TY - JOUR
AU - Lee, Soo Jae
AB - Abstract The scaffolding protein Salvador (Sav) plays a key role in the Hippo (Hpo) signalling pathway, which controls tissue growth by inhibiting cell proliferation and promoting apoptosis. Dysregulation of the Hippo pathway contributes to cancer development. Since the identification of the first Sav gene in 2002, very little is known regarding the molecular basis of Sav-SARAH mediating interactions due to its insolubility. In this study, refolding of the first Sav (known as WW45)-SARAH provided insight into the biochemical and biophysical properties, indicating that WW45-SARAH exhibits properties of a disordered protein, when the domain was refolded at a neutral pH. Interestingly, WW45-SARAH shows folded and rigid conformations relative to the decrease in pH. Further, diffracting crystals were obtained from protein refolded under acidic pH, suggesting that the refolded WW45 protein at low pH has a homogeneous and stable conformation. A comparative analysis of molecular properties found that the acidic-stable fold of WW45-SARAH enhances a heterotypic interaction with Mst2-SARAH. In addition, using an Mst2 mutation that disrupts homotypic dimerization, we showed that the monomeric Mst2-SARAH domain could form a stable complex of 1:1 stoichiometric ratio with WW45 refolded under acidic pH. Hippo pathway, low-pH stabilization, refolding, SARAH domain, WW45 Drosophila Salvador (Sav) is a core component of the Hippo (Hpo) signalling pathway that plays a pivotal role in organ size control and tumour suppression (1–4). This highly conserved pathway, discovered in Drosophila melanogaster, contains a core kinase cascade in which Hpo recruits Sav as a scaffolding molecule to trigger cell cycle arrest and apoptosis through the phosphorylation and activation of Warts (3, 5, 6). The human homologue of Sav (known as WW45), Mst1/Mst2 and RASSF1 share a common interaction motif known as SARAH, mediating homo/hetero-dimerization that are core processes of Hpo signalling transmission (7, 8). WW45 binds Mst1/Mst2 directly via the SARAH domain (9). Also, Mst1/2 forms a SARAH domain-mediated homodimer and a heterodimer with RASSF-SARAH domain (10, 11). Recently, some information is available on the structures of the SARAH domains, showing that the SARAH domains form a homodimer/heterodimer by a head-to-tail interaction (11–13). It has been reported that WW45 binds to the SARAH domain of RASSF1A and acts as a tumour suppressor effector of RASSF1A with Hippo pathway-independent functions. Knockout mouse studies have implicated WW45 as a potential tumour suppressor in mammalian systems (1, 14). However, WW45-mediated Hippo signalling and tumourigenesis are not fully understood, as structural studies are limited. Recently, Ni et al. (11) have identified important roles of RASSF5s in regulating Mst1/2 in vitro at the molecular level. The complex structure shows that Mst2 undergoes autoactivation through phosphorylation and forms a heterodimer with RASSF5 via SARAH interaction, which disrupts Mst2 homodimer and blocks Mst2 autoactivation (13). However, its applicability to the structural information for WW45-SARAH is limited, as the perturbed signals of the recombinant WW45-SARAH differ from the Rassf5 when in complex formation with 15N-labelled Mst1 (10,12). Although the mechanism for activating the Mst1/2-WW45 complex has yet to be defined, Mst2 binding to WW45 leads to the signal transduction through the Hippo pathway (15). For understanding of molecular mechanism in the Hippo signalling, it is necessary to perform molecular analysis for the Mst-WW45 homo/hetero-dimerization via SARAH. Recombinant WW45-SARAH domain was overexpressed in Escherichia coli (E. coli) inclusion bodies compared to other SARAH domains. Therefore, it is important to understand how WW45-SARAH could be solubilized and refolded from inclusion bodies to obtain a properly folded protein for molecular studies. This study focuses on increasing overall solubility and stability of the WW45-SARAH protein. Our data revealed that a refolded protein is partially unfolded in a neutral pH (pH 7.5), although soluble WW45-SARAH was successfully obtained by Triton X-100. We observed that the refolded WW45-SARAH undergoes pH-dependent conformational change and forms a stable complex with Mst2 SARAH in an acidic pH. Also, the diffracting crystals were obtained from WW45-SARAH protein refolded in an acidic pH, implying that low pH-driven folding of WW45-SARAH is stable and homogeneous. Here we reported the results of a comparative