A Single Point Mutation in Mitochondrial Hsp70 Cochaperone Mge1

Dec 7, 2016 - School of Chemistry, University of Hyderabad, Hyderabad 500046, T.S., India. ABSTRACT: Mge1, a yeast homologue of Escherichia coli. GrpE...
1 downloads 0 Views 4MB Size
Article pubs.acs.org/biochemistry

A Single Point Mutation in Mitochondrial Hsp70 Cochaperone Mge1 Gains Thermal Stability and Resistance Adinarayana Marada,† Srinivasu Karri,† Swati Singh,‡ Praveen Kumar Allu,† Yerranna Boggula,† Thanuja Krishnamoorthy,† Lalitha Guruprasad,‡ and Naresh Babu V Sepuri*,† †

Department of Biochemistry, School of Life Sciences, University of Hyderabad, Hyderabad 500046, T.S., India School of Chemistry, University of Hyderabad, Hyderabad 500046, T.S., India



ABSTRACT: Mge1, a yeast homologue of Escherichia coli GrpE, is an evolutionarily conserved homodimeric nucleotide exchange factor of mitochondrial Hsp70. Temperaturedependent reversible structural alteration from a dimeric to a monomeric form is critical for Mge1 to act as a thermosensor. However, very limited information about the structural component or amino acid residue(s) that contributes to thermal sensing of Mge1/GrpE is available. In this report, we have identified a single point mutation, His167 to Leu (H167L), within the hinge region of Mge1 that confers thermal resistance to yeast. Circular dichroism, cross-linking, and refolding studies with recombinant proteins show that the Mge1 H167L mutant has increased thermal stability compared to that of wild-type Mge1 and also augments Hsp70-mediated protein refolding activity. While thermal denaturation studies suggest flexibility in the mutant, ionic quenching studies and limited proteolysis analysis reveal a relatively more rigid structure compared to that of the wild type. Intriguingly, Thermus thermophilus GrpE has a leucine at the corresponding position akin to the Mge1 mutant, and thermophilus proteins are wellknown for their rigidity and hydrophobicity. Taken together, our results show that the yeast Mge1 H167L mutant functionally and structurally mimics T. thermophilus GrpE.

T

sensor in Escherichia coli.7 In response to thermal stress, the active dimeric GrpE dissociates into a monomer and fails to interact with DnaK.8 The NEF family is represented by candidates from archaea, eubacteria, and eukaryotic related organelles such as chloroplasts and mitochondria. Mge1, a yeast mitochondrial homologue of E. coli GrpE, has also been shown to undergo a reversible structural alteration during thermal stress.8 The crystal structure of the E. coli GrpE−DnaK asymmetry complex reveals a GrpE homodimer attached to DnaK at its C-terminal end.9 Each GrpE molecule in the homodimer consists of an N-terminal α-helix domain, central two-helix bundle domains, and a C-terminal β-sheet domain. The β-sheet domain interacts with the nucleotide binding domain of DnaK and is responsible for decreasing the affinity of DnaK for ADP. The temperature-dependent structural transition in E. coli GrpE and yeast Mge1 is attributed to the unfolding of long N-terminal helices followed by destabilization of the β-sheet.10 However, in the case of Thermus thermophilus GrpE, melting of the β-sheet domain at permissive transition temperature induces the unwinding of N-terminal helices during thermal stress.7

