BGB 15025

Pervanadate induces Mammalian Ste20 Kinase 3 (MST3) tyrosine phosphorylation but not activation

Wei-Chih Kan a,b,1, Te-Ling Lu c,1, Pin Ling d, Te-Hsiu Lee b, Chien-Yu Cho e, Chi-Ying F. Huang f, Wen-Yih Jeng e,
Yui-Ping Weng g, Chun-Yen Chiang h,i, Jin Bin Wu c, Te-Jung Lu b,⁎
a Department of Nephrology, Chi-Mei Medical Center, Tainan 701, Taiwan
b Department of Medical Laboratory Science and Biotechnology, Chung Hwa University of Medical Technology, Tainan 701, Taiwan
c School of Pharmacy, Tsuzuki Institute for Traditional Medicine, China Medical University, Taichung 404, Taiwan
d Department of Microbiology and Immunology, College of Medicine, National Cheng Kung University, Tainan 701, Taiwan
e Department of Biochemistry and Molecular Biology, College of Medicine, National Cheng Kung University, Tainan 701, Taiwan
f Institute of Biopharmaceutical Sciences, National Yang-Ming University, Taipei 112, Taiwan
g Graduate Institute of Biological Science and Technology, Chung Hwa University of Medical Technology, Tainan, Taiwan
h Department of Cardiology, Chi-Mei Medical Center, Tainan, Taiwan
i Department of Optometry, Chung Hwa University of Medical Technology, Tainan, Taiwan

a b s t r a c t
The yeast Ste20 (sterile) protein kinase, which is a serine/threonine kinase, responds to the stimulation of the G proteincoupled receptor (GPCR) pheromone receptor. Ste20 protein kinase serves as the critical component that links signaling from the GPCR/G proteins to the mitogen-activated protein kinase (MAPK) cascade in yeast. The yeast Ste20p functions as a MAP kinase kinase kinase kinase (MAP4K) in the pheromone response. Ste20-like kinases are structurally conserved from yeast to mammals. The mechanism by which MAP4K links GPCR to the MAPK pathway is less clearly defined in vertebrates. In addition to MAP4K, the tyrosine kinase cascade bridges G proteins and the MAPK pathway in vertebrate cells. Mammalian Ste20 Kinase 3 (MST3) has been categorized into the Ste20 family and has been reported to function in the regulation of cell polarity and migration. However, whether MST3 tyrosine phosphorylation regulates diverse signaling pathways is unknown. In this study, the tyrosine phosphatase inhibitor pervanadate was found to induce MST3 tyrosine phosphorylation in intact cells, and the activity of tyrosine-phosphorylated MST3 was measured. This tyrosine-directed phosphorylation was in- dependent of MST3 activity. Parameters including protein conformation, Triton concentration and ionic concen- tration influenced the sensitivity of MST3 activity. Taken together, our data suggests that the serine/threonine kinase MST3 undergoes tyrosinedirected phosphorylation. The tyrosine-phosphorylated MST3 may create a docking site for the structurally conserved SH2/SH3 (Src Homology 2 and 3) domains within the Src oncoprotein. The unusual tyrosinephosphorylated MST3 may recruit MST3 to various signaling components.
Abbreviations: Cys, cysteine; DMEM, Dulbecco’s modified Eagle’s medium; EGFR, epidermal growth factor receptor; GPCR, G protein-coupled receptor; HEK 293, human embryonic kidney 293; HPK1, hematopoietic progenitor kinase 1; MST, mammalian Ste20 kinase; MAP4K, MAP kinase kinase kinase kinase; MAP3K, MAP kinase kinase ki- nase; MAP2K, MAP kinase kinase; MAPK, MAP kinase; MDCK, Madin-Darby canine kidney; NMR, nuclear magnetic resonance; OA, okadaic acid; PAK, p21-activated kinase; PLD1, phospholipase D1; PSP, protein serine/threonine phosphatase; PTP, protein tyrosine phos- phatase; PV, pervanadate; SH2, Src Homology 2; SH3, Src Homology 3; Ste20, sterile20; Triton, Triton X-100; T178, threonine-178; T328, threonine-328.
Keywords:
MST3
Tyrosine phosphorylation Kinase assay
Triton
Ionic strength

1. Introduction
Yeast cells respond to pheromones by activating a signal transduc- tion pathway that involves a transmembrane G-protein-coupled recep- tor (GPCR) and G proteins [1]. The signal from the G proteins to the mitogen-activated protein kinase (MAPK) cascade is best understood in yeast, in which a Ste20 (sterile) protein kinase serves as the critical linking component [2]. Yeast Ste20p is located downstream of the GPCR/G proteins and functions as a MAP kinase kinase kinase kinase (MAP4K), which directly phosphorylates and activates the yeast MAP kinase kinase kinase (MAP3K) Ste11p. Ste11p sequentially phosphory- lates and activates MAP kinase kinase (MAP2K) and MAPK [3]. The mechanism by which MAP4K links GPCRs to the MAPK pathway is less clearly defined in vertebrates. A large emerging group of mammalian Ste20-like kinases, such as HPK1 (hematopoietic progenitor kinase 1), PAK (p21-activated kinase) and MST1–4 (Mammalian Ste20 Kinase), has been identified by specificity for the distinct catalytic domains of Ste20-like kinases [4–6]. HPK1 and PAK have been reported to function as MAP4Ks [6,7]. In addition, MST4 has been shown to mediate MAPK pathway activation [8]. These data suggested that mammalian Ste20-like kinases serve as component that link G proteins to the MAPK cascade in vertebrates.