analysis of molecular and functional properties of WW45-SARAH. Gel-filtration analysis at a low pH and SDS–PAGE trace of Mst2-WW45 SARAH complex showed a 1:1 stoichiometry of an Mst2-WW45 complex. Mutagenesis studies based on the crystal structure of Mst2-SARAH revealed that monomeric Mst2-SARAH domain enhances a stable binding with WW45 refolded under acidic pH. These results provide a starting point for a mechanistic understanding of how WW45 mediates Hippo signalling. Materials and Methods Cloning, expression and purification Human WW45-SARAH (residues 314–383), referred to as WW45-SARAH forward, was cloned into bacterial expression vector pET28a (Novagen) modified to a TEV cleavage site and overexpressed in E. coli BL21(DE3). Escherichia coli cells were cultured in Luria Broth medium (LB) and 50 μg/ml kanamycin at 37°C until OD600 reached 0.5–0.6. Target protein was then induced by incubating cells with 0.1 mM IPTG at 37°C for 6 h. Cells were harvested by centrifugation at 6,000g for 15 min at 4°C, washed in buffer A (50 mM Tris–HCl pH 7.5, containing 5% glycerol, 0.5 M NaCl, 30 mM imidazole) and resuspended in the same buffer containing 1% Triton X-100 (USB Co.) and 0.1 mM PMSF. Cells were sonicated and collected by centrifugation at 18,000g for 1 h at 4°C. This step was repeated twice to fully extract the targeted protein from pellets. The supernatant was incubated for 20 min with Ni-NTA agarose (Qiagen, USA), and then the bead was extensively washed with buffer A containing 0.2% Triton X-100. The fusion proteins were eluted with buffer B (50 mM Tris–HCl pH 7.5, containing 5% glycerol, 0.5 M NaCl, 300 mM imidazole, and 0.2% Triton X-100) and cleaved overnight with TEV protease at 4°C. His-tag removed WW45-SARAH was concentrated to 2 ml using a 5 kDa cutoff membrane (Amicon Ultra-15 filter; Millipore). For further purification and refolding, size-exclusion chromatography (SEC) was performed twice using HiLoad16/60 Superdex 200 (GE Healthcare, USA) pre-equilibrated in citrate buffer, 20 mM sodium citrate pH 3.5, 50 mM NaCl, 1 mM DTT, and 10% glycerol. Purified proteins were frozen in liquid nitrogen and stored at −80°C. Purification of human Mst2-SARAH was carried out, as described in a previous study (16). The SARAH complex of Mst2 and WW45 was formed by mixing equal amounts of purified proteins. Unbound proteins were removed by gel filtration on Superdex 200 HiLoad 16/60. An analytical 10/300 Superdex 200 column was equilibrated with 20 mM Tris–HCl (pH 7.5), 50 mM NaCl, 1 mM DTT, and 10% glycerol at 4°C. Fractions containing the complex were analysed by SDS–PAGE. Point mutants were produced by PCR-based mutagenesis using the Muta-Direct Site Directed Mutagenesis Kit (Intron, Korea). All plasmid inserts were verified by DNA sequencing (Solgent Co., Korea). A calibration curve was obtained using the molecular weight calibration kits (GE Healthcare, USA) on Superdex 200 HiLoad 16/60. The size standards used were β-amylase (200 kDa), alcohol dehydrogenase (150 kDa), albumin (66 kDa), carbonic anhydrase (29 kDa), and cytochrome C (12.4 kDa). CD spectroscopy CD spectra were measured at 25°C with a J815 spectrometer using 1 mm light path cuvettes (Jasco). Purified proteins were analysed at a concentration of 0.1 mg/ml in various pH buffers. Wavelength scans were monitored from 250 to 190 nm with an average of five measurements. Secondary structure content was determined using K2D2, an algorithm-based neural network online software (17). All experiments were performed in triplicate with the proteins prepared on different dates. Dynamic light scattering The dynamic light scattering (DLS) experiment was realized with a protein concentration of 1.0 mg/ml in different buffers using DynaPro NanoStar dynamic light scattering instrument (Wyatt Technology, USA) at 25°C; 20 mM Tris–HCl pH 7.5, 20 mM Bis–Tris pH 6.5, 20 mM Bis–Tris pH 5.5, 20 mM sodium acetate pH 4.5, 20 mM sodium citrate pH 3.5. Samples were previously filtered through a 0.22 μm pore size filter (Nalgene). Each measure is the mean of 10 runs and was repeated three times. Crystallization and preliminary diffraction studies of WW45-SARAH Crystallization of WW45-SARAH was screened at 25°C by the hanging-drop vapor diffusion technique using a PEG screening kit designed for crystallizing the protein-protein complex from our laboratory. Protein solution (1 μl) was mixed with a reservoir solution (1 μl) and equilibrated against 400 μl reservoir solution. A24-well plate (Falcon, USA) was sealed with siliconized cover slides (MARIENFELD, Germany) and