- Email: [email protected]
The Synthesis of an O,C-Trisaccharide: The O,C-Analog of Methyl 4′-O-β-d -glucopyranosylgentiobioside
The cysteine-cluster motif of c-Src: Its role for the heavy metal-mediated activation of kinase Blackwell Publishing Asia
Takeshi Senga, Hitoki Hasegawa, Miwa Tanaka, Mohammad Aminur Rahman, Satoko Ito and Michinari Hamaguchi1 Division of Cancer Biology, Nagoya University School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya 466-8550, Japan (Received September 6, 2007/Revised November 9, 2007/Accepted November 19, 2007/Online publication January 2, 2008)
We have previously reported that c-Src is activated by mercuric chloride (HgCl2). We investigated the mechanism of this activation and found that in vitro activation of c-Src by HgCl2 did not require tyrosine residues at 416 and 527. Both SH2 and SH3 domains of c-Src were dispensable for the activation by HgCl2. In contrast, iodoacetoamide (IAA) that binds to thiol side chain of cysteine blocked the activation of c-Src by HgCl2. To obtain more clues, each cysteine residue of c-Src was replaced with alanine. Of six cysteine residues in the kinase domain of c-Src, Cys483 and Cys498 located in the C-terminal portion as a cysteine-cluster (CC) motif were critical for the activation. In addition, other Src family kinases, Yes and Lyn, were activated by treatment with HgCl2, and cysteine residues, especially those correspond to Cys498 of Src in the CC motif of these kinases, were also required for the activation of the kinases by HgCl2. In addition to these observations, treatment of cells with HgCl2 dramatically activated the wild-type c-Src, whereas it could not activate the mutant form of Src with a substitution of Cys498. Taken together, our results disclose that cysteine residues in the CC motif of c-Src, Cys483 and Cys498, act as a module for the activation of the kinase by a heavy metal compound, mercuric chloride. (Cancer Sci 2008; 99: 571–575)
c
-Src, a cellular counterpart of viral oncogene product, v-Src, is a non-receptor tyrosine kinase distributed widely in tissues.(1) Activation of c-Src is frequently observed in various human carcinomas including those of breast and colon carcinomas and is critical for their metastasis,(2) suggesting the importance of Src signaling in human malignant tumors. Regulation of kinase activity of c-Src has been well characterized. Phosphorylation of tyrosine 527 (Tyr527) located near the Cterminus appears to down-regulate the kinase. Point mutation of Tyr527 renders c-Src active and highly oncogenic, (3–5) whereas targeted disruption of C-terminal SRC kinase (CSK),(6) the kinase that phosphorylates Tyr527, caused activation of c-Src.(7) Studies of the three-dimensional structure of c-Src revealed that interaction between SH2 and phosphorylated Tyr527 together with SH3 and linker region folded the molecule in a closed, inactive state.(8,9) In addition to the Tyr527-based regulation, increasing evidence suggests that the c-Src kinase has a redox-linked regulatory mechanism.(10) One of the models of the redox-linked regulation, we found, is the activation of c-Src by mercuric chloride (HgCl2).(11) Treatment of purified c-Src with mercuric chloride changed Vmax and Km of the kinase, suggesting the kinase activation to be a direct one. (11) Mercuric chloridetreatment of cells activates signal pathways leading to tyrosine phosphorylation of cellular proteins,(12) and activation of mitogenactivated protein (MAP) kinases,(13–16) which in turn induce the expression of cytokines,(17) unregulated cell growth and DNA synthesis.(13,18) Consistently, treatment of c-Src with mercuric chloride activates the kinase.(11) However, the molecular mechanism by which mercuric chloride activates c-Src remains largely unclear. Mercuric ions are widely known as a protein-modifying agent specific for the sulfhydryl group and have an exceptionally high doi: 10.1111/j.1349-7006.2007.00714.x © 2008 Japanese Cancer Association
association constant with sulfhydryl groups in proteins.(19,20) Indeed, we found that N-acetylcysteine could interfere with the activation of c-Src by mercuric chloride both in vitro and in vivo.(11) As a step to understand the molecular mechanism of HgCl2-dependent activation of c-Src, we explored the effect of HgCl2 on a series of mutant c-Src. In this report, we show evidence that the activation of c-Src by HgCl2 requires neither SH2/SH3 domains nor phosphorylation on Tyr416 and Tyr527, but does require some of cysteine residues of c-Src. Among nine cysteine residues scattered over c-Src, Cys483 and Cys498 located in the cysteine-cluster (CC) motif, but not others, are critical for the HgCl2-dependent kinase activation. Other Src family kinases such as Lyn and c-Yes also have similar CC motifs and require cysteine residues at similar positions to be activated by HgCl2. Furthermore, we show that the activation of c-Src by HgCl2 in vivo also requires Cys498. Materials and Methods Cell culture and transfection. COS-7 cells and SYF cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS). COS-7 cells were transfected with diethylaminoethyl (DEAE) dextran. Briefly, plasmid DNA was mixed with transfection reagent (DMEM with 10% NuSerum, 0.5 mg/mL of DEAE dextran, 0.1 mM of chloroquine) and added to cells. Three hours after transfection, 10% dimethyl sulfoxide (DMSO) in phosphate-buffered saline (PBS) was added and incubated for 2 min, and then replaced with fresh media. Two days later, expression of Src was confirmed by immunoblotting. SYF cells are cell lines derived from mouse embryonic fibroblast in which c-src, c-yes and lyn genes are target-disrupted. For transfection to SYF cells, Lipofectamine 2000 (Invitrogen) was used according to the manufacturer’s protocol. Construction of mutant c-Src, c-Yes and Lyn kinases. Full length of c-src was amplified by polymerase chain reaction (PCR) and ligated into pcDNA3 (Invitrogen). lyn and yes were kindly provided by Dr T Yamamoto (University of Tokyo). Substitution of amino acids was performed as previously described.(21) Point mutation of each mutant gene was confirmed by sequencing. ΔSH2SH3 src was constructed by fusing PCR-amplified fragments of membrane-binding domain and kinase domain of c-src. In vitro kinase assay. In vitro kinase assay was performed as described previously.(22) In brief, cells were lyzed in RIPA buffer (20 mM Tris-HCl pH 7.4, 150 mM NaCl, 1% NP-40, 1% deoxycholic acid, 0.1% sodium dodecyl sulfate) and immunoprecipitated with anti-Src antibody. Immunoprecipitates were washed, suspended with kinase buffer (10 mM Tris-HCl pH 7.4, 5 mM MgCl2) containing [γ32 p] ATP (370 kBq) (NEN, Wilmington, DE, USA) and 1.5 μg of acid-treated enolase as an exogenous substrate. The kinase reaction was performed in the
1
To whom correspondence should be addressed. E-mail: [email protected]
Cancer Sci |
March 2008 | vol. 99 | no. 3 | 571–575
presence or absence of HgCl2 for 10 min at 30°C and subjected to electrophoresis. Relative activities were assayed by BAS 2000 system. Iodoacetoamide (IAA) was added to the immunoprecipitates and incubated for 10 min on ice. After washing three times with kinase buffer to eliminate these chemicals, kinase reaction was performed. To examine the kinase activity after HgCl2 stimulation in vivo, SYF cells were transfected with plasmids that express either wild type c-Src or C498A Src. Twenty-four hours later, cells were stimulated with HgCl2 for 2 min, and Src was immunoprecipitated and subjected to in vitro kinase assay; 1 mM of L-cysteine was added during immunoprecipitation to inhibit the further modification of Src by HgCl2. Anti-Src monoclonal antibody 327 was used for immunoprecipitation of c-Src. Anti-Lyn and anti-Yes was purchased from BD Biosciences. Results Activation of c-Src by HgCl2 is sulfhydryl-dependent but does not require SH2, SH3, Tyr416 and Tyr527. We first confirmed the effect
of HgCl2 treatment on the kinase activity of c-Src in vitro. c-Src was immunoprecipitated from either COS-7 cells transfected with c-src expressing plasmid or Sf9 cells infected with recombinant baculovirus containing c-src gene. The in vitro kinase assay was performed with enolase as an exogenous substrate. As shown in Fig. 1(a), addition of HgCl2 to the kinase reaction mixtures enhanced the catalytic activity of c-Src from both sources as we previously reported.(11) In the following experiments, we used c-Src immunoprecipitated from COS-7 cells for the in vitro kinase assay. Mercuric ions are known to bind to free sulfhydryl groups of proteins with a high association constant.(19,20) We next investigated
Fig. 1. Activation of the c-Src kinase in vitro by mercuric chloride (HgCl2) is sulfhydryl-dependent. (a) Immunoprecipitated c-Src, or c-Src purified from Sf9 cells were subjected to in vitro kinase assay in the presence or absence of HgCl2. (b) Inhibition of HgCl2-dependent activation of c-Src by l-cysteine. (c) Inhibition of HgCl2-dependent activation of c-Src by iodoacetoamide. (d) Activation of mutant Src kinases by HgCl2. Y416F has substitution of Tyr416 to phenylalanine and Y527F has Tyr527 to alanine. ΔSH2/SH3 is deleted of SH2 and SH3 domains of cSrc. (e) Effects of heavy metals on the kinase activity of c-Src.