Several specific scaffold proteins that link GPCRs/G proteins to distinct MAPK modules contribute to the context-specific and spatial- temporal regulation of the MAPK cascade [9–11]. The tyrosine kinase cascade is involved in G protein signaling and the MAPK pathway in ver- tebrate cells. For example, a GPCR transactivates the EGFR (epidermal growth factor receptor)/MAPK pathway by recruiting the SH2 (Src Homology 2) domains to tyrosine-phosphorylated EGFR tails [12]. Src is a tyrosine kinase and belongs to a family of non-receptor tyrosine kinases called Src family kinases. The SH2 domain is a structurally conserved domain within the Src oncoprotein. The SH3 (Src Homology 3) domains of tyrosine kinases or adaptor proteins, such as Grb2, recog- nize proline-rich motifs with the consensus sequence Pro-X-X-Pro (P- X-X-P) in downstream molecules [13–18]. These associated proteins transduce signals to the downstream MAPK pathway. In addition, EGF has been reported to phosphorylate HPK1 tyrosine. The proline-rich motif of HPK1 associates with Grb2, which is recruited to EGFR [19]. These results suggest that tyrosine kinase is involved in the interaction between MAP4K and the adaptor Grb2.
MST3 (EC# 2.7.11.1) is a member of the serine/threonine Ste20 fam- ily; members of this family transfer phosphates to the oxygen atom of a serine or threonine side chain in substrate proteins. MST3 threonine- 178 (T178) in the activation loop between the subdomains VII (DFG [Asp-Phe-Gly]) and VIII (APE [Ala-Pro-Glu]) of Ste20-like kinases is a conserved autophosphorylation site. The mutant T178A-MST3, in which the threonine at codon 178 is replaced with an alanine, does not have MST3 activity and the ability to regulate cell polarity and mi- gration [20]. Analysis of the MST3 sequence revealed that MST3 has a P-K-R-P (residues 356 to 359 at the C-terminus) proline-rich motif. Whether MST3 serves as a MAP4K, through a proline-rich motif close to an SH3 domain of a tyrosine kinase, and whether MST3 is regulated by a tyrosine kinase has not been reported. Vanadate is commonly used as a protein-tyrosine phosphatase (PTP) inhibitor. Vanadate, which has a structure similar to phosphate, forms a thiol-vanadyl ester linkage that resembles the thiol-phosphate linkage. Pervanadate (PV), a complex formed between Na3VO4 and H2O2, has been shown to be an irreversible inhibitor of PTPs by oxidizing the catalytic cysteine (Cys) residue. PV is more effective than vanadate in increasing the level of cellular tyrosine phosphorylation [21–24]. To investigate whether MST3 is tyrosine phosphorylated and whether this phosphorylation af- fects MST3 activity, PV was chosen to examine whether MST3 could be tyrosine phosphorylated in this study.
The most common approach to studying kinase activity is to evalu- ate the ability of the kinase to transfer a phosphor-group to a synthetic peptide. The kinase is first extracted from cells, then immunoprecipita- tion and incubated with the substrate peptide. A Triton lysis buffer is generally used for kinase assays from cells. A nonionic detergent in the Triton lysis buffer, such as NP-40 or Triton X-100 (Triton), is less harsh than ionic detergents, such as SDS and sodium deoxycholate. Other components in the Triton lysis buffer, such as the salt concentra- tion, also affect the success of kinase assays. However, MST3 activity was not optimized in these buffers; conditions allowing the optimization of the protein conformation, the Triton concentration and the salt concen- tration were examined to determine the best buffer conditions for MST3 activity.

2. Materials and methods
Preparation of Pervanadate (PV) — Sodium orthovanadate (Na3VO4, 1.84 g) was dissolved in 90 ml of water, and then, the pH was adjusted to 10 using either 1 N NaOH or 1 N HCl with stirring. At pH 10, the solu- tion was yellow. The solution was boiled until it was colorless. It was then cooled to room temperature. The pH was readjusted to 10, and the boiling and cooling of the solution were repeated until it was color- less and the pH was stabilized at 10. The final volume was adjusted to 100 ml with water to a final concentration of 100 mM Na3VO4. The PV solution (10 mM) was prepared by adding 10 μl of 100 mM Na3VO4 and 30 μl of 100 mM hydrogen peroxide (diluted from a 30% stock in water, pH 7.3) to 100 μl of H2O for a 15 min period. As PV is unstable, 10 mM PV was diluted in serum-free Dulbecco’s modified Eagle’s medi- um (DMEM) to obtain a final concentration of 100 μM, and the solution was added to the cells immediately after preparation [21].