he Hsp70 class of proteins consists of evolutionarily conserved ATP-dependent molecular chaperones. These members are involved in protein folding, prevention of protein aggregation, stress response, and remodeling of protein complexes.1,2 Hsp70 protein comprises a nucleotide binding N-terminal domain and a substrate binding C-terminal domain. These two domains are connected through a disordered loop.3 The binding and hydrolysis of ATP result in a conformational change in Hsp70 that modulates its affinity for the substrate. In the ATP-bound state, the affinity of Hsp70 for the substrate is low with fast on/off kinetics due to formation of a tense state (T state), whereas in the ADP-bound state, the affinity of Hsp70 for the substrate is high with slow on/off kinetics due to formation of a relaxed state (R state).4 Hsp70 has low intrinsic ATPase activity with slow ADP dissociation kinetics, and hence, it depends on cochaperones for stimulation of ATPase activity and recycling of ATP for ADP.5 Although individual components involved in the Hsp70 chaperone cycle have been well studied, the communication between these components still needs to be explored in detail. In general, thermal stress, a transient increase in temperature, induces the expression of chaperones and cochaperones.6 In addition to overexpression, temperature-dependent structural modulation or alterations in chaperones have been observed.4 GrpE, an evolutionarily conserved nucleotide exchange factor (NEF) of DnaK (Hsp70), has been identified as a thermal © XXXX American Chemical Society

Received: March 14, 2016 Revised: December 6, 2016 Published: December 7, 2016 A

DOI: 10.1021/acs.biochem.6b00232 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

of 50 nm/s). Three spectral scans were averaged after subtracting the baseline spectrum obtained using the buffer control. Thermal denaturation curves were obtained from 20 to 70 °C with a scan rate of 60 °C/h by monitoring ellipticity at 222 nm. The collected data were analyzed by plotting temperature against ellipticity at 222 nm, and Tm was calculated by fitting the data into two-state equations as described previously.8,18 Sequence Analyses and Structural Modeling of Mge1. The Mge1 protein sequence from Saccharomyces cerevisiae (CAA99452.1) was used as query in the NCBI database. The homologous sequences from mesophiles and thermophiles were obtained from BLASTP searches.19 Multiple-sequence alignment was performed using Clustal Omega,20 while phylogenetic trees were generated with MEGA 6.06 using the “Maximum Likelihood” method with the following parameters: uniform rates of site-specific mutation, partial deletion of gaps and missing data, and the nearest-neighbor interchange heuristic tree inference method. The LG amino acid substitution model was used as the best fit amino acid substitution model based on BIC (Bayesian information criterion) and AIC (Akaike information criterion) scores. The model with the lowest BIC and AIC values was thought to describe the best substitution pattern. Each phylogeny was evaluated using the bootstrap method with 500 replications. Multiple-sequence alignment and the phylogenetic tree were generated using Clustal Omega.20 The template structure for building the three-dimensional (3D) model of S. cerevisiae Mge1 was obtained using BLASTP searches19 against the Protein Data Bank.21 The homodimer model was built using MODELER22 that builds the protein structures by the satisfaction of spatial restraints. This method was incorporated into the Discovery Studio (D.S.) 2.5 suite of software. The reliability of the stereochemical geometry of the model was tested using PROCHECK.23 The template and model structures were superimposed using MAPSCI24 to examine the extent of similarity that is estimated by the root-meansquare deviation (RMSD) of Cα atoms of both proteins. In both chains A and B of Mge1, a single amino acid mutation, H167L, was incorporated using the “protein modeling” protocol and the “build mutation” module in D.S. 2.5, and the protein structures were energy-minimized. In Vitro Cross-Linking of Recombinant Proteins. The cross-linking assay was performed as described previously.13 Purified recombinant proteins were cross-linked using bis(sulfosuccinimidyl) suberate (BS3; Pierce) at room temperature in 20 mM sodium phosphate (pH 7.4), 150 mM NaCl, 20 mM HEPES, and 50 mM borate; 2.5 μg of purified proteins was preincubated at different temperatures (25, 37, and 42 °C) for 20 min, and BS3 was added to a final concentration of 0.1 mM. The reaction mixture was left for 30 min at the indicated temperature and the reaction quenched by the addition of 40 mM Tris-HCl (pH 7.5), and samples were were analyzed by sodium dodecyl sulfate−polyacrylamide gel electrophoresis (SDS−PAGE). ATPase Assay of mHsp70. The ATPase activity of mHsp70 was monitored as described previously.25 Recombinant mHsp70 protein (2.5 μg) was incubated with 7.5 μg of wild-type or Mge1 H167L mutant protein in an ATPase assay buffer [50 mM HEPES-KOH (pH 7.2), 5 mM MgCl2, and 100 mM KCl] containing 50 nM denatured G6PDH and 1 μg of Mdj1. A control assay was also included wherein mHsp70 protein was incubated in a buffer in the absence of Mge1