silicon grease, then stored at 20°C. Crystals were obtained in 0.1 M sodium citrate pH 3.2, 17% PEG400, 15% 2-propanolcondition. The presence of WW45-SARAH in crystals was confirmed by SDS–PAGE of harvested crystals. Diffraction was tested using a Rayonix MX-225HE CCD detector, supported by NSRRC on the BL44XU beamline at SPring-8 (Harima, Japan). Pull-down binding assay Equal amounts of GST fusion Mst2 SARAH wild-type (wt) and mutants were immobilized in 20 μl of GST resin. Resin was incubated with 50 μg of WW45-SARAH in binding buffer 20 mM Tris–HCl pH 7.5, 300 mM NaCl, and 10 mM β-mercaptoethanol, 5% glycerol for 1 h at 4°C. After washing a minimum of 3 times with binding buffer, bound proteins were eluted by boiling for 1 min with SDS–PAGE sample buffer and resolved by 20% SDS–PAGE. Results and Discussion Low pH-dependent folding of WW45-SARAH To obtain soluble WW45-SARAH for molecular study, a two-step refolding procedure was established. His6-tagged WW45-SARAH was overexpressed in E. coli cells and extracted from inclusion bodies by Triton X-100. Soluble protein was bound to Ni-NTA resin and eluted with a buffer containing imidazole. For further purification, the Superdex 200 16/60 column was used twice as an additional chromatography step. SDS–PAGE analysis of samples from the final step revealed that the protein had >95% purity (Fig. 1A). Fig. 1 View largeDownload slide CD spectra of the purified WW45-SARAH at various pHs. (A) Purification of WW45-SARAH. Fractions from the final gel-filtration step with a Superdex 200 were analysed by SDS–PAGE. Gel was stained using Coomassie Brilliant Blue R250. (B) pH dependence of far-UV CD spectra of WW45-SARAH. Data were monitored at 25°C. Spectral contribution from the buffer was subtracted from protein spectra. Fig. 1 View largeDownload slide CD spectra of the purified WW45-SARAH at various pHs. (A) Purification of WW45-SARAH. Fractions from the final gel-filtration step with a Superdex 200 were analysed by SDS–PAGE. Gel was stained using Coomassie Brilliant Blue R250. (B) pH dependence of far-UV CD spectra of WW45-SARAH. Data were monitored at 25°C. Spectral contribution from the buffer was subtracted from protein spectra. Far-UV CD spectra were used for analysing the secondary structure content of a protein by K2D2; algorithm-based neural network online software. A comparative change in far-UV CD spectra of WW45-SARAH at various pHs is shown in Fig. 1B. Secondary structure contents are presented in Table I. In a neutral pH (pH 7.5), WW45-SARAH appears to be a highly unfolded protein with a secondary structure content of 37% helices and 45% random coil (Table I). Surprisingly, a decreased pH induces a drastic change in the spectrum shape (Fig. 1B). At pH 3.5, the CD signal of WW45-SARAH is typical of alpha-helices with ∼98% alpha-helix content, which displays double minima at 208 and 222 nm. WW45-SARAH is known as an alpha-helical coiled coil motif (9). Structural studies have shown conserved features of the SARAH domain that form a bundle of two helices with short coiled coil motif (11–13). Dutta et al. (18) reported that folding of coiled coil motifs, which is natively unfolded in physiological pH and temperature, can often be induced by changes in environmental conditions such as low pH and low temperature. In many cases the folding and stability of proteins is strongly related to the protonation state of acid-based amino acids. The pKa values of these residues in folded proteins can be strongly influenced by the local environment (19). Extracellular pH is substantially and consistently lower in physiological and pathological situations including inflammation response against pathogens, autoimmune diseases and tumours (20). The correlation between changes in a cellular and subcellular pH environment is not fully understood. However, alterations in locally environmental pH affect protein folding, presumably which can couple with at least partial protonation of the acidic residues to change. Our data revealed that decrease in pH can induce the helical folding of WW45-SARAH. We have carefully speculated that low pH-driven homodimeric state of WW45-SARAH might be increasingly favored on interaction with other SARAHs in a locally reduced pH environment, due to intracellular events. Absolutely, further studies and evidences are needed to clarify this speculation. Table I. Secondary structure content of WW45-SARAH analysed using the K2D2 method Conditions  α-Helix (%)  β-sheet (%)  Random coil (%)  pH 7.5  37.18  17.48  45.18  pH 6.5  69.46  