572
the effect of l-cysteine and IAA) on the activation of c-Src by HgCl2. As shown in Fig. 1(b), addition of l-cysteine to the reaction mixtures blocked the HgCl2-dependent activation of c-Src. Some of the SH-alkylating agents such as IAA and N-ethylmaleimide (NEM) bind to sulfhydryl groups of cysteine residues of v-Src but do not inhibit kinase activity.(23) We found that pretreatment of c-Src with IAA did not inactivate the kinase but blocked the activation by HgCl2 (Fig. 1c). A similar inhibitory effect on the HgCl2-dependent activation was also observed with NEM (data not shown). Autophosphorylation of tyrosine 416 of c-Src (Tyr416) and dephosphorylation of tyrosine 527 (Tyr527) appears to be crucial for the activation of c-Src.(3) To investigate the involvement of phosphorylation of these tyrosine residues in the HgCl2-dependent activation of c-Src, we constructed mutant Src proteins, Y416F Src and Y527F Src, whose Tyr416 and Tyr527 were replaced with phenylalanines by site-directed mutagenesis, respectively. Although Y527F Src had higher kinase activity as compared to that of wild-type c-Src, addition of HgCl2 to reaction mixture further activated the catalytic activity of Y416F Src and Y527F Src (Fig. 1d). To confirm and extend these observations, another mutant that lacks SH2 and SH3 domains (ΔSH2/SH3) of c-Src was constructed. ΔSH2/SH3 Src has a deletion of residues 86–246 of c-Src. As shown in Fig. 1(d), ΔSH2/SH3 Src was activated by HgCl2 to a level similar to that of wild-type c-Src. These results imply that Tyr416, Tyr527 and the SH2/SH3 domains are dispensable for the activation of c-Src by HgCl2. We also examined effects of other heavy metals such as lead, copper and iron on the kinase activity of c-Src. As shown in Fig. 1(e), these metals did not affect the catalytic activity of c-Src at the concentration where HgCl2 clearly activated the kinase. Identification of cysteine residues critical for the activation of c-Src by HgCl2. Since HgCl2 is a sulfhydryl group modifier and IAA
could block the activation of c-Src, we next explored the role of each cysteine residue in HgCl2-dependent activation by constructing a series of mutant c-Src in which each cysteine residue in the kinase domain of c-Src was replaced with alanine by site-directed mutagenesis. Of nine cysteine residues in c-Src, Cys185, Cys238 and Cys245 are in the SH2 domain and the other six cysteine residues are in the kinase domain. Since the SH2 domain was dispensable for HgCl2-dependent activation, cysteine residues in the kinase domain were replaced by alanines. C277A, C400A, C483A, C487A, C496A and C498A Src proteins have substitutions of Cys277, Cys400, Cys483, Csy487, Cys496 and Cys498 to alanine, respectively. Among these six cysteines in the kinase domains, Cys483, Cys487, Cys496 and Cys498, are located near the C-terminus and form a cluster,(24) so that we named this motif as the CC (cysteine-cluster) motif (Fig. 2). Other members of the Src-family kinase have similar cysteine clusters, although Src has the largest number of cysteine residues in this motif. It should be noted that this CC motif was critical for cell transformation by v-Src.(21) Based on the number of cysteine residues in the CC motif, the Src-family kinases can be classified into three groups (Fig. 2). Group 1, including Src, Yes and Fyn, has both cysteine residues corresponding to Cys 483 and Cys498 of c-Src. Group 2, including Lck, Lyn and Hck, has the cysteine residue corresponding to Cys 498 of c-Src, but lacks the one corresponding to Cys483. Group 3 (Fgr) lacks both the residues corresponding to Cys 483 and Cys498 of c-Src. The mutant src genes were ligated into pcDNA3 vector and transiently expressed in COS-7 cells. Expression of each mutant was confirmed by immunoblotting and catalytic activity was examined by in vitro kinase assay (Fig. 3). Except for C498A, these mutant Src proteins showed similar levels of kinase activity. Expression of C498A Src was similar to that of others, but its kinase activity was reduced to half of wild-type c-Src (Fig. 3). We examined the effect of HgCl2 treatment on the kinase activity of these mutant Src proteins. As shown in doi: 10.1111/j.1349-7006.2007.00714.x © 2008 Japanese Cancer Association
Fig. 2. Structure of the cysteine-cluster motif of Src family kinases.
Fig. 4. Activation of mutant c-Src by mercuric chloride (HgCl2) (a) Activation of the wild-type and mutant Src kinases by HgCl2 was assayed by in vitro kinase assay. (b) Kinase activities of C498A Src and C498A/Y527F Src. (c) Kinase activation of ΔSH2/SH3C277/400/487/496 A Src, ΔSH2/SH3C483 A Src and ΔSH2/SH3C498 A Src by HgCl2.
and Cys498 to alanines, respectively. As shown in Fig. 4(c), ΔSH2/SH3C277/400/487/496A Src, which has only two cysteine residues at 483 and 498, was activated by HgCl2. In contrast, ΔSH2/SH3C483A Src and rSH2/SH3C498ASrc were resistant to the activation by HgCl2. Fig. 3. Kinase activity of mutant c-Src. Expression of the wild-type and mutant c-Src kinases in COS7 were examined by immunoblotting with anti-Src (lower panel). Src proteins were immunoprecipitated and their kinase activities were assayed (upper panel).