Preparation of okadaic acid (OA)—An OA stock solution (60 μM) was prepared by adding 10 μg of OA to 200 μl dimethyl sulfoxide (DMSO). Then, the 60 μM OA stock solution was diluted in serum-free DMEM to obtain a final concentration of 600 nM.
Cell culture and treatment — Madin-Darby canine kidney (MDCK) cells, originating from the renal collecting duct, or human embryonic kidney cells (HEK293) were cultured in DMEM medium supplemented with 10% fetal calf serum (FCS) and 100 U/ml penicillin/streptomycin (all purchased from Gibco-BRL) and incubated at 37 °C in a humidified atmosphere containing 5% CO2. When the cells reached 90% confluency, they were serum-starved overnight before starting PV or OA treatment. One hundred micromolar PV or 600 nM OA in serum-free DMEM was added to the MDCK or HEK293 cells for 20 min or 1 h, respectively.
Buffer formulations—The ice-cold Tris buffer contained 25 mM Tris- HCl, 150 mM NaCl, 5 mM KCl, and 0.7 mM Na2HPO4; the Triton lysis buffer contained 150 mM NaCl, 1% Triton, 20 mM HEPES, 2 mM EDTA, 500 μM DTT, 25 mM beta-glycerophosphate, 10% glycerol, and protease and phosphatase inhibitors (100 μM Na3VO4); the hypotonic lysis buffer contained 20 mM HEPES, 2 mM EDTA, 500 μM DTT, 25 mM beta- glycerophosphate, 10% glycerol, and protease and phosphatase inhibitors (100 μM Na3VO4); the kinase wash buffer contained 20 mM HEPES, 500 μM DTT, 5 mM beta-glycerophosphate, and 100 μM Na3VO4; and the kinase reaction buffer contained 10 mM peptide substrate, 5 mM MnCl2 cofactor, 20 μCi [γ-33P]-ATP (2500 Ci/mmol), and 20 μM unlabeled ATP.
MST3 immunoprecipitation and activity assay—A synthetic peptide containing the MST3 activation loop was used as the substrate peptide. Three buffers were required for the MST3 kinase assay: (1) Triton lysis buffer that contained Triton to release MST3 from cells in order to perform the subsequent immunoprecipitation and to wash the MST3- antibody complexes; (2) kinase wash buffer to replace the lysis buffer components in order to prepare for the following kinase reaction; and (3) kinase reaction buffer that contained the co-factors manganese and ATP (cold and radio-activate ATP), as well as a peptide substrate to report the MST3 activity. MST3 activity was measured by the detection of the ability of MST3 to transfer [γ-33P]-ATP to an MST3 peptide substrate. The peptide substrate sequence used was TQIKRNTFVGTPFWMAPEVIKQS, which mapped to residues 172–194 of MST3 and was conserved in the activation loop of Ste20 kinases. Tagged-MST3 from cell lysates or active purified MST3 from baculovirus (in cell-free system) was immunoprecipitated using the appropriate antibody. When the cells reached 90% confluence, they were washed three times with ice-cold Tris buffer on ice. The cells were lysed in the Triton lysis buffer or the hypotonic lysis buffer. Cell homogenates were centrifuged at 10,000 ×g for 15 min. The supernatant was incubat- ed for 2 h at 4 °C with the appropriate antibody and then incubated with protein G sepharose (Amersham Pharmacia Biotech, NJ, USA) for 1.5 h at 4 °C. The immune complexes were washed three times with the lysis buffer, followed by three washes with cold kinase wash buffer. An equal volume of 2 × kinase reaction buffer was added to each sample, and the samples incubated for 20 min at 30 °C. The mixed samples were then dropped onto circular Whatman P81 phosphocellulose paper. The P81 paper was washed three times with 0.5% phosphoric acid for 3 min to remove any unbound [γ-33P]-ATP. The radioactivity of [γ-33P]-incorporated peptide substrate on the P81 paper was quantified by the Cerenkov method using scintillation [25–28].
Western Blot Analysis—Samples were resolved by SDS-PAGE and transferred to a nitrocellulose membrane. Western blot was performed as described [29]. Anti-HA (12-CA5, Sigma, St. Louis, MO) was used at 1:5000, anti-MST3 (kindly provided by Ming-Derg Lai, NCKU, Taiwan) was used at 1:5000, and anti-4G10 (immunogen: Phosphotyramine- KLH, Upstate Biotechnology, NY) was used at 1:3000.

3. Results

3.1. MST3 is tyrosine phosphorylated upon PV treatment
To determine whether MST3 was tyrosine phosphorylated, we treat- ed cells with the tyrosine phosphatase inhibitor PV [30–33]. We transfected HEK293 or MDCK cells with MST3-HA and examined MST3 tyrosine residue phosphorylation after PV treatment. The results demonstrate that the PV treatment decreased MST3-HA mobility by SDS-PAGE, suggesting that MST3 was post-translationally modified upon PV treatment in intact cells (Fig. 1A). To determine whether the MST3 post-translational modification was caused by tyrosine phosphorylation, we immunoprecipitated MST3-HA and probed with the anti-phosphotyrosine antibody 4G10. The position of the tyrosine- phosphorylated MST3 is indicated on the gel (Fig. 1B, lane 2, arrow- head). In addition, we probed MST3-HA-transfected cell lysate with the 4G10 antibody. The PV treatment led to global tyrosine phosphory- lation, indicating that PV serves as a tyrosine phosphatase inhibitor (Fig. 1B, lane 4). These results demonstrated that an upstream tyrosine kinase could phosphorylate and regulate MST3 in intact cells.