1.85  4.9  pH 5.5  94.89  0.05  5.06  pH 4.5  95.03  0.07  4.9  pH 3.5  99.8  0.02  0.18  Conditions  α-Helix (%)  β-sheet (%)  Random coil (%)  pH 7.5  37.18  17.48  45.18  pH 6.5  69.46  1.85  4.9  pH 5.5  94.89  0.05  5.06  pH 4.5  95.03  0.07  4.9  pH 3.5  99.8  0.02  0.18  View Large Low pH-induced alterations in molecular topology- sufficient for crystallization Size exclusion chromatography (SEC) was performed to explore whether oligomeric behaviors of WW45-SARAH would be altered by pH. A single peak for 80 ml was eluted at pH 3.5 with an apparent molecular weight (MW) of 37 kDa: ∼4-fold higher than MW of a monomer (8.6 kDa) calculated from the amino acid sequence (Fig. 2A). A peak for 75 ml eluted at pH 7.5 revealed higher MW of ∼53 kDa, implying that oligomer formation is mediated (Fig. 2B). It seems that WW45-SARAH presents a homodimer with MW 37 kDa in pH 3.5 (calculated MW of a homodimer, ∼17.2 kDa) on the basis of Mst2-SARAH, which has an approximate 2-fold higher MW than calculated from the amino acid sequence (Fig. 2A and B); 17 kDa in pH 7.5 as a homodimer (calculated MW, 11.2 kDa) and 9 kDa in pH 3.5 as a monomer (calculated MW, 5.6 kDa). The relationship between elution volume and MW is shown in Table II. Most proteins tend to be globular, but those containing alpha-helical coiled-coil motifs commonly form extended rods. An SEC column contains porous beads and the movement of a protein in and out of small pores during a column running is influenced by the size and shape of a protein. Relative to a globular protein of the same mass, a rod-shaped protein will appear bigger by SEC because the protein explores less volume and elutes earlier (21, 22). Taken together, SEC analyses suggest that the decreasing buffer pH results in a change of oligomeric states of WW45-SARAH, glutathione transferase (GST) and Mst2-SARAH, which undergo dissociation to a homodimer or a monomer from an oligomer or a homodimer at pH 3.5, respectively. Fig. 2 View largeDownload slide Effect of pH on the WW45- and Mst2-SARAH interaction. Elution profiles of WW45- and Mst2-SARAH complex were compared with uncomplexed WW45-SARAH and Mst2-SARAH at pH 3.5 and pH 7.5, respectively. The experiment was performed on a Superdex 200 HiLoad 16/60 and analytical 10/300 Superdex 200 column. Gel-filtration elution profiles. (A) Complexed and uncomplexed proteins in pH 3.5, (B) uncomplexed proteins in pH 7.5 and (C) mixed but uncomplexed proteins in pH 7.5. The WW45 and Mst2-SARAH was eluted as a single peak consistent with a complex at pH 3.5, but not at pH 7.5. SDS–PAGE analysis indicates a 1:1 stoichiometry for the complex. (D) Dynamic light scattering of WW45-SARAH. Intensity-Size distributions showed an apparent hydrodynamic radius of particles at different pHs. Fig. 2 View largeDownload slide Effect of pH on the WW45- and Mst2-SARAH interaction. Elution profiles of WW45- and Mst2-SARAH complex were compared with uncomplexed WW45-SARAH and Mst2-SARAH at pH 3.5 and pH 7.5, respectively. The experiment was performed on a Superdex 200 HiLoad 16/60 and analytical 10/300 Superdex 200 column. Gel-filtration elution profiles. (A) Complexed and uncomplexed proteins in pH 3.5, (B) uncomplexed proteins in pH 7.5 and (C) mixed but uncomplexed proteins in pH 7.5. The WW45 and Mst2-SARAH was eluted as a single peak consistent with a complex at pH 3.5, but not at pH 7.5. SDS–PAGE analysis indicates a 1:1 stoichiometry for the complex. (D) Dynamic light scattering of WW45-SARAH. Intensity-Size distributions showed an apparent hydrodynamic radius of particles at different pHs. Table II. Size exclusion chromatography data Sample  pH  Elution volume (ml)  CalculatedMW (kDa)  Amino acid sequence based MW (kDa)  WW45-SARAH  3.5  80.41  37  17.2, Homodimer  Mst2-SARAH  3.5  100.28  9  5.6, Monomer  WW45- and Mst2- SARAH complex  3.5  84.65  27  13.6, Heterodimer  GST  3.5  85.12  25.2  23.7, Monomer  WW45-SARAH  7.5  75.42  53  8.6, Monomer  Mst2-SARAH  7.5  92.43  17  11.2, Homodimer  GST  7.5  79.12  41  47.4, Homodimer  Sample  pH  Elution volume (ml)  CalculatedMW (kDa)  Amino acid sequence based MW (kDa)  WW45-SARAH  3.5  80.41  37  17.2, Homodimer  Mst2-SARAH  3.5  100.28  9  5.6, Monomer  WW45- and Mst2- SARAH complex  3.5  84.65  27  13.6, Heterodimer  GST  3.5  85.12  25.2  23.7, Monomer  WW45-SARAH  7.5  75.42  53  8.6, Monomer  Mst2-SARAH  7.5  92.43  17  11.2, Homodimer  GST  7.5  79.12  41  47.4, Homodimer  View Large pH-dependent size distributions were studied through dynamic light scattering (DLS) measurements. DLS is