Fig. 4(a), C277A, C400A, C487A and C496A Src were activated by HgCl2 to the levels similar to that of wild-type c-Src. In contrast, C483A and C498A Src were resistant to HgCl 2dependent activation. Although C483A Src was activated slightly by HgCl2, its activation was less than two-fold. On the other hand, C498A Src was completely resistant to the activation by HgCl2. Since C498A Src showed lower catalytic activity compared to any other mutant Src (Fig. 3), a possibility remained that the resistance of this protein to HgCl2 was non-specific. To exclude the possibility, we constructed another mutant Src (C498A/ Y527F Src) in which Tyr527 was replaced by phenylalanine in addition to the substitution of Cys 498 to alanine. As shown in Fig. 4(b), the kinase activity of C498A/Y527F Src was around 10 times higher than that of C498A Src. These results suggest that, although C498A had relatively low kinase activity, it could still be activated by the mutation of Tyr527 and that the resistance of C498A to HgCl2 is specific. These results also suggest that Tyr527-dependent regulation and HgCl2-dependent regulation of c-Src are mutually independent. To confirm that cysteines at 483 and 498 are not only critical but also sufficient for the activation of c-Src by HgCl2, we constructed another series of mutant Src. ΔSH2/SH3C277/400/487/ 496A Src has substitution of Cys277, Cys400, Cys487 and Cys496 to alanine together with deletion of SH2 and SH3 domains. ΔSH2/SH3C483A Src and rSH2/SH3C498A Src are ΔSH2/SH3 Src-derived mutants which have substitution of Cys483 Senga et al.
Requirement of cysteine residues in Yes and Lyn for HgCl2-dependent activation. As the cysteine residues critical for the HgCl2-
mediated activation were partly conserved among the Src family tyrosine kinases (Fig. 2), we investigated whether other members of the Src family tyrosine kinases could also be activated by HgCl2. Yes, which we classified as Group 1 together with c-Src, has three cysteines in the CC motif of the kinase domain. Cysteines at 491, 495 and 506 of Yes correspond to cysteines at 483, 487 and 498 of c-Src, respectively. As shown in Fig. 5(a), Yes was strongly activated by HgCl2. Both C491A Yes and C506A Yes that have substitution of either Cys491 or Cys506 to alanine are resistant to the activation by HgCl2. In contrast to Group 1 kinases, Lyn, which belongs to Group 2, has only two cysteines in the CC motif at 468 and 479 that correspond to cysteines at 487 and 498 of c-Src. Although Lyn lacks cysteine corresponding to Cys483 of c-Src, Lyn is activated by HgCl2. To investigate whether Lyn requires Cys479 corresponding to Cys498 of c-Src to be activated by HgCl2, we constructed Lyn mutant, C479A Lyn that has alanine in place of Cys479. As shown in Fig. 5(b), C479A Lyn became relatively resistant to HgCl2 treatment, although it showed weak activation. These results suggest that not only c-Src but also other members of the Src-family kinase can be activated by HgCl2, at least in part, in a manner dependent on the cysteine residues in the CC motif. Role of the CC motif in in vivo activation of c-Src by HgCl2. To examine whether the CC motif is indeed involved in HgCl2dependent activation of c-Src in vivo, we transiently expressed c-Src and C498A Src in SYF cells in which the c-src, c-yes and c-fyn genes were target-disrupted. It has been known that treatment of cells with HgCl2 dramatically activates tyrosine phosphorylation of cellular proteins. Addition of HgCl2 to SYF cells did not increase tyrosine phosphorylation of cellular proteins. However, we observed dramatic increase of tyrosine phosphorylation of proteins when SYF cells transfected with c-src Cancer Sci | March 2008 | vol. 99 | no. 3 | 573 © 2008 Japanese Cancer Association
expressing plasmid were treated with HgCl2 (Fig. 6a). Interestingly, expression of C498A Src in SYF cells did not show any increase of tyrosine phosphorylation of proteins when treated with HgCl2 (Fig. 6a). We checked catalytic activity of c-Src and C498A Src after HgCl2 treatment. Src proteins were immunoprecipitated in the presence of 1 mM l-cysteine to avoid the in vitro activation of kinase during immunoprecipitation. Kinase activities of the immunoprecipitates were assayed in vitro with enolase as an exogenous substrate. As shown in Fig. 6(b), kinase activity of wild-type c-Src was activated by HgCl2 approximately 10-fold compared to the untreated c-Src. In contrast, C498A Src was resistant to the activation by HgCl2. Discussion
Fig. 5. Activation of the Yes and Lyn kinases by mercuric chloride (HgCl2) (a) Activation of Yes and its mutants by HgCl2. C491A Yes has substitution of Cys491 to alanine and 506 A Yes has Cys506 to alanine. (b) Activation of Lyn and C479A Lyn by HgCl2. C479A Lyn has substitution of Cys 479 to alanine.
Fig. 6. Activation of the Src kinase and tyrosine phosphorylation of proteins in cells transfected with wild-type and mutant forms of c-src by mercuric chloride (HgCl2) (a) SYF cells, in which endogenous c-src, cyes, and fyn were disrupted, were transfected with wild-type c-src or C498A src. The day after transfection, cells were stimulated at the indicated concentration of HgCl2 for 2 min and cells were lyzed and blotted with antiphosphotyrosine and anti-Src antibodies. (b). SYF cells transfected with the indicated plasmids were stimulated with HgCl2 and subjected to in vitro kinase assay.