3.2. Effect of protein conformation on MST3 activity
To investigate whether MST3 tyrosine phosphorylation affected MST3 activity, we performed an MST3 kinase assay. We transfected HEK293 cells with HA-MST3 or HA-T178A-MST3. The MST3 kinase assay revealed that HA-T178A-MST3 activity, compared with HA- MST3 activity, was lost (Fig. 2, columns 1 and 2). Because MST3 has been reported to be autophosphorylated at threonine 178, we treated cells with OA, a potent protein serine/threonine phosphatase inhibitor, to evaluate MST3 activity [34,35]. However, the OA treatment did not increase MST3 activity compared with controls (Fig. 2, columns 1 and 3). These data indicated that the MST3 kinase assay could successfully distinguish between wild type HA-MST3 and mutant HA-T178A-MST3 activity but could not successfully detect OA-induced MST3 activity, suggesting that the assay was not optimized. To determine whether protein conformation affected MST3 activity, we used both N- and C-terminally tagged MST3 (HA-MST3, or MST3-HA) to determine which tagged MST3 allowed for higher MST3 activity. Using the MST3 kinase assay described above, we found that MST3-HA had higher activity than HA-MST3 (Fig. 3, columns 1 and 2). This result may have occurred because the HA antibody binds to the C-terminal site of MST3-HA during immunoprecipitation, leaving the MST3 N-terminal catalytic domain unmasked and allowing it to more easily transfer a phospho-group to the substrate. In contrast, HA antibody binding to the N-terminus of HA-MST3 during immunoprecipitation masked the N-terminal kinase domain. To determine whether we had optimized the MST3 kinase assay, we transfected HEK293 cells with MST3-HA or K53R-MST3-HA (kinase-dead MST3) to evaluate MST3 activity. The re- sults demonstrated that MST3-HA had higher activity than K53R-MST3- HA (Fig. 3, columns 2 and 3). However, the MST3 kinase assay still could not detect OA-induced MST3 activation (Fig. 3, columns 2 and 4).

3.3. Effects of Triton on MST3 activity
Although MST3-HA had higher activity, we still not detect the induc- tion of MST3-HA activity upon OA treatment (Fig. 3, columns 2 and 4). We hypothesized that the components in the Triton lysis buffer affected MST3 activity during MST3 protein extraction from cells and the subse- quent immunoprecipitation. Therefore, we used baculovirus-expressed His-MST3 (in cell-free system) to evaluate MST3 activity. We added active purified His-MST3 directly into the Triton lysis buffer or the kinase wash buffer to immunoprecipitate His-MST3. After immunopre- cipitation, we added MST3 to the kinase reaction buffer to measure MST3 activity. MST3 activity was significantly reduced when the purified His-MST3 was exposed to the lysis buffer. However, we observed a prominent peak of activity upon exposure to the kinase wash buffer (Fig. 4A). The results showed that the components of the lysis buffer but not the kinase reaction buffer affected MST3 activity. To determine whether the Triton in the lysis buffer affected MST3 activ- ity, we used different Triton concentrations in the Triton lysis buffer to immunoprecipitate His-MST3. His-MST3 activity was inhibited by approximately 90% in the presence of 1% Triton in lysis buffer (Fig. 4B, column 2). Although His-MST3 had higher activity in 0.2% Triton than in 1% Triton (Fig. 4B, column 1 and 2), His-MST3 activity did not reach the same level as when it was directly added to the kinase wash buffer for precipitation (Fig. 4B, column 3), suggesting that Triton had adverse effects on MST3 activity.

3.4. Hypotonic lysis buffer was optimal for MST3 activity
To extract MST3 from cells without Triton, we lysed the cells and precipitated MST3 in the hypotonic lysis buffer. We washed the precip- itated MST3-antibody complex with the hypotonic lysis buffer and incu- bated it in the kinase reaction buffer. MST3 had the highest activity when the cells were lysed in the hypotonic lysis buffer (Fig. 5, column 1). In addition to Triton, salt has been reported to affect kinase activity. To examine whether NaCl affected MST3 activity, we added 150 mM NaCl to the hypotonic lysis buffer when we washed the MST3- antibody complex. The results showed that MST3 activity was inhibited by approximately 40%, indicating that NaCl also had an adverse effect on MST3 activity (Fig. 5, column 3). When both 150 mM NaCl and Triton were added to the wash buffer, MST3 activity was reduced to approxi- mately 20% (Fig. 5, column 2), indicating that both NaCl and Triton have adverse effects on MST3 activity.