a convenient method to determine monodispersity of macromolecules in solution (23). The hydrodynamic radius (Rh) of WW45-SARAH in different pH conditions were: 8.68 nm with 41.2% polydispersity (Pd; pH 7.5), 7.93 nm with 31.98% Pd (pH 6.5), 6.4 nm with 26.1% Pd (pH 5.5), 5.28 nm with 29.7% Pd (pH 4.5) and 3.22 nm with 20.8% Pd (pH 3.5; Table III). Pd values <30% are indicative of a homogenous species in the solution. Scattering intensity of the major peak was significantly increased from 80.8% to 97.2% with decreasing pH, indicating that conformational change and stabilization of WW45-SARAH can be attributed to low pH. In pH 3.5, the plot shows that WW45-SARAH contains a small aggregate (Fig. 2D). Further, MWs determined by DLS revealed a lower trend as pH was reduced, although these sizes are significantly larger than the expected MW (8.6 kDa) determined from the amino acid sequence (Table III). A reduction in protein size may be due to the changes in local dynamics of the protein, which are induced by low-pH. Higher MW of coiled-coil motif compared to globular proteins is a characteristic feature that should be considered when analysing DLS data (24, 25). The results of the DLS experiments are consistent with CD measurements that suggest the conformational change and stabilization of WW45–SARAH could be driven by an acidic condition. In addition, DLS is primarily used to assess the aggregation state of a sample and to measure polydispersity, which is predictive of crystallizability (26, 27). Taken together, our data show that the conformation of WW45-SARAH refolded in an acidic condition is more likely to crystallize due to forming monodisperse in solution. It is well known that randomly aggregating protein molecules or polydisperse systems rarely produce crystals. Table III. Dynamic light scattering data for WW45-SARAH in various pHs Conditions  Radius (nm)  Pd (%)  Mw (kDa)  Intensity (%)  Mass (%)  pH 7.5  8.68  41.2  529  80.8  98.8  pH 6.5  7.93  31.9  428  88.3  99.1  pH 5.5  6.4  26.1  259  88.3  99.6  pH 4.5  5.28  29.7  165  85.6  99.9  pH 3.5  3.22  20.8  52.1  97.2  100  Conditions  Radius (nm)  Pd (%)  Mw (kDa)  Intensity (%)  Mass (%)  pH 7.5  8.68  41.2  529  80.8  98.8  pH 6.5  7.93  31.9  428  88.3  99.1  pH 5.5  6.4  26.1  259  88.3  99.6  pH 4.5  5.28  29.7  165  85.6  99.9  pH 3.5  3.22  20.8  52.1  97.2  100  View Large Our aim was to purify the recombinant protein available as a molecular basis of functional studies. Therefore, we attempted to obtain crystals suitable for X-ray analysis using refolded WW45-SARAH protein. Crystals diffracting to 6 Å resolution (Fig. 3A) were obtained with a hanging-drop at 20°C using a reservoir solution containing 0.1 M sodium citrate pH 3.2, 17% PEG400, 15% 2-propanol (Fig. 3B). Unfortunately, we could not get over diffraction limit. The presence of a WW45-SARAH domain in crystals was confirmed by SDS–PAGE analysis of harvested crystals (Fig. 3A). Therefore, the data suggests that the acidic pH-induced folding is suitable for enhancing homogeneity and stability of the WW45-SARAH protein that helps facilitate protein crystallization. Fig. 3 View largeDownload slide A crystal and diffraction image of WW45 SARAH. (A) Crystals were obtained in 0.1 M sodium citrate pH 3.2, 17% PEG400, 15% 2-propanol. The presence of WW45-SARAH in crystals was confirmed by SDS–PAGE of harvested crystals; lane 1, size marker, lane 2, WW45-SARAH used for crystallization, lane 3, dissolved crystals for SDS–PAGE. (B) Diffraction image was gained on a BL44XU beamline at SPring-8 (Harima, Japan) at 6 angstrom. Fig. 3 View largeDownload slide A crystal and diffraction image of WW45 SARAH. (A) Crystals were obtained in 0.1 M sodium citrate pH 3.2, 17% PEG400, 15% 2-propanol. The presence of WW45-SARAH in crystals was confirmed by SDS–PAGE of harvested crystals; lane 1, size marker, lane 2, WW45-SARAH used for crystallization, lane 3, dissolved crystals for SDS–PAGE. (B) Diffraction image was gained on a BL44XU beamline at SPring-8 (Harima, Japan) at 6 angstrom. Low pH-induced folding of WW45-SARAH facilitates heterotypic interaction with Mst2-SARAH The effect of low pH-driven folding on the complex formation with Mst2 via SARAH was compared at pH 3.5 and pH 7.5 using SEC. Surprisingly, a single peak at ∼85 ml was found to elute at pH 3.5 with an apparent MW of 27 kDa (Fig. 2A and Table II), while peaks corresponding to WW45- and Mst2-SARAH were separately observed at pH 7.5 (Fig. 2C). This complex peak was behind the elution peak of a single-purified WW45-SARAH. These