574
Heavy metals are one of the most toxic forms of environmental pollutants causing life-threatening problems on a worldwide scale. They have accumulated to hazardous levels in the air, land and water from the sludge produced by industries and population centers.(25) Among multiple heavy metal pollutants, mercuric chloride (HgCl2) is unique for its multifarious biological effects, including acute renal failure,(26) autoimmune glomerulonephritis,(27) systemic vasculitis,(28) and autoimmune diseases similar to human systemic lupus erythematosus.(29) Mercuric chloride induces a dramatic activation of the immune system and autoantibody production in susceptible animals. (30) It can activate signal pathways leading to tyrosine phosphorylation of cellular proteins,(12) and activation of MAP kinases,(13–16) which in turn induce the expression of cytokines,(17) unregulated cell growth and DNA synthesis.(13,18) In this paper, we report a novel regulatory mechanism of cSrc by HgCl2 that requires specific cysteine residues in the CC motif. Neither SH2/SH3 nor Tyr527 were required for the activation of c-Src by HgCl2, whereas kinase domain alone was sufficient for the activation. Mutational analysis found that Cys483 and Cys498 are critical for the activation by HgCl2. Point mutation of tyrosine at 527 of C498A Src that was resistant to the activation by HgCl 2 to phenylalanine clearly activated the kinase. These results suggest that the CC motifbased regulation of c-Src is independent of the Tyr527-based regulation of c-Src. Based on the presence of the cysteine residues corresponding to Cys483 and Cys498 of c-Src, the Src family kinases could be classified into three groups (Fig. 2). Group 1, including Src and Yes, has the cysteine residues corresponding to Cys483 and Cys498 of Src. As expected, the Yes kinase responded well to the activation by HgCl 2 and required both of the cysteine residues, suggesting that the CC motif-based activation of kinase by HgCl2 to be a common characteristic among the Group 1 kinases. In contrast to Group 1 kinases, Lyn, a Group 2 kinase, responded to the activation by HgCl2, despite the absence of cysteine corresponding to Cys483 of Src (Fig. 5b). The mechanism of kinase activation by HgCl2 might differ in part between Group 1 and Group 2 kinases, and some other cysteine residues might substitute the role of Cys483 of Src in cases of Group 2 kinase. Replacement of Cys479 of Lyn to alanine, however, converted the kinase to be resistant to HgCl2, again suggesting the importance of the CC motif for the HgCl 2-dependent regulation of Group 2 kinases. We have previously reported that mutation of cysteine residues in the CC motif rendered v-Src refractory to the inhibition by Herbimycin A, indicating that modification of cysteine residues could also inactivate the catalytic activity. Other SH-alkylating agents, N-[p-(2-benzimidazolyl)phenyl] maleimide (BIPM) and N-(9-acridinyl) maleimide (NAM), could also inactivate v-Src dependent on the modification of cysteine residues in the CC motif.(31) These results suggest that, depending on the type of modification, modification of cysteine doi: 10.1111/j.1349-7006.2007.00714.x © 2008 Japanese Cancer Association
residues in the CC motif could either activate or inactivate catalytic activity of Src. A previous report showed that monovalent mercuric compound such as p-chloromercuribenzoate (PCMB) did not change any kinase activity of c-Src as bivalent mercuric ions (Hg2+) did.(11) Positively charged bivalent mercuric ions bound to cysteine may affect adjacent charged amino acids, which could distort the structure of c-Src and enhance the catalytic activity. Otherwise, Hg2+ might bind two adjacent sulfhydryl groups and bridge two cysteines to increase the kinase activity of c-Src. Although further studies are required to elucidate the mechanism of kinase activation by mercuric ions, our results strongly suggest the unique function of the CC motif as a bifunctional regulatory unit. A large body of evidence suggests that HgCl 2 is a potent activator of the intracelluar signaling pathways leading to tyrosine
References 1 Martin GS. The hunting of the Src. Nat Rev Mol Cell Biol 2001; 2: 467–75. 2 Alvarez RH, Kantarjian HM, Cortes JE. The role of Src in solid and hematologic malignancies: development of new-generation Src inhibitors. Cancer 2006; 107: 1918–29. 3 Cartwright CA, Eckhart W, Simon S, Kaplan PL. Cell transformation by pp60c-src mutated in the carboxy-terminal regulatory domain. Cell 1987; 49: 83–91. 4 Kmiecik TE, Shalloway D. Activation and suppression of pp60c-src transforming ability by mutation of its primary sites of tyrosine phosphorylation. Cell 1987; 49: 65–73. 5 Piwnica-Worms H, Saunders KB, Roberts TM, Smith AE, Cheng SH. Tyrosine phosphorylation regulates the biochemical and biological properties of pp60c-src. Cell 1987; 49: 75–82. 6 Nada S, Okada M, MacAuley A, Cooper JA, Nakagawa H. Cloning of a complementary DNA for a protein-tyrosine kinase that specifically phosphorylates a negative regulatory site of p60c-src. Nature 1987; 351: 69– 72. 7 Imamoto A, Soriano P. Disruption of the csk gene, encoding a negative regulator of Src family tyrosine kinases, leads to neural tube defects and embryonic lethality in mice. Cell 1991; 73: 1117–24. 8 Xu W, Harrison SC, Eck MJ. Three-dimensional structure of the tyrosine kinase c-Src. Nature 1993; 385: 595–602. 9 Sicheri F, Moarefi I, Kuriyan J. Crystal structure of the Src family tyrosine kinase Hck. Nature 1997; 385: 602–9. 10 Giannoni E, Buricchi F, Raugei G, Ramponi G, Chiarugi P. Intracellular reactive oxygen species activate Src tyrosine kinase during cell adhesion and anchorage-dependent cell growth. Mol Cell Biol 1997; 25: 6391–403. 11 Pu M et al. Evidence of a novel redox-linked activation mechanism for the Src kinase which is independent of tyrosine 527-mediated regulation. Oncogene 2005; 13: 2615–22. 12 Rahman SM et al. Redox-linked ligand-independent cell surface triggering for extensive protein tyrosine phosphorylation. FEBS Lett 1996; 317: 35– 8. 13 Akhand AA, Kato M, Suzuki H, Miyata T, Nakashima I. Level of HgCl2mediated phosphorylation of intracellular proteins determines death of thymic T-lymphocytes with or without DNA fragmentation. J Cell Biochem 1993; 71: 243–53. 14 Yano T et al. The repetitive activation of extracellular signal-regulated kinase is required for renal regeneration in rat. Life Sci 1998; 62: 2341–7. 15 Turney KD, Parrish AR, Orozco J, Gandolfi AJ. Selective activation in the MAPK pathway by Hg (II) in precision-cut rabbit renal cortical slices. Toxicol Appl Pharmacol 1998; 160: 262–70.
Senga et al.
phosphorylation of cellular proteins which in turn induce unregulated cell growth, DNA synthesis and apoptosis. We found that wild-type c-Src was activated by treatment of cells with HgCl2 within 2 min, whereas C498A Src was resistant to the activation (Fig. 6b). These results suggest that the CC motif of c-Src is involved in the pathological disorders induced by HgCl 2. Our results also suggest that phenotype of human tumors with overexpressed c-Src might be modified under mercuric pollution. In summary, our results disclose that Cys483 and Cys498 in the CC motif of c-Src act as sensors for the activation of the kinase by a redox-linked mechanism. Further studies, including the structural analysis of the CC motif under stimulation by HgCl2, are required for the complete comprehension of the unique redox-linked regulatory mechanism of Src.
16 Matsuoka M, Wispriyono B, Iryo Y, Igisu H. Mercury chloride activates cJun N-terminal kinase and induces c-jun expression in LLC-PK1 cells. Toxicol Sci 1999; 53: 361–8. 17 Fournie GJ et al. Induction of autoimmunity through bystander effects. Lessons from immunological disorders induced by heavy metals. J Autoimmun 2000; 16: 319–26. 18 Akhand AA et al. Magnitude of protein tyrosine phosphorylation-linked signals determines growth versus death of thymic T lymphocytes. Eur J Immunol 2001; 27: 1254–9. 19 Stricks W, Kolthoff IM, Bush DG, Kuroda PK. Experimental study of the polarographic cancer test and of the sulfhydryl titration for the differentiation between normal, cancerous and other pathological sera. J Lancet 1997; 73: 328–35. 20 Sperling R, Burstein Y, Steinberg IZ. Selective reduction and mercuration of cystine IV–V in bovine pancreatic ribonuclease. Biochemistry 1953; 8: 3810–20. 21 Senga T et al. Clustered cysteine residues in the kinase domain of v-Src: critical role for protein stability, cell transformation and sensitivity to herbimycin A. Oncogene 1969; 19: 273–9. 22 Hamaguchi M et al. p60v-src causes tyrosine phosphorylation and inactivation of the N-cadherin-catenin cell adhesion system. Embo J 2000; 12: 307–14. 23 Uehara Y, Fukazawa H, Murakami Y, Mizuno S. Irreversible inhibition of vsrc tyrosine kinase activity by herbimycin A and its abrogation by sulfhydryl compounds. Biochem Biophys Res Commun 1993; 163: 803–9. 24 Veillette A, Dumont S, Fournel M. Conserved cysteine residues are critical for the enzymatic function of the lymphocyte-specific tyrosine protein kinase p56lck. J Biol Chem 1989; 268: 17547–53. 25 Nriagu JO, Pacyna JM. Quantitative assessment of worldwide contamination of air, water and soils by trace metals. Nature 1993; 333: 134–9. 26 Pfaller W, Gstraunthaler G, Willinger CC. Morphology of renal tubular damage from nephrotoxins. Toxicol Lett 1988; 53: 39–43. 27 Kanfer A. Role of coagulation in glomerular injury. Toxicol Lett 1990; 46: 83–92. 28 Luzina IG, Handwerger BS. Lessons from animal models of vasculitis. Curr Rheumatol Rep 1989; 2: 369–75. 29 White KL, David DW, Butterworth LF, Klykken PC. Assessment of autoimmunity-inducing potential using the brown Norway rat challenge model. Toxicol Lett 2000; 112–113: 443–51. 30 Bagenstose LM, Salgame P, Monestier M. Murine mercury-induced autoimmunity: a model of chemically related autoimmunity in humans. Immunol Res 2000; 20: 67–78. 31 Oo ML et al. Cysteine residues in the C-terminal lobe of Src: their role in the suppression of the Src kinase. Oncogene 1999; 22: 1411–17.