The role of the N-terminal region (33-residue disordered region) remains controversial.9 It has been shown that the formation of a four-helix bundle from the C-terminal but not the long N-terminal α-helical tail is important for the stability of the dimer as GrpE constructs (1−138 and 88−197) without the N-terminal tail domain are able to form a dimer.11 However, the long N-terminal helix appears to be important for the thermal sensor property as it helps in thermal adaptation by shifting the DnaK−substrate complex toward the high-affinity R state.4 The R state is reduced in E. coli that expresses the R40C mutant of GrpE.4 In addition, a recent report suggests that disordered long pair helices mimic substrates and compete for the DnaK substrate binding site.12 These observations suggest that both N- and C-terminal domains are important in the regulation of Hsp70 function and thermal sensing. In this report, we show that the Mge1 H167L mutant exhibits thermal stability when compared to the wild type. Mutant Mge1 H167L shows resistance toward thermal stress by enhancing the ATPase and refolding activities of Hsp70. Biochemical and biophysical studies show that the Mge1 mutant attains an altered conformation. Our studies show that Mge1 serves as a model protein to illustrate structural divergence and evolutionary process toward adaptation.



MATERIALS AND METHODS Construction of Mge1 Mutant Strains. For in vivo studies, we used haploid strain yNB65 wherein the genomic deletion of Mge1 is complemented by ectopic expression of Mge1 from a pTEF-2μURA-Mge1 plasmid as previously described.13 The high-copy number URA3 plasmid in the yNB65 strain was replaced with either pNB186 (WT MGE1pTEF-2μLEU) or pNB189 (MGE1-H167L-pTEF-2μLEU) by plasmid shuffling to generate the yNB67 (wild-type Mge1) or yNB70 (Mge1 H167L) mutant strain, respectively. These strains were treated with 5-FOA for two generations to evict the URA3 plasmid, and the loss was confirmed by a lack of growth on the SC-Ura plate. Yeast Media. Standard yeast protocols were used for culturing yeast strains. Wild-type yeast strains were grown in YPD that contained 1% yeast extract, 2% peptone, and 2% dextrose or in synthetic dextrose (SD) minimal medium that contained 0.73% yeast nitrogen base with amino acids, 0.4% ammonium sulfate, and 2% glucose. For isolation of mitochondria, semisynthetic lactose medium containing 2% lactose with yeast extract, ammonium chloride, calcium chloride, magnesium chloride, potassium phosphate, and glucose was used as described previously.14,15 Yeast cells were grown at 30 °C. Yeast transformations were performed by following the standard lithium acetate method.16 Protein Expression and Purification. Wild-type Mge1, mutant H167L, Mdj1, and mtHsp70 were expressed in E. coli and purified as described previously.13,17 Briefly, BL21 cells were transformed with plasmids encoding wild-type MGE1 and mutant MGE1. For the expression of recombinant proteins, bacterial cultures were induced with 1 mM IPTG (isopropyl thiogalactoside) and recombinant proteins purified using NiNTA beads in phosphate buffer (pH 7.4). Circular Dichroism Spectroscopy Analysis. Circular dichroism (CD) spectra in the far-ultraviolet region (190− 250 nm) for wild-type and mutant Mge1 proteins (5 μM) in phosphate buffer were recorded using a circular dichroic spectrometer (Jasco J-810) at different temperatures ranging from 25 to 70 °C (2 mm path length quartz cell, response time B