3.5. The effect of ionic strength on MST3 activity
To further examine the effect of NaCl on MST3 activity, we added various NaCl concentrations (0–150 mM) to the kinase wash buffer after precipitating the MST3-antibody complex. We found that MST3 activity was maintained at its maximum value in 0–50 mM NaCl but de- creased as the NaCl concentration increased above 50 mM. An increase in ionic strength above 100 mM decreased MST3 activity by 50% (Fig. 6, black bar). To determine whether other ionic strengths besides NaCl could inhibit MST3 activity, we replaced NaCl with KCl, which has an equivalent ionic strength. We did not find any difference in MST3 activity when we replaced NaCl with KCl. KCl concentrations above 100 mM also lowered MST3 activity by 50% (Fig. 6, gray bar). High con- centrations of either NaCl or KCl inhibited MST3 activity, suggesting that MST3 did not have an ionic strength preference for its activity (Fig. 6).

3.6. MST3 activity does not require tyrosine phosphorylation
We determined that the hypotonic lysis buffer was the best buffer for the MST3 activity analysis from cell lysates. Therefore, we used the hypotonic lysis buffer to detect MST3 activity after PV or OA treatment. Using the modified MST3 kinase assay, we found that OA treatment increased MST3 activity by 3–4-fold (Fig. 7, columns 2 and 3). K53R- MST3-HA showed low activity (Fig. 7, column 4). However, we did not observe a significant increase or decrease in MST3 activity after PV treatment (Fig. 7, column 1). These results suggest that MST3 activity is independent of its tyrosine phosphorylation.

4. Discussion
In this study, we found that the tyrosine phosphatase inhibitor, PV, induced MST3 tyrosine phosphorylation in cells. This finding suggests that tyrosine phosphorylation controls MST3 regulation in intact cells. Hence, we examined MST3 activity with this unusual tyrosine modification to further determine MST3 regulation. Several parameters affected MST3 activity in MST3 kinase assays. Our results showed that the presence of non-ionic detergent (Triton) and the ionic strength (NaCl or KCl) of buffers had adverse effects on MST3 activity. When 150 mM NaCl or KCl was added to the buffer, MST3 activity decreased, both in HEK293 cells exogenously expressing MST3-HA and in active purified baculovirus-expressed MST3 (data not shown). We modified the MST3 kinase assay by reducing the non-ionic detergent concentra- tion and ionic strength of the lysis buffer and found that MST3 activity increased by approximately 3–4-fold upon OA treatment. However, we did not observe an increase or decrease in MST3 activity upon PV treatment.
Okadaic acid (OA) is used to inhibit protein-serine/threonine phos- phatase (PSP), including PP1, PP2A and PP2B [36]. OA had been reported to inhibit PP2A by interacting with Cys 269 in the PP2A catalytic subunit [37]. In OA-treated cells, MST3 had an approximately 5-fold increase of kinase activity; therefore, OA was chosen to serve as the positive control for the MST3 kinase assay in this paper [38]. Activation of MST3 by OA stimulated autophosphorylation of MST3-T178 in the catalytic domain with simultaneous cis-autophosphorylation of MST3-T328 in the regulatory domain. MST3-T178 phosphorylation increased MST3 kinase activity, but this activity was independent of MST3-T328 phosphoryla- tion. MST3-T328 phosphorylation was necessary for the formation of the activated MST3-MO25 scaffold complex. Inactivated MST3 co- immunoprecipitated with the Golgi protein. However phosphorylated MST3-T178/T328 dissociated from the Golgi protein, resulting in MST3 and MO25 association [39]. The phosphorylation of MST3 at different residues played different roles in the regulation of MST3. MST3 tyrosine phosphorylation implicates MST3 in more diverse signaling processes.
Upon PV treatment, MST3 activity was not increased or decreased in our study. Vanadate forms different species depending on its concentra- tion and the pH value. These different species of vanadium have been reported to induce a variety of biological effects. We prepared PV at pH 10, with the monomer being the favored species. However, addition- al oligomers or PV decomposition may occur in limited concentrations in the initial prepared solution or upon dilution into the cell medium at physiologic pH. Decavanadate inhibits the rate of G-actin polymeriza- tion at several micromolar as observed by 51V NMR (nuclear magnetic resonance) spectroscopy studies [40]. Vanadate oligomers also affects catalase activity, lipid peroxidation, mitochondrial superoxide anion production and oxygen species (ROS) in the biological system [41]. These results indicated that different vanadate species might be respon- sible for the lack of change of tyrosine-phosphorylated MST3 activity upon PV treatment in this study, which is yet to be elucidated.