Cancer Sci | March 2008 | vol. 99 | no. 3 | 575 © 2008 Japanese Cancer Association
Takeshi Senga, Hitoki Hasegawa, Miwa Tanaka, Mohammad Aminur Rahman, Satoko Ito and Michinari Hamaguchi1 Division of Cancer Biology, Nagoya University School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya 466-8550, Japan (Received September 6, 2007/Revised November 9, 2007/Accepted November 19, 2007/Online publication January 2, 2008)
We have previously reported that c-Src is activated by mercuric chloride (HgCl2). We investigated the mechanism of this activation and found that in vitro activation of c-Src by HgCl2 did not require tyrosine residues at 416 and 527. Both SH2 and SH3 domains of c-Src were dispensable for the activation by HgCl2. In contrast, iodoacetoamide (IAA) that binds to thiol side chain of cysteine blocked the activation of c-Src by HgCl2. To obtain more clues, each cysteine residue of c-Src was replaced with alanine. Of six cysteine residues in the kinase domain of c-Src, Cys483 and Cys498 located in the C-terminal portion as a cysteine-cluster (CC) motif were critical for the activation. In addition, other Src family kinases, Yes and Lyn, were activated by treatment with HgCl2, and cysteine residues, especially those correspond to Cys498 of Src in the CC motif of these kinases, were also required for the activation of the kinases by HgCl2. In addition to these observations, treatment of cells with HgCl2 dramatically activated the wild-type c-Src, whereas it could not activate the mutant form of Src with a substitution of Cys498. Taken together, our results disclose that cysteine residues in the CC motif of c-Src, Cys483 and Cys498, act as a module for the activation of the kinase by a heavy metal compound, mercuric chloride. (Cancer Sci 2008; 99: 571–575)
c
-Src, a cellular counterpart of viral oncogene product, v-Src, is a non-receptor tyrosine kinase distributed widely in tissues.(1) Activation of c-Src is frequently observed in various human carcinomas including those of breast and colon carcinomas and is critical for their metastasis,(2) suggesting the importance of Src signaling in human malignant tumors. Regulation of kinase activity of c-Src has been well characterized. Phosphorylation of tyrosine 527 (Tyr527) located near the Cterminus appears to down-regulate the kinase. Point mutation of Tyr527 renders c-Src active and highly oncogenic, (3–5) whereas targeted disruption of C-terminal SRC kinase (CSK),(6) the kinase that phosphorylates Tyr527, caused activation of c-Src.(7) Studies of the three-dimensional structure of c-Src revealed that interaction between SH2 and phosphorylated Tyr527 together with SH3 and linker region folded the molecule in a closed, inactive state.(8,9) In addition to the Tyr527-based regulation, increasing evidence suggests that the c-Src kinase has a redox-linked regulatory mechanism.(10) One of the models of the redox-linked regulation, we found, is the activation of c-Src by mercuric chloride (HgCl2).(11) Treatment of purified c-Src with mercuric chloride changed Vmax and Km of the kinase, suggesting the kinase activation to be a direct one. (11) Mercuric chloridetreatment of cells activates signal pathways leading to tyrosine phosphorylation of cellular proteins,(12) and activation of mitogenactivated protein (MAP) kinases,(13–16) which in turn induce the expression of cytokines,(17) unregulated cell growth and DNA synthesis.(13,18) Consistently, treatment of c-Src with mercuric chloride activates the kinase.(11) However, the molecular mechanism by which mercuric chloride activates c-Src remains largely unclear. Mercuric ions are widely known as a protein-modifying agent specific for the sulfhydryl group and have an exceptionally high doi: 10.1111/j.1349-7006.2007.00714.x © 2008 Japanese Cancer Association
association constant with sulfhydryl groups in proteins.(19,20) Indeed, we found that N-acetylcysteine could interfere with the activation of c-Src by mercuric chloride both in vitro and in vivo.(11) As a step to understand the molecular mechanism of HgCl2-dependent activation of c-Src, we explored the effect of HgCl2 on a series of mutant c-Src. In this report, we show evidence that the activation of c-Src by HgCl2 requires neither SH2/SH3 domains nor phosphorylation on Tyr416 and Tyr527, but does require some of cysteine residues of c-Src. Among nine cysteine residues scattered over c-Src, Cys483 and Cys498 located in the cysteine-cluster (CC) motif, but not others, are critical for the HgCl2-dependent kinase activation. Other Src family kinases such as Lyn and c-Yes also have similar CC motifs and require cysteine residues at similar positions to be activated by HgCl2. Furthermore, we show that the activation of c-Src by HgCl2 in vivo also requires Cys498. Materials and Methods Cell culture and transfection. COS-7 cells and SYF cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS). COS-7 cells were transfected with diethylaminoethyl (DEAE) dextran. Briefly, plasmid DNA was mixed with transfection reagent (DMEM with 10% NuSerum, 0.5 mg/mL of DEAE dextran, 0.1 mM of chloroquine) and added to cells. Three hours after transfection, 10% dimethyl sulfoxide (DMSO) in phosphate-buffered saline (PBS) was added and incubated for 2 min, and then replaced with fresh media. Two days later, expression of Src was confirmed by immunoblotting. SYF cells are cell lines derived from mouse embryonic fibroblast in which c-src, c-yes and lyn genes are target-disrupted. For transfection to SYF cells, Lipofectamine 2000 (Invitrogen) was used according to the manufacturer’s protocol. Construction of mutant c-Src, c-Yes and Lyn kinases. Full length of c-src was amplified by polymerase chain reaction (PCR) and ligated into pcDNA3 (Invitrogen). lyn and yes were kindly provided by Dr T Yamamoto (University of Tokyo). Substitution of amino acids was performed as previously described.(21) Point mutation of each mutant gene was confirmed by sequencing. ΔSH2SH3 src was constructed by fusing PCR-amplified fragments of membrane-binding domain and kinase domain of c-src. In vitro kinase assay. In vitro kinase assay was performed as described previously.(22) In brief, cells were lyzed in RIPA buffer (20 mM Tris-HCl pH 7.4, 150 mM NaCl, 1% NP-40, 1% deoxycholic acid, 0.1% sodium dodecyl sulfate) and immunoprecipitated with anti-Src antibody. Immunoprecipitates were washed, suspended with kinase buffer (10 mM Tris-HCl pH 7.4, 5 mM MgCl2) containing [γ32 p] ATP (370 kBq) (NEN, Wilmington, DE, USA) and 1.5 μg of acid-treated enolase as an exogenous substrate. The kinase reaction was performed in the
1
To whom correspondence should be addressed. E-mail: [email protected]
Cancer Sci |