DOI: 10.1021/acs.biochem.6b00232 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry protein. Reactions were initiated by addition of 0.05 mM [γ-32P]ATP (3000 Ci/mmol) followed by incubation at 25 °C. The reactions were quenched at different intervals by addition of 4 M formic acid, 2 M LiCl, and 36 mM ATP. All the reactions were analyzed by autoradiography as described previously.25 Glucose-6-phosphate Dehydrogenase Refolding Assay. Glucose-6-phosphate dehydrogenase (G6PDH; Sigma) at 25 μM was chemically denatured by being incubated with 6 M GdmCl containing 50 mM Tris-HCl (pH 7.5), 5 mM MgCl2, and 2 mM DTT buffer for 24 h.26 Refolding of G6PDH was performed as described previously.26 Briefly, denatured G6PDH was diluted to 250 nM in a buffer containing 10 mM ATP and proteins Ssc1 (3.5 μM), Mdj1 (1 μM), and Mge1 (1 μM) and incubated for various amounts of time. All the proteins used in this assay except G6PDH were recombinant proteins that were purified as described previously.17 The refolding kinetics of G6PDH was assessed by measuring the enzyme activity spectrophotometrically at 340 nm in a reaction mixture containing 5 nM G6PDG, 100 mM Tris-HCl (pH 7.5), 3 mM glucose 6-phosphate, and 150 μM NAD. Reduction of NAD was monitored and enzyme activity calculated as described previously.27 Quenching Analysis. Quenching studies of wild-type and mutant purified Mge1 were conducted by recording the intrinsic tyrosine fluorescence using a fluorescence spectrophotometer (FluoroMax-3) in the range of 300−400 nm using an excitation wavelength of 275 nm, with 5.0 nm excitation and emission slit widths. The assay samples contained 10 μM purified protein in 100 mM phosphate-buffered saline (pH 7.2). Potassium iodide was added to the assay after the initial fluorescence scan as an ionic quencher, while acrylamide was used as a neutral quencher. The fluorescence scans were once again monitored. A baseline scan was obtained using buffer as a control. Three scans of each spectrum were averaged and normalized using a baseline scan. Stern−Volmer constants were obtained by Augusteyn et al.28 Surface Hydrophobicity Studies. A surface hydrophobicity probe, 4,49-dianilino-1,19-binaphthyl-5,59-disulfonate (bis-ANS), was used to monitor the surface hydrophobicity by obtaining the spectra in the range of 400−600 nm with excitation at 390 nm and 2.5 nm slit widths for both excitation and emission. Spectra were recorded by adding 30 μM bis-ANS to protein (200 μg/mL) as described previously.29 Limited Proteolysis Assay. Purified proteins, both wildtype Mge1 and mutant H167L (1 mg/mL), in phosphatebuffered saline (pH 7.4) were incubated with trypsin at ratio of 1:100 (w/w) at 25 °C for different amounts of time. The reactions were stopped with addition of 2× SDS sample buffer.30 Samples from three independent experiments were resolved via SDS−PAGE and proteins quantified by densitometry using ImageJ (National Institutes of Health, Bethesda, MD).

Figure 1. Thermal unfolding of Mge1 and the Mge1 H167L mutant. (A) Position of the histidine amino acid residue in the hinge region of Mge1 in the modeled structure. (B) Recombinant purified wild-type, Mge1 H167L mutant, Hsp70, and Mdj1 proteins were separated via SDS−PAGE and Coomassie-stained. (C) Secondary structure spectra of wild-type Mge1 and its H167L mutant were scanned in the far-UV range using a CD spectrometer. (D) Secondary structure content of wild-type and Mge1 H167L proteins that were scanned via far-UV CD spectra from 20 to 70 °C. (E) CD signal at 222 nm recorded for wild type and Mge1 H167L mutant proteins at the indicated temperatures. (F) Cross-linking of wild-type and Mge1 H167L proteins (2.5 μg) was performed as described in Materials and Methods at different temperatures (25, 37, and 42 °C). The cross-linking products were resolved via 10% SDS−PAGE and stained with Coomassie Brilliant Blue. (G) Dimer:monomer ratios of cross-linked Mge1 protein quantified from panel F using ImageJ.