Vanadate interacts with glucose and NAD to form glucose-6-vanadate or NADV, respectively. The NADV appeared to be the NADP an- alog with high Kat/Km for alcohol dehydrogenase [42,43]. Vanadate was also reported to interact with phenol and N-acetyltyrosine ethyl ester in aqueous solution using 51 V NMR spectroscopy. The value of the equi- librium constant of esterification of phenol by vanadate is over 4 orders of magnitude larger than that of esterification of phenol by phosphate. The rapid vanadate esterification of phenol provides an interesting explanation for the action of vanadate on systems that involve the activation of enzymes by phosphorylation on tyrosine [44]. In addition, vanadate is a potent insulin-mimetic agent that activates the insulin receptor kinase [45–47]. It has been proposed that vanadate acts as a phosphate analog to esterify the insulin receptor hydroxyl groups on ty- rosine residues, imitating the effect of a tyrosine kinase enzyme [44,48]. The formation of MST3 vanadation on serine/threonine/tyrosine resi- dues of MST3 upon PV treatment was intriguing. Vanadate is mostly known as a specific inhibitor of phosphotyrosyl phosphatases (PTPases). PV appears to be more effective than vanadate in increasing the level of cellular tyrosine phosphorylation [21–24]. When adipocytes were incubated with isotope [32P]-containing medium, PV stimulated [32P] incorporation into the insulin receptor, indicating that the insulin receptor was phosphorylated [22]. Additionally, in our experiment, MST3 tyrosine phosphorylation was detected by 4G10 antibody in intact cells treated with PV. PV helps to elucidate that MST3 tyrosine phosphorylation in intact cells, suggesting that tyrosine phosphoryla- tion controls MST3 regulation in intact cells.
During cell lysis disruption, to avoid substantially diminished quan- tity of MST3 phosphorylation for the kinase assay, (i) Na3VO4, (ii) DTT,
(iii) HEPES, and (iv) EDTA were included in the cell lysis buffer.
(i) Na3VO4 served as a PTP inhibitor after cell disruption. (ii) To inhibit the function of PTPs, the thiol group of PTPs had to be reduced to form thiol-vanadate ester linkage with Na3VO4. DTT was added as a reducing reagent to protect the thiol groups of PTPs [49,50]. In addition, DTT could also retain the kinase activity. The thiol groups of Cys residues in kinases are important for interacting with ATP [51]. Conversion of these thiol groups to oxidized species might change the catalytic prop- erties of kinases, including MST3 [52]. (iii) Because HEPES was one of the few buffers that does not interact with Na3VO4, HEPES was chosen for use in the lysis buffer. (iv) EDTA in the buffer makes enzymes, such as proteases or DNAses, unavailable to function by chelating diva- lent cations. Huyer et al. also reported that when 100 μM Na3VO4 and 2 mM EDTA (same concentration as we used in the lysis buffer) were present during cell lysis, a much lower level of total phosphotyrosyl pro- teins was observed compared with a lysate without EDTA in the buffer [21]. Although, in such formulation, the endogenous PTPs were not completely inhibited, the MST3 tyrosine phosphorylation could still be measured by antibody in our study. This result suggested that the amount of Na3VO4 affected by EDTA might be negligible when detecting phosphorylation status. However, it might affect the MST3 kinase activity as we demonstrated here.
Membrane protein fractions can be divided into Triton-soluble and Triton-insoluble fractions. Phospholipase D1 (PLD1) in the Triton- soluble fraction showed little tyrosine phosphorylation, but PLD1 in the Triton-insoluble fraction was heavily tyrosine phosphorylated [53]. In addition, the activity of EGF-stimulated protein in the 0.15% Triton- insoluble fraction of A431 cells was nearly 3-fold greater than that of the 0.15% Triton-soluble fraction [54]. Furthermore, when ionic strength was increased in the homogenization buffer by adding NaCl or KCl, dynamin and P130 were released from the membrane into the cytosol, allowing for their phosphorylation by different proteins [55,56]. There- fore, the Triton concentration and the ionic strength in the lysis buffer have a striking effect on the protein distribution. MST3 has been found to exhibit several different subcellular localizations. MST3 contains both a nuclear localization sequence and a nuclear export sequence, allowing MST3 to continuously shuttle between the nucleus and the cytoplasm. Because both MST3 extracted from cells and purified MST3 were sensitive to Triton and NaCl, the effect of Triton and NaCl on MST3 activity was less likely due to different spatial localization by Triton solubility or ionic strength of the buffers.
Bovine RNAase A shifted from its monomeric to dimeric form in response to ionic strength. Monomeric RNAase A is significantly more active than dimeric RNAase A [23]. Incubation of dimeric creatine kinase with high NaCl or LiCl concentrations results in dissociation of the subunits and enzyme inactivation [57]. These results indicated that salt concentration affects kinase oligomerization and kinase activity. OA-treated MST maintains its phosphorylation and tends to remain as a monomer, whereas unphosphorylated MST forms a dimer. MST dimerization contributes to its cytoplasmic retention [35]. Our data showed that high NaCl or KCl concentrations inhibited MST3 activity (Fig. 6). Thus, we propose that intramolecular MST3 dynamics (mono- mer to dimer) depend on the ionic strength in the solution and affects MST3 activity.