March 2008 | vol. 99 | no. 3 | 571–575
presence or absence of HgCl2 for 10 min at 30°C and subjected to electrophoresis. Relative activities were assayed by BAS 2000 system. Iodoacetoamide (IAA) was added to the immunoprecipitates and incubated for 10 min on ice. After washing three times with kinase buffer to eliminate these chemicals, kinase reaction was performed. To examine the kinase activity after HgCl2 stimulation in vivo, SYF cells were transfected with plasmids that express either wild type c-Src or C498A Src. Twenty-four hours later, cells were stimulated with HgCl2 for 2 min, and Src was immunoprecipitated and subjected to in vitro kinase assay; 1 mM of L-cysteine was added during immunoprecipitation to inhibit the further modification of Src by HgCl2. Anti-Src monoclonal antibody 327 was used for immunoprecipitation of c-Src. Anti-Lyn and anti-Yes was purchased from BD Biosciences. Results Activation of c-Src by HgCl2 is sulfhydryl-dependent but does not require SH2, SH3, Tyr416 and Tyr527. We first confirmed the effect
of HgCl2 treatment on the kinase activity of c-Src in vitro. c-Src was immunoprecipitated from either COS-7 cells transfected with c-src expressing plasmid or Sf9 cells infected with recombinant baculovirus containing c-src gene. The in vitro kinase assay was performed with enolase as an exogenous substrate. As shown in Fig. 1(a), addition of HgCl2 to the kinase reaction mixtures enhanced the catalytic activity of c-Src from both sources as we previously reported.(11) In the following experiments, we used c-Src immunoprecipitated from COS-7 cells for the in vitro kinase assay. Mercuric ions are known to bind to free sulfhydryl groups of proteins with a high association constant.(19,20) We next investigated
Fig. 1. Activation of the c-Src kinase in vitro by mercuric chloride (HgCl2) is sulfhydryl-dependent. (a) Immunoprecipitated c-Src, or c-Src purified from Sf9 cells were subjected to in vitro kinase assay in the presence or absence of HgCl2. (b) Inhibition of HgCl2-dependent activation of c-Src by l-cysteine. (c) Inhibition of HgCl2-dependent activation of c-Src by iodoacetoamide. (d) Activation of mutant Src kinases by HgCl2. Y416F has substitution of Tyr416 to phenylalanine and Y527F has Tyr527 to alanine. ΔSH2/SH3 is deleted of SH2 and SH3 domains of cSrc. (e) Effects of heavy metals on the kinase activity of c-Src.
572
the effect of l-cysteine and IAA) on the activation of c-Src by HgCl2. As shown in Fig. 1(b), addition of l-cysteine to the reaction mixtures blocked the HgCl2-dependent activation of c-Src. Some of the SH-alkylating agents such as IAA and N-ethylmaleimide (NEM) bind to sulfhydryl groups of cysteine residues of v-Src but do not inhibit kinase activity.(23) We found that pretreatment of c-Src with IAA did not inactivate the kinase but blocked the activation by HgCl2 (Fig. 1c). A similar inhibitory effect on the HgCl2-dependent activation was also observed with NEM (data not shown). Autophosphorylation of tyrosine 416 of c-Src (Tyr416) and dephosphorylation of tyrosine 527 (Tyr527) appears to be crucial for the activation of c-Src.(3) To investigate the involvement of phosphorylation of these tyrosine residues in the HgCl2-dependent activation of c-Src, we constructed mutant Src proteins, Y416F Src and Y527F Src, whose Tyr416 and Tyr527 were replaced with phenylalanines by site-directed mutagenesis, respectively. Although Y527F Src had higher kinase activity as compared to that of wild-type c-Src, addition of HgCl2 to reaction mixture further activated the catalytic activity of Y416F Src and Y527F Src (Fig. 1d). To confirm and extend these observations, another mutant that lacks SH2 and SH3 domains (ΔSH2/SH3) of c-Src was constructed. ΔSH2/SH3 Src has a deletion of residues 86–246 of c-Src. As shown in Fig. 1(d), ΔSH2/SH3 Src was activated by HgCl2 to a level similar to that of wild-type c-Src. These results imply that Tyr416, Tyr527 and the SH2/SH3 domains are dispensable for the activation of c-Src by HgCl2. We also examined effects of other heavy metals such as lead, copper and iron on the kinase activity of c-Src. As shown in Fig. 1(e), these metals did not affect the catalytic activity of c-Src at the concentration where HgCl2 clearly activated the kinase. Identification of cysteine residues critical for the activation of c-Src by HgCl2. Since HgCl2 is a sulfhydryl group modifier and IAA
could block the activation of c-Src, we next explored the role of each cysteine residue in HgCl2-dependent activation by constructing a series of mutant c-Src in which each cysteine residue in the kinase domain of c-Src was replaced with alanine by site-directed mutagenesis. Of nine cysteine residues in c-Src, Cys185, Cys238 and Cys245 are in the SH2 domain and the other six cysteine residues are in the kinase domain. Since the SH2 domain was dispensable for HgCl2-dependent activation, cysteine residues in the kinase domain were replaced by alanines. C277A, C400A, C483A, C487A, C496A and C498A Src proteins have substitutions of Cys277, Cys400, Cys483, Csy487, Cys496 and Cys498 to alanine, respectively. Among these six cysteines in the kinase domains, Cys483, Cys487, Cys496 and Cys498, are located near the C-terminus and form a cluster,(24) so that we named this motif as the CC (cysteine-cluster) motif (Fig. 2). Other members of the Src-family kinase have similar cysteine clusters, although Src has the largest number of cysteine residues in this motif. It should be noted that this CC motif was critical for cell transformation by v-Src.(21) Based on the number of cysteine residues in the CC motif, the Src-family kinases can be classified into three groups (Fig. 2). Group 1, including Src, Yes and Fyn, has both cysteine residues corresponding to Cys 483 and Cys498 of c-Src. Group 2, including Lck, Lyn and Hck, has the cysteine residue corresponding to Cys 498 of c-Src, but lacks the one corresponding to Cys483. Group 3 (Fgr) lacks both the residues corresponding to Cys 483 and Cys498 of c-Src. The mutant src genes were ligated into pcDNA3 vector and transiently expressed in COS-7 cells. Expression of each mutant was confirmed by immunoblotting and catalytic activity was examined by in vitro kinase assay (Fig. 3). Except for C498A, these mutant Src proteins showed similar levels of kinase activity. Expression of C498A Src was similar to that of others, but its kinase activity was reduced to half of wild-type c-Src (Fig. 3). We examined the effect of HgCl2 treatment on the kinase activity of these mutant Src proteins. As shown in doi: 10.1111/j.1349-7006.2007.00714.x © 2008 Japanese Cancer Association
Fig. 2. Structure of the cysteine-cluster motif of Src family kinases.