GrpE has been implicated in allosteric communication between its N-terminal helix and C-terminal β-sheet that is required for efficient interaction with DnaK.4,31 On the basis of the location of H167 and it being highly conserved, we hypothesized that H167 is probably crucial for thermal sensing. To understand the importance of histidine at position 167 in thermal sensing and its influence on the ATPase and refolding kinetics of Hsp70, we expressed and purified recombinant wild-type Mge1, Mge1 H167L mutant, Hsp70, and Mdj1 proteins in bacteria as described in Materials and Methods and as shown (Figure 1B). The single point mutation of H167L did not alter the secondary structure, which is evident from the observations made using far-UV CD spectrometry (Figure 1C). Gradual increases in temperature result in unfolding of proteins due to a conformational change in its structure. Thermally induced unfolding provides valuable information about the stability of a protein. Despite the H167L mutation in Mge1 having no effect on its secondary structure, we assessed the thermally induced unfolding of recombinant wild-type and H167L mutant proteins in vitro. We exposed both wild-type and Mge1 H167L mutant proteins to an increasing temperature (from 20 to 70 °C) while simultaneously recording a series of CD spectra (Figure 1D). The mean residual ellipticity at 222 nm was calculated and plotted for wild-type and mutant proteins as a function of an increasing temperature (Figure 1E). It is evident from the spectra that the isodichroic points of both wild-type and Mge1 mutant H167L proteins are at 203 nm, and this does not change with an increasing temperature. This reflects a transition from the folded dimeric state to an unfolded



RESULTS The Mge1 H167L Mutant Is Resistant to Thermal Stress. In a mutational screen for oxidative stress resistant variants of Mge1,13 we found that a histidine to leucine substitution at position 167 of Mge1 exhibits thermal resistance. Multiple-sequence alignment of GrpE/Mge1 from different organisms revealed that the histidine amino acid at position 167 is highly conserved, and curiously, it is found to be located in the hinge region (Figure 1A). The hinge region in C