data show the possibility that WW45-SARAH existed as a homodimer in pH 3.5 and could be readily disrupted by monomeric Mst2-SARAH binding and forming a heterodimer with Mst2-SARAH. Ni et al. (11) showed that the homodimer of Mst2-SARAH is disrupted by RASSF5-SARAH binding. In addition, SEC measurement and SDS–PAGE traces of the single peak relevant to the WW45- and Mst-SARAH complex produced 1:1 stoichiometric ratios. This result shows that low pH-induced folding of WW45-SARAH could affect WW45-and Mst2-SARAH complexation supporting the existence of functional WW45-SARAH in pH 3.5. As described earlier, low pH-driven conformation of WW45-SARAH may allow a specific and coordinated protein–protein interaction with a binding partner or self-oligomerization. The folding and interactions for pH-dependent coiled-coil formation have been previously reported (18). Par-4, a coiled-coil protein essential to apoptotic pathways of various cell types, exists in the unfolded monomer under natural conditions, but undergoes conformational changes at low pH and low temperature, suggesting that a Par-4 leucine zipper (LZ) is in a stable state (18). Neuroglobin (Ngb) is also a folded and functional protein at pH 2.0 (28). Although the detailed low pH-induced protein stability mechanism of these proteins is not yet elucidated, it might be required to exert its specific role under various conditions resulting in cellular or subcellular acidification. Mst2-SARAH appears to undergo a similar oligomeric dissociation to WW45-SARAH in pH 3.5 (Fig. 2A and B). To verify that the oligomeric state of Mst2-SARAH contributes to the interaction between WW45 and Mst2- SARAH, a pull-down assay was performed using purified WW45 and Mst2 proteins. Structure-based mutants for disrupting the dimer interface were designed on the basis of Mst2-SARAH structure [PDBID: 4HKD]. GST-tagged Mst2-SARAH mutants M459A and L466A, which are known to impair homodimerization of Mst2, showed significantly stronger binding to WW45-SARAH purified at pH 3.5 compared to the wt Mst2-SARAH homodimer (Fig. 4A left, lane 4 and 5). This interaction was definitely reduced when the WW45-SARAH oligomer purified at pH 7.5 was examined (Fig. 4A left, lane 2). In contrast, WW45-SARAH purified in pH 7.5 showed similar interactions with Mst2-SARAH wt or mutants (Fig. 4A, right). Stability of WW45-SARAH folding was confirmed using far-UV CD to determine whether a low pH-induced folding of WW45-SARAH is altered by the buffer change to pH 7.5. It showed that a low pH-folded WW45-SARAH maintained a secondary structure at pH 7.5 (Fig. 4B). WW45-SARAH is known as an alpha-helical coiled coil motif. We suppose that the coiled coil packing induced by low pH contains a hydrophobic interior, resulting in the conformational stabilization and structural rigidity during the buffer exchange. Hydrophobic interactions of at the oligomeric interface of coiled coil domain are the driving force for folding and oligomerization of coiled coils (18,). Taken together, our data suggest that WW45-SARAH undergoes conformational changes from an oligomeric state in pH 3.5 which facilitates the stable complex formation between WW45- and Mst2-SARAH. These results are consistent with SEC data that complex formation is enhanced by dimerization of WW45-SARAH in an acidic condition. Mst1/2 can form a heterodimer with RASSF through the SARAH domain (29). The oligomeric state of SARAH domain-containing proteins is important for their cellular activities (30). Although Ni et al. (11) suggested that WW45-SARAH can form a heterotetramer with Mst1/2-SARAH), Hauri et al. (31) proposed a heterodimeric interaction between WW45- and Mst1/2-SARAH. Nevertheless, the complexation module of Hippo signalling components depending on SARAH domain is not fully elucidated. Further studies on the structure of WW45-SARAH domain and WW45/Mst2-SARAH complex are required to flesh out some of these mechanisms. Fig. 4 View largeDownload slide GST pull-down analysis for the interaction WW45-SARAH and Mst2-SARAH wt and mutants. (A) Effects of GST Mst2-SARAH wt and mutations, disrupting the dimer interface, on WW45-SARAH binding. (B) Far-UV spectra of WW45-SARAH. Spectra were recorded in buffer containing 20 mM sodium citrate pH 3.5, and then in buffer exchanged by 20 mM Tris–HCl pH 7.5. Low pH-induced folding of WW45-SARAH was not altered by buffer exchange with 20 mM Tris–HCl pH 7.5. Fig. 4 View largeDownload slide GST pull-down analysis for the interaction WW45-SARAH and Mst2-SARAH wt and mutants. (A) Effects of GST