Fig. 4. Activation of mutant c-Src by mercuric chloride (HgCl2) (a) Activation of the wild-type and mutant Src kinases by HgCl2 was assayed by in vitro kinase assay. (b) Kinase activities of C498A Src and C498A/Y527F Src. (c) Kinase activation of ΔSH2/SH3C277/400/487/496 A Src, ΔSH2/SH3C483 A Src and ΔSH2/SH3C498 A Src by HgCl2.
and Cys498 to alanines, respectively. As shown in Fig. 4(c), ΔSH2/SH3C277/400/487/496A Src, which has only two cysteine residues at 483 and 498, was activated by HgCl2. In contrast, ΔSH2/SH3C483A Src and rSH2/SH3C498ASrc were resistant to the activation by HgCl2. Fig. 3. Kinase activity of mutant c-Src. Expression of the wild-type and mutant c-Src kinases in COS7 were examined by immunoblotting with anti-Src (lower panel). Src proteins were immunoprecipitated and their kinase activities were assayed (upper panel).
Fig. 4(a), C277A, C400A, C487A and C496A Src were activated by HgCl2 to the levels similar to that of wild-type c-Src. In contrast, C483A and C498A Src were resistant to HgCl 2dependent activation. Although C483A Src was activated slightly by HgCl2, its activation was less than two-fold. On the other hand, C498A Src was completely resistant to the activation by HgCl2. Since C498A Src showed lower catalytic activity compared to any other mutant Src (Fig. 3), a possibility remained that the resistance of this protein to HgCl2 was non-specific. To exclude the possibility, we constructed another mutant Src (C498A/ Y527F Src) in which Tyr527 was replaced by phenylalanine in addition to the substitution of Cys 498 to alanine. As shown in Fig. 4(b), the kinase activity of C498A/Y527F Src was around 10 times higher than that of C498A Src. These results suggest that, although C498A had relatively low kinase activity, it could still be activated by the mutation of Tyr527 and that the resistance of C498A to HgCl2 is specific. These results also suggest that Tyr527-dependent regulation and HgCl2-dependent regulation of c-Src are mutually independent. To confirm that cysteines at 483 and 498 are not only critical but also sufficient for the activation of c-Src by HgCl2, we constructed another series of mutant Src. ΔSH2/SH3C277/400/487/ 496A Src has substitution of Cys277, Cys400, Cys487 and Cys496 to alanine together with deletion of SH2 and SH3 domains. ΔSH2/SH3C483A Src and rSH2/SH3C498A Src are ΔSH2/SH3 Src-derived mutants which have substitution of Cys483 Senga et al.
Requirement of cysteine residues in Yes and Lyn for HgCl2-dependent activation. As the cysteine residues critical for the HgCl2-
mediated activation were partly conserved among the Src family tyrosine kinases (Fig. 2), we investigated whether other members of the Src family tyrosine kinases could also be activated by HgCl2. Yes, which we classified as Group 1 together with c-Src, has three cysteines in the CC motif of the kinase domain. Cysteines at 491, 495 and 506 of Yes correspond to cysteines at 483, 487 and 498 of c-Src, respectively. As shown in Fig. 5(a), Yes was strongly activated by HgCl2. Both C491A Yes and C506A Yes that have substitution of either Cys491 or Cys506 to alanine are resistant to the activation by HgCl2. In contrast to Group 1 kinases, Lyn, which belongs to Group 2, has only two cysteines in the CC motif at 468 and 479 that correspond to cysteines at 487 and 498 of c-Src. Although Lyn lacks cysteine corresponding to Cys483 of c-Src, Lyn is activated by HgCl2. To investigate whether Lyn requires Cys479 corresponding to Cys498 of c-Src to be activated by HgCl2, we constructed Lyn mutant, C479A Lyn that has alanine in place of Cys479. As shown in Fig. 5(b), C479A Lyn became relatively resistant to HgCl2 treatment, although it showed weak activation. These results suggest that not only c-Src but also other members of the Src-family kinase can be activated by HgCl2, at least in part, in a manner dependent on the cysteine residues in the CC motif. Role of the CC motif in in vivo activation of c-Src by HgCl2. To examine whether the CC motif is indeed involved in HgCl2dependent activation of c-Src in vivo, we transiently expressed c-Src and C498A Src in SYF cells in which the c-src, c-yes and c-fyn genes were target-disrupted. It has been known that treatment of cells with HgCl2 dramatically activates tyrosine phosphorylation of cellular proteins. Addition of HgCl2 to SYF cells did not increase tyrosine phosphorylation of cellular proteins. However, we observed dramatic increase of tyrosine phosphorylation of proteins when SYF cells transfected with c-src Cancer Sci | March 2008 | vol. 99 | no. 3 | 573 © 2008 Japanese Cancer Association
expressing plasmid were treated with HgCl2 (Fig. 6a). Interestingly, expression of C498A Src in SYF cells did not show any increase of tyrosine phosphorylation of proteins when treated with HgCl2 (Fig. 6a). We checked catalytic activity of c-Src and C498A Src after HgCl2 treatment. Src proteins were immunoprecipitated in the presence of 1 mM l-cysteine to avoid the in vitro activation of kinase during immunoprecipitation. Kinase activities of the immunoprecipitates were assayed in vitro with enolase as an exogenous substrate. As shown in Fig. 6(b), kinase activity of wild-type c-Src was activated by HgCl2 approximately 10-fold compared to the untreated c-Src. In contrast, C498A Src was resistant to the activation by HgCl2. Discussion
Fig. 5. Activation of the Yes and Lyn kinases by mercuric chloride (HgCl2) (a) Activation of Yes and its mutants by HgCl2. C491A Yes has substitution of Cys491 to alanine and 506 A Yes has Cys506 to alanine. (b) Activation of Lyn and C479A Lyn by HgCl2. C479A Lyn has substitution of Cys 479 to alanine.
Fig. 6. Activation of the Src kinase and tyrosine phosphorylation of proteins in cells transfected with wild-type and mutant forms of c-src by mercuric chloride (HgCl2) (a) SYF cells, in which endogenous c-src, cyes, and fyn were disrupted, were transfected with wild-type c-src or C498A src. The day after transfection, cells were stimulated at the indicated concentration of HgCl2 for 2 min and cells were lyzed and blotted with antiphosphotyrosine and anti-Src antibodies. (b). SYF cells transfected with the indicated plasmids were stimulated with HgCl2 and subjected to in vitro kinase assay.
