) spectroscopy

4.3. Thermal denaturation by CD spectroscopy

4.4. Differential scanning calorimetry (DSC)

4.5. Isothermal urea denaturation monitored by CD spectroscopy

4.6. Proteolysis by thermolysin

4.7. Hydrogen/Deuterium Exchange Mass Spectrometry (HDX-MS)

4.8. Computational analyses of the F25 mutational effects

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1