The SH2 domains of tyrosine kinase interacted with phosphotyrosine residues of receptor tyrosine kinase tails, whereas tyrosine kinase SH3 domains recognized proline-rich motifs with the consensus sequence Pro-X-X-Pro (P-X-X-P) of their interacting partners [13–18]. The proline-rich motifs of the Ste20 family member HPK1 interacted with Grb2 SH3 domains in Cos1 cells. EGF stimulation induced HPK1 tyrosine phosphorylation and its recruitment into a stable complex with autophosphorylated EGF receptor. Kinase-dead HPK1 could still interact with the EGF receptor, suggesting that HPK1 served as an adapter in EGFR signaling [19]. Our results showed that MST3 was tyrosine phos- phorylated in intact cells upon PV treatment (Fig. 7), but tyrosine phos- phorylation had no effect on MST3 activity (Fig. 7). Peptide sequence comparison revealed a proline-rich motif in MST3 (356P-K-R-P359). Through its proline-rich motif, MST3 may interact with the SH2 and SH3 domains of tyrosine kinase, allowing for its tyrosine phosphorylation. It is also reasonable to hypothesize that the proline-rich motif of MST3 brought it in close proximity of the SH3 domains of adaptor proteins and that the adaptor recruited tyrosine kinase to phosphorylate MST3.
We did not observe any change in MST3 activity upon tyrosine-phosphorylation. However, MST3 tyrosine phosphorylation may play a role in adaptor protein recruitment for subsequent signal transduction. The exact cellular function of MST3 tyrosine phosphorylation is un- known. The identification of novel binding adaptor proteins that traffic BGB 15025 MST3 to signaling microdomains in various subcellular compart- ments will elucidate distinct aspects of MST3 signaling.

References
[1] M. Versele, K. Lemaire, J.M. Thevelein, EMBO Rep. 2 (2001) 574–579.
[2] E.A. Elion, Curr. Opin. Microbiol. 3 (2000) 573–581.
[3] I. Dan, N.M. Watanabe, A. Kusumi, Trends Cell Biol. 11 (2001) 220–230.
[4] E. Yamada, K. Tsujikawa, S. Itoh, Y. Kameda, Y. Kohama, H. Yamamoto, Biochim. Biophys. Acta 1495 (2000) 250–262.
[5] C.M. Pombo, T. Force, J. Kyriakis, E. Nogueira, M. Fidalgo, J. Zalvide, Front. Biosci. 12 (2007) 850–859.
[6] P. Ling, T.J. Lu, C.J. Yuan, M.D. Lai, Cell. Signal. 20 (2008) 1237–1247.
[7] G. Manning, D.B. Whyte, R. Martinez, T. Hunter, S. Sudarsanam, Science 298 (2002) 1912–1934.
[8] J.L. Lin, H.C. Chen, H.I. Fang, D. Robinson, H.J. Kung, H.M. Shih, Oncogene 20 (2001) 6559–6569.
[9] Z.G. Goldsmith, D.N. Dhanasekaran, Oncogene 26 (2007) 3122–3142.
[10] O.A. Coso, M. Chiariello, G. Kalinec, J.M. Kyriakis, J. Woodgett, J.S. Gutkind, J. Biol. Chem. 270 (1995) 5620–5624.
[11] D.N. Dhanasekaran, K. Kashef, C.M. Lee, H. Xu, E.P. Reddy, Oncogene 26 (2007) 3185–3202.
[12] D.K. Luttrell, L.M. Luttrell, Oncogene 23 (2004) 7969–7978.
[13] A. Musacchio, T. Gibson, V.P. Lehto, M. Saraste, FEBS Lett. 307 (1992) 55–61.
[14] T. Pawson, P. Olivier, M. Rozakis-Adcock, J. McGlade, M. Henkemeyer, Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci. 340 (1993) 279–285.
[15] N. Goudreau, F. Cornille, M. Duchesne, F. Parker, B. Tocque, C. Garbay, B.P. Roques, Nat. Struct. Biol. 1 (1994) 898–907.
[16] W.A. Lim, F.M. Richards, Nat. Struct. Biol. 1 (1994) 221–225.
[17] S. Feng, J.K. Chen, H. Yu, J.A. Simon, S.L. Schreiber, Science 266 (1994) 1241–1247.
[18] H. Yu, M.K. Rosen, T.B. Shin, C. Seidel-Dugan, J.S. Brugge, S.L. Schreiber, Science 258 (1992) 1665–1668.
[19] M. Anafi, F. Kiefer, G.D. Gish, G. Mbamalu, N.N. Iscove, T. Pawson, J. Biol. Chem. 272 (1997) 27804–27811.
[20] T.J. Lu, W.Y. Lai, C.Y. Huang, W.J. Hsieh, J.S. Yu, Y.J. Hsieh, W.T. Chang, T.H. Leu, W.C. Chang, W.J. Chuang, M.J. Tang, T.Y. Chen, T.L. Lu, M.D. Lai, J. Biol. Chem. 281 (2006) 38405–38417.
[21] G. Huyer, S. Liu, J. Kelly, J. Moffat, P. Payette, B. Kennedy, G. Tsaprailis, M.J. Gresser, C. Ramachandran, J. Biol. Chem. 272 (1997) 843–851.
[22] I.G. Fantus, S. Kadota, G. Deragon, B. Foster, B.I. Posner, Biochemistry 28 (1989) 8864–8871.
[23] M. Libonati, G. Gotte, Biochem. J. 380 (2004) 311–327.
[24] D. Heffetz, I. Bushkin, R. Dror, Y. Zick, J. Biol. Chem. 265 (1990) 2896–2902.