Mst2-SARAH wt and mutations, disrupting the dimer interface, on WW45-SARAH binding. (B) Far-UV spectra of WW45-SARAH. Spectra were recorded in buffer containing 20 mM sodium citrate pH 3.5, and then in buffer exchanged by 20 mM Tris–HCl pH 7.5. Low pH-induced folding of WW45-SARAH was not altered by buffer exchange with 20 mM Tris–HCl pH 7.5. Conclusion In this study, we focused on understanding the molecular and functional properties of refolded WW45-SARAH to perform a molecular study. Low pH conditions significantly affected the folding conformation of WW45-SARAH. We could obtain the diffracting crystals of WW45-SARAH, implying that the homogeneity and stability of WW45-SARAH, purified in an acidic pH, were suitable for structure determination. Monomeric formation of Mst2-SARAH and WW45-SARAH folding in an acidic pH facilitated the heterotypic interaction between WW45 and Mst2. Acknowledgements Synchrotron radiation experiments were performed at the BL44XU of SPring-8, Japan (2013A6500, 2013B6500) and the Protein Beamline 7A, Pohang Light Source, Korea. Funding This research was supported by a National Research Foundation of Korea (NRF) grant, funded by the Korea government (MEST) (2008-0062275, 2011-0017405), and research grant from Chungbuk National University in 2011 to L.S.J. Conflict of Interest None declared. References 1 Tapon N,  Harvey KF,  Bell DW,  Wahrer DC,  Schiripo TA,  Haber DA,  Hariharan IK.  salvador promotes both cell cycle exit and apoptosis in Drosophila and is mutated in human cancer cell lines,  Cell. ,  2002, vol.  110 (pg.  467- 478) Google Scholar CrossRef Search ADS PubMed  2 Kango-Singh M,  Nolo R,  Tao C,  Verstreken P,  Hiesinger PR,  Bellen HJ,  Halder G.  Shar-pei mediates cell proliferation arrest during imaginal disc growth in Drosophila,  Development. ,  2002, vol.  129 (pg.  5719- 5730) Google Scholar CrossRef Search ADS PubMed  3 Halder G,  Johnson RL.  Hippo signaling: growth control and beyond,  Development. ,  2011, vol.  138 (pg.  9- 22) Google Scholar CrossRef Search ADS PubMed  4 Staley BK,  Irvine KD.  Hippo signaling in Drosophila: recent advances and insights,  Dev. Dyn. ,  2012, vol.  241 (pg.  3- 15) Google Scholar CrossRef Search ADS PubMed  5 Cai J,  Zhang N,  Zheng Y,  de Wilde RF,  Maitra A,  Pan D.  The Hippo signaling pathway restricts the oncogenic potential of an intestinal regeneration program,  Genes Dev. ,  2010, vol.  24 (pg.  2383- 2388) Google Scholar CrossRef Search ADS PubMed  6 Pan D.  The hippo signaling pathway in development and cancer,  Dev. Cell. ,  2010, vol.  19 (pg.  491- 505) Google Scholar CrossRef Search ADS PubMed  7 Scheel H,  Hofmann K.  A novel inter action motif, SARAH, connects three classes of tumor suppressor,  Curr. Biol. ,  2003, vol.  13 (pg.  R899- R900) Google Scholar CrossRef Search ADS PubMed  8 Pfeifer GP,  Dammann R,  Tommasi S.  RASSF proteins,  Curr. Biol. ,  2010, vol.  20 (pg.  R344- R345) Google Scholar CrossRef Search ADS PubMed  9 Callus BA,  Verhagen AM,  Vaux DL.  Association of mammalian sterile twenty kinases, Mst1 and Mst2, with hSalvador via C‐terminal coiled‐coil domains, leads to its stabilization and phosphorylation,  FEBS J. ,  2006, vol.  273 (pg.  4264- 4276) Google Scholar CrossRef Search ADS PubMed  10 Avruch J,  Zhou D,  Fitamant J,  Bardeesy N,  Mou F,  Barrufet LR.  Protein kinases of the Hippo pathway: regulation and substrates,  Semin. Cell Dev. Biol. ,  2012, vol.  23 (pg.  770- 784) Google Scholar CrossRef Search ADS PubMed  11 Ni L,  Li S,  Yu J,  Min J,  Brautigam CA,  Tomchick DR,  Pan D,  Luo X.  Structural basis for autoactivation of human Mst2 kinase and its regulation by RASSF5,  Structure. ,  2013, vol.  21 (pg.  1757- 1768) Google Scholar CrossRef Search ADS PubMed  12 Hwang E,  Ryu K-S,  Pääkkönen K,  Güntert P,  Cheong H-K,  Lim D-S,  Lee J-O,  Jeon YH,  Cheong C.  Structural insight into dimeric interaction of the SARAH domains from Mst1 and RASSF family proteins in the apoptosis pathway,  Proc. Natl. Acad. Sci. ,  2007, vol.  104 (pg.  9236- 9241) Google Scholar CrossRef Search ADS   13 Liu G,  Shi Z,  Jiao S,  Zhang Z,  Wang W,  Chen C,  Hao Q,  Zhang M,  Feng M,  Xu L.  Structure of MST2 SARAH domain provides insights into its interaction with RAPL,  J. Struct. Biol. ,  2014, vol.  185 (pg.  366- 374) Google Scholar CrossRef Search ADS PubMed  14 Wu D,  Wu S.  A conserved serine residue regulates the stability of Drosophila Salvador and human WW domain-containing adaptor 45 through proteasomal degradation,  Biochem. Biophys. Res. Commun. ,  2013, vol.  433 (pg.  538- 541) Google Scholar CrossRef Search ADS PubMed  15 Yu F-X,  Zhang Y,  Park HW,  Jewell JL,  Chen Q,  Deng Y,  Pan D,  Taylor SS,  Lai Z-C,  Guan K-L.  