DOI: 10.1021/acs.biochem.6b00232 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry monomeric state. The spectrum obtained at lower temperatures (20−30 °C) mainly represents the folded structure as most of the secondary structure is retained in both wild-type and mutant proteins. An increase in temperature from 35 to 70 °C causes a substantial loss of secondary structure of wild-type Mge1 protein when compared to that of Mge1 H167L. Mean residual ellipticity curves indicate that wild-type Mge1 has a clear cooperative thermal transition with a midpoint at 310 K (37 °C). In contrast, Mge1 H167L exhibits noncooperative thermal unfolding with the midpoint shifted to 314 K (41 °C). The fitting equation was used to calculate the Tm value as described previously.8 Consistent with the midpoint transition temperatures, wild-type Mge1 denatures at ∼310 K whereas the Tm value for the mutant Mge1 protein is ∼314 K. Like E. coli GrpE,32 Mge1 also exists as a dimer in solution.33 To investigate the thermal stress on oligomeric states of wild type and Mge1 H167L proteins, we incubated purified recombinant wild-type and mutant Mge1 proteins at 25, 37, and 42 °C for 20 min followed by cross-linking with the BS3 cross-linker. The samples were resolved via SDS−PAGE, and the gel was Coomassie-stained. In the absence of a cross-linking reagent, wild-type Mge1 migrates as a monomer with a molecular weight of 27 kDa. However, in the presence of BS3, Mge1 migrates as an ∼60 kDa homodimer. In addition, there is a direct correlation between the increase in temperature and the amount of monomer formation. Dimeric Mge1 is completely dissociated into a monomer at 42 °C even in the presence of a cross-linker (Figure 1F). These results suggest that thermal stress causes the disordering of secondary structure followed by dissociation of dimeric Mge1 into monomers. In the presence of a cross-linker, Mge1 H167L exists as a homodimer with a molecular weight of ∼60 kDa. However, unlike wild-type Mge1, Mge1 H167L is partially resistant to thermally induced dissociation of the dimeric form into a monomer at 42 °C (Figure 1F). Quantification of the dimeric to monomeric ratio of wild-type and Mge1 H167L mutant proteins at different temperatures shows that the mutant retains a significant quantity of the more dimeric form compared to the amount of wild-type Mge1 at 42 °C (Figure 1G). This result is consistent with the increase in thermal stability reflected by the shift in the midpoint temperature of denaturation. Taken together, the aforementioned results show that the secondary structure of Mge1 impacts its thermal stability in vitro and most importantly that the H167 amino acid within its hinge region is crucial for thermal regulation of Mge1 in vitro. The Mge1 H167L Strain Shows Thermal Resistance under Nonpermissive Temperatures. To understand the in vivo significance of Mge1 in thermal stress, we generated chromosomal Mge1-deleted haploid yeast strains yNB67 and yNB70 expressing wild-type Mge1 or Mge1 H167L from highcopy number plasmids, pNB186 and pNB189, respectively. We compared the phenotype of the parent BY4741 transformed with the vector with those of its derivative strains, yNB67 and yNB70, on YPD plates. Exponentially growing yeast cells from all three strains were subjected to heat shock for 30 min at 42 °C, and 10 μL of serially diluted cells was spotted on YPD plates followed by incubation at different temperatures (30 and 37 °C). All three strains exhibited comparable growth on YPD plates at 30 °C without heat shock (Figure 2A). However, wildtype Mge1-expressing strains BY4741 and yNB67 exhibited diminished levels of growth at 37 °C when they were subjected to without or with heat shock at 42 °C prior to being plated

Figure 2. Mge1 H167L mutant strain that shows resistance to thermal stress. (A−D) Yeast strains BY4741 transformed with vector, yNB67 (expressing MGE1), and yNB70 (expressing MGE1 H167L) were grown overnight in YPD medium. One OD of cells that were kept at room temperature or 37 °C or heat shocked at 42 °C for 30 min were serially diluted 10-fold; 10 μL of each suspension was spotted on YPD plates, and the plates were incubated at 30 or 37 °C for 2 days.

(Figure 2B,C). In contrast, we find that under the same conditions, the yNB70 strain expressing Mge1 H167L exhibits growth significantly better than that of BY4741 or yNB67. These observations clearly suggest that the H167L mutation within the hinge region of Mge1 protein has a positive effect on cellular adaptation to thermal stress. The Mge1 H167L Mutant Influences Hsp70 Function in Vitro. Thermal stress induces structural alteration in Mge1 and affects its interaction with Hsp70 and ATPase activity.8 As Mge1 H167L shows structural stability at higher temperatures, we investigated if there was any effect on Hsp70 or its ATPase activity. We studied the steady state ATPase activity of Hsp70 in the presence of equivalent amounts of purified recombinant Mge1 or Mge1 H167L proteins as described in Materials and Methods. Wild-type and Mge1 H167L proteins were incubated with Hsp70 as described in Materials and Methods to quantify the steady state ATPase activity (Figure 3A). The Mge1 H167L mutant shows moderately enhanced ATPase activity compared to that of wild-type Mge1. To examine if the Mge1 H1671 mutant has any physiological effect on Hsp70 functions, we studied the refolding kinetics of glucose-6-phosphate dehydrogenase (G6PDH) that is mediated by Hsp70 in the presence of an equimolar concentration of either wild-type or Mge1 H167L proteins. We used the wellestablished refolding assay wherein chemically denatured heat stable G6PDH is used as a substrate and quantified spontaneous refolding by measuring its enzymatic activity spectrophotometrically as described in Materials and Methods. The percent of G6PDH reactivation in the absence of chaperones was set as a control, and it was observed to be