[25] A. Ruggiero, J.P. Holland, J.S. Lewis, J. Grimm, J. Nucl. Med. 51 (2010) 1123–1130. [26] H.H. Ross, Anal. Chem. Am. Chem. Soc. 41 (1969) 1260–1265.
[27] J.V. Jelley, Br. J. Appl. Phys. 6 (1955) 227.
[28] P.A. Cherenkov, Dokl. Akad. Nauk SSSR 2 (1934) 451.
[29] T.-J. Lu, C.-Y.F. Huang, C.-J. Yuan, Y.-C. Lee, T.-H. Leu, W.-C. Chang, T.-L. Lu, W.-Y. Jeng, M.-D. Lai, J. Inorg. Biochem. 99 (2005) 1306–1313.
[30] E. Tsiani, E. Bogdanovic, A. Sorisky, L. Nagy, I.G. Fantus, Diabetes 47 (1998) 1676–1686.
[31] S.J. Haque, Q. Wu, W. Kammer, K. Friedrich, J.M. Smith, I.M. Kerr, G.R. Stark, B.R. Williams, Proc. Natl. Acad. Sci. U. S. A. 94 (1997) 8563–8568.
[32] J.P. Secrist, L.A. Burns, L. Karnitz, G.A. Koretzky, R.T. Abraham, J. Biol. Chem. 268 (1993) 5886–5893.
[33] S.H. Jang, S.A. Hwang, M. Kim, S.H. Yun, M.S. Kim, S.S. Karnik, C. Lee, Mol. Cells 28 (2009) 25–30.
[34] M.C. Louzao, M.R. Vieytes, L.M. Botana, Mini Rev. Med. Chem. 5 (2005) 207–215. [35] K.K. Lee, S. Yonehara, J. Biol. Chem. 277 (2002) 12351–12358.
[36] A. Takai, M. Murata, K. Torigoe, M. Isobe, G. Mieskes, T. Yasumoto, Biochem. J. 284 (Pt 2) (1992) 539–544.
[37] M. Nagao, H. Shima, M. Nakayasu, T. Sugimura, Mutat. Res. 333 (1995) 173–179.
[38] M.R. Stegert, A. Hergovich, R. Tamaskovic, S.J. Bichsel, B.A. Hemmings, Mol. Cell. Biol. 25 (2005) 11019–11029.
[39] S.J. Fuller, L.J. McGuffin, A.K. Marshall, A. Giraldo, S. Pikkarainen, A. Clerk, P.H. Sugden, Biochem. J. 442 (2012) 595–610.
[40] S. Ramos, M. Manuel, T. Tiago, R. Duarte, J. Martins, C. Gutierrez-Merino, J.J. Moura,
M. Aureliano, J. Inorg. Biochem. 100 (2006) 1734–1743.
[41] M. Aureliano, R.M. Gandara, J. Inorg. Biochem. 99 (2005) 979–985.
[42] A.F. Nour-Eldeen, M.M. Craig, M.J. Gresser, J. Biol. Chem. 260 (1985) 6836–6842.
[43] D.C. Crans, R.W. Marshman, R. Nielsen, I. Felty, J. Org. Chem. Am. Chem. Soc. 58 (1993) 2244–2252.
[44] A.S. Tracey, M.J. Gresser, Proc. Natl. Acad. Sci. U. S. A. 83 (1986) 609–613.
[45] I.G. Fantus, R. George, S. Tang, P. Chong, M.J. Poznansky, Diabetes 45 (1996) 1084–1093.
[46] S. Kadota, I.G. Fantus, G. Deragon, H.J. Guyda, B. Hersh, B.I. Posner, Biochem. Biophys. Res. Commun. 147 (1987) 259–266.
[47] M. Bernier, D.M. Laird, M.D. Lane, J. Biol. Chem. 263 (1988) 13626–13634.
[48] R.M. Cruz-Mireles, I. Ortega-Blake, Plant Physiol. 96 (1991) 91–97.
[49] D.J. Kemble, G. Sun, Proc. Natl. Acad. Sci. U. S. A. 106 (2009) 5070–5075.
[50] S. Kumar, B. Zhou, F. Liang, W.Q. Wang, Z. Huang, Z.Y. Zhang, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 7943–7948.
[51] R. Dhand, I. Hiles, G. Panayotou, S. Roche, M.J. Fry, I. Gout, N.F. Totty, O. Truong, P. Vicendo, K. Yonezawa, et al., EMBO J. 13 (1994) 522–533.
[52] C. Klomsiri, P.A. Karplus, L.B. Poole, Antioxid. Redox Signal. 14 (2011) 1065–1077.
[53] W.T. Ho, Z. Xie, Z.J. Zhao, J.H. Exton, Cell. Signal. 17 (2005) 691–699.
[54] A.M. Gronowski, P.J. Bertics, Endocrinology 136 (1995) 2198–2205.
[55] R.A. Feldman, E. Wang, H. Hanafusa, J. Virol. 45 (1983) 782–791.
[56] R. Lundmark, S.R. Carlsson, J. Biol. Chem. 279 (2004) 42694–42702.
[57] F. Couthon, E. Clottes, C. Vial, Biochim. Biophys. Acta 1339 (1997) 277–288.