Protein kinase A activates the Hippo pathway to modulate cell proliferation and differentiation,  Genes Dev. ,  2013, vol.  27 (pg.  1223- 1232) Google Scholar CrossRef Search ADS PubMed  16 Song J,  Hong H,  Choi S,  Lee YH,  Yamashita E,  Bae S-C,  Park IY,  Lee SJ.  Crystallization and preliminary X-ray crystallographic study of the human MST2 SARAH domain,  Acta Crystall. Secti. F. ,  2011, vol.  67 (pg.  1403- 1405) Google Scholar CrossRef Search ADS   17 Perez-Iratxeta C,  Andrade-Navarro MA.  K2D2: estimation of protein secondary structure from circular dichroism spectra,  BMC Struct. Biol. ,  2008, vol.  8 pg.  25  Google Scholar CrossRef Search ADS PubMed  18 Dutta K,  Alexandrov A,  Huang H,  Pascal SM.  pH‐induced folding of an apoptotic coiled coil,  Protein Sci. ,  2001, vol.  10 (pg.  2531- 2540) Google Scholar CrossRef Search ADS PubMed  19 Di Russo NV,  Estrin DA,  Martí MA,  Roitberg AE.  pH-Dependent conformational changes in proteins and their effect on experimental pKas: the case of nitrophorin 4,  PLoS Comput. Biol. ,  2012, vol.  8 pg.  e1002761  Google Scholar CrossRef Search ADS PubMed  20 Park SY,  Bae DJ,  Kim MJ,  Piao ML,  Kim IS.  Extracellular low pH modulates phosphatidylserine-dependent phagocytosis in macrophages by increasing stabilin-1 expression,  J. Biol. Chem. ,  2012, vol.  287 (pg.  11261- 11271) Google Scholar CrossRef Search ADS PubMed  21 Guthrie CFG. ,  Guide to Yeast Genetics and Molecular Biology ,  1991 San Diego, CA Academic Press, Inc 22 Beck K,  Brodsky B.  Supercoiled protein motifs: the collagen triple-helix and the α-helical coiled coil,  J. Struct. Biol. ,  1998, vol.  122 (pg.  17- 29) Google Scholar CrossRef Search ADS PubMed  23 Schmitz KS. ,  An Introduction to Dynamic Light Scattering by Macromolecules ,  1990 New York. Academic Press 24 Michielsen EC,  Diris JH,  Kleijnen VW,  Wodzig WK,  Dieijen-Visser V,  Marja P.  Interpretation of cardiac troponin T behaviour in size-exclusion chromatography,  Clin. Chem. Lab. Med. ,  2006, vol.  44 (pg.  1422- 1427) Google Scholar CrossRef Search ADS PubMed  25 Peteranderl R,  Rabenstein M,  Shin Y-K,  Liu CW,  Wemmer DE,  King DS,  Nelson HC.  Biochemical and biophysical characterization of the trimerization domain from the heat shock transcription factor,  Biochemistry. ,  1999, vol.  38 (pg.  3559- 3569) Google Scholar CrossRef Search ADS PubMed  26 George A,  Wilson WW.  Predicting protein crystallization from a dilute solution property,  Acta Crystall. Sect. D. ,  1994, vol.  50 (pg.  361- 365) Google Scholar CrossRef Search ADS   27 Bonnete F,  Vivares D.  Interest of the normalized second virial coefficient and interaction potentials for crystallizing large macromolecules,  Acta Crystall. Sect. D. ,  2002, vol.  58 (pg.  1571- 1575) Google Scholar CrossRef Search ADS   28 Picotti P,  Dewilde S,  Fago A,  Hundahl C,  De Filippis V,  Moens L,  Fontana A.  Unusual stability of human neuroglobin at low pH–molecular mechanisms and biological significance,  FEBS J. ,  2009, vol.  276 (pg.  7027- 7039) Google Scholar CrossRef Search ADS PubMed  29 Hwang E,  Cheong HK,  Ul Mushtaq A,  Kim HY,  Yeo KJ,  Kim E,  Lee WC,  Hwang KY,  Cheong C,  Jeon YH.  Structural basis of the heterodimerization of the MST and RASSF SARAH domains in the Hippo signalling pathway,  Acta Crystall. Sect. D. ,  2014, vol.  70 (pg.  1944- 1953) Google Scholar CrossRef Search ADS   30 Makbul C,  Constantinescu Aruxandei D,  Hofmann E,  Schwarz D,  Wolf E,  Herrmann C.  Structural and thermodynamic characterization of Nore1-SARAH: a small, helical module important in signal transduction networks,  Biochemistry. ,  2013, vol.  52 (pg.  1045- 1054) Google Scholar CrossRef Search ADS PubMed  31 Hauri S,  Wepf A,  Drogen A,  Varjosalo M,  Tapon N,  Aebersold R,  Gstaiger M.  Interaction proteome of human Hippo signaling: modular control of the co‐activator YAP1,  Mol. Syst. Biol. ,  2013, vol.  9 pg.  713  Google Scholar CrossRef Search ADS PubMed  © The Authors 2015. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved
TI - Low pH-driven folding of WW45-SARAH domain leads to stabilization of the WW45-Mst2 complex
JF - The Journal of Biochemistry
DO - 10.1093/jb/mvv031
DA - 2015-03-25
UR - https://www.deepdyve.com/lp/oxford-university-press/low-ph-driven-folding-of-ww45-sarah-domain-leads-to-stabilization-of-CDFTRIX0iS
SP - 181
EP - 188
VL - 158
IS - 3
DP - DeepDyve
ER -