574
Heavy metals are one of the most toxic forms of environmental pollutants causing life-threatening problems on a worldwide scale. They have accumulated to hazardous levels in the air, land and water from the sludge produced by industries and population centers.(25) Among multiple heavy metal pollutants, mercuric chloride (HgCl2) is unique for its multifarious biological effects, including acute renal failure,(26) autoimmune glomerulonephritis,(27) systemic vasculitis,(28) and autoimmune diseases similar to human systemic lupus erythematosus.(29) Mercuric chloride induces a dramatic activation of the immune system and autoantibody production in susceptible animals. (30) It can activate signal pathways leading to tyrosine phosphorylation of cellular proteins,(12) and activation of MAP kinases,(13–16) which in turn induce the expression of cytokines,(17) unregulated cell growth and DNA synthesis.(13,18) In this paper, we report a novel regulatory mechanism of cSrc by HgCl2 that requires specific cysteine residues in the CC motif. Neither SH2/SH3 nor Tyr527 were required for the activation of c-Src by HgCl2, whereas kinase domain alone was sufficient for the activation. Mutational analysis found that Cys483 and Cys498 are critical for the activation by HgCl2. Point mutation of tyrosine at 527 of C498A Src that was resistant to the activation by HgCl 2 to phenylalanine clearly activated the kinase. These results suggest that the CC motifbased regulation of c-Src is independent of the Tyr527-based regulation of c-Src. Based on the presence of the cysteine residues corresponding to Cys483 and Cys498 of c-Src, the Src family kinases could be classified into three groups (Fig. 2). Group 1, including Src and Yes, has the cysteine residues corresponding to Cys483 and Cys498 of Src. As expected, the Yes kinase responded well to the activation by HgCl 2 and required both of the cysteine residues, suggesting that the CC motif-based activation of kinase by HgCl2 to be a common characteristic among the Group 1 kinases. In contrast to Group 1 kinases, Lyn, a Group 2 kinase, responded to the activation by HgCl2, despite the absence of cysteine corresponding to Cys483 of Src (Fig. 5b). The mechanism of kinase activation by HgCl2 might differ in part between Group 1 and Group 2 kinases, and some other cysteine residues might substitute the role of Cys483 of Src in cases of Group 2 kinase. Replacement of Cys479 of Lyn to alanine, however, converted the kinase to be resistant to HgCl2, again suggesting the importance of the CC motif for the HgCl 2-dependent regulation of Group 2 kinases. We have previously reported that mutation of cysteine residues in the CC motif rendered v-Src refractory to the inhibition by Herbimycin A, indicating that modification of cysteine residues could also inactivate the catalytic activity. Other SH-alkylating agents, N-[p-(2-benzimidazolyl)phenyl] maleimide (BIPM) and N-(9-acridinyl) maleimide (NAM), could also inactivate v-Src dependent on the modification of cysteine residues in the CC motif.(31) These results suggest that, depending on the type of modification, modification of cysteine doi: 10.1111/j.1349-7006.2007.00714.x © 2008 Japanese Cancer Association
residues in the CC motif could either activate or inactivate catalytic activity of Src. A previous report showed that monovalent mercuric compound such as p-chloromercuribenzoate (PCMB) did not change any kinase activity of c-Src as bivalent mercuric ions (Hg2+) did.(11) Positively charged bivalent mercuric ions bound to cysteine may affect adjacent charged amino acids, which could distort the structure of c-Src and enhance the catalytic activity. Otherwise, Hg2+ might bind two adjacent sulfhydryl groups and bridge two cysteines to increase the kinase activity of c-Src. Although further studies are required to elucidate the mechanism of kinase activation by mercuric ions, our results strongly suggest the unique function of the CC motif as a bifunctional regulatory unit. A large body of evidence suggests that HgCl 2 is a potent activator of the intracelluar signaling pathways leading to tyrosine
References 1 Martin GS. The hunting of the Src. Nat Rev Mol Cell Biol 2001; 2: 467–75. 2 Alvarez RH, Kantarjian HM, Cortes JE. The role of Src in solid and hematologic malignancies: development of new-generation Src inhibitors. Cancer 2006; 107: 1918–29. 3 Cartwright CA, Eckhart W, Simon S, Kaplan PL. Cell transformation by pp60c-src mutated in the carboxy-terminal regulatory domain. Cell 1987; 49: 83–91. 4 Kmiecik TE, Shalloway D. Activation and suppression of pp60c-src transforming ability by mutation of its primary sites of tyrosine phosphorylation. Cell 1987; 49: 65–73. 5 Piwnica-Worms H, Saunders KB, Roberts TM, Smith AE, Cheng SH. Tyrosine phosphorylation regulates the biochemical and biological properties of pp60c-src. Cell 1987; 49: 75–82. 6 Nada S, Okada M, MacAuley A, Cooper JA, Nakagawa H. Cloning of a complementary DNA for a protein-tyrosine kinase that specifically phosphorylates a negative regulatory site of p60c-src. Nature 1987; 351: 69– 72. 7 Imamoto A, Soriano P. Disruption of the csk gene, encoding a negative regulator of Src family tyrosine kinases, leads to neural tube defects and embryonic lethality in mice. Cell 1991; 73: 1117–24. 8 Xu W, Harrison SC, Eck MJ. Three-dimensional structure of the tyrosine kinase c-Src. Nature 1993; 385: 595–602. 9 Sicheri F, Moarefi I, Kuriyan J. Crystal structure of the Src family tyrosine kinase Hck. Nature 1997; 385: 602–9. 10 Giannoni E, Buricchi F, Raugei G, Ramponi G, Chiarugi P. Intracellular reactive oxygen species activate Src tyrosine kinase during cell adhesion and anchorage-dependent cell growth. Mol Cell Biol 1997; 25: 6391–403. 11 Pu M et al. Evidence of a novel redox-linked activation mechanism for the Src kinase which is independent of tyrosine 527-mediated regulation. Oncogene 2005; 13: 2615–22. 12 Rahman SM et al. Redox-linked ligand-independent cell surface triggering for extensive protein tyrosine phosphorylation. FEBS Lett 1996; 317: 35– 8. 13 Akhand AA, Kato M, Suzuki H, Miyata T, Nakashima I. Level of HgCl2mediated phosphorylation of intracellular proteins determines death of thymic T-lymphocytes with or without DNA fragmentation. J Cell Biochem 1993; 71: 243–53. 14 Yano T et al. The repetitive activation of extracellular signal-regulated kinase is required for renal regeneration in rat. Life Sci 1998; 62: 2341–7. 15 Turney KD, Parrish AR, Orozco J, Gandolfi AJ. Selective activation in the MAPK pathway by Hg (II) in precision-cut rabbit renal cortical slices. Toxicol Appl Pharmacol 1998; 160: 262–70.
Senga et al.
phosphorylation of cellular proteins which in turn induce unregulated cell growth, DNA synthesis and apoptosis. We found that wild-type c-Src was activated by treatment of cells with HgCl2 within 2 min, whereas C498A Src was resistant to the activation (Fig. 6b). These results suggest that the CC motif of c-Src is involved in the pathological disorders induced by HgCl 2. Our results also suggest that phenotype of human tumors with overexpressed c-Src might be modified under mercuric pollution. In summary, our results disclose that Cys483 and Cys498 in the CC motif of c-Src act as sensors for the activation of the kinase by a redox-linked mechanism. Further studies, including the structural analysis of the CC motif under stimulation by HgCl2, are required for the complete comprehension of the unique redox-linked regulatory mechanism of Src.
16 Matsuoka M, Wispriyono B, Iryo Y, Igisu H. Mercury chloride activates cJun N-terminal kinase and induces c-jun expression in LLC-PK1 cells. Toxicol Sci 1999; 53: 361–8. 17 Fournie GJ et al. Induction of autoimmunity through bystander effects. Lessons from immunological disorders induced by heavy metals. J Autoimmun 2000; 16: 319–26. 18 Akhand AA et al. Magnitude of protein tyrosine phosphorylation-linked signals determines growth versus death of thymic T lymphocytes. Eur J Immunol 2001; 27: 1254–9. 19 Stricks W, Kolthoff IM, Bush DG, Kuroda PK. Experimental study of the polarographic cancer test and of the sulfhydryl titration for the differentiation between normal, cancerous and other pathological sera. J Lancet 1997; 73: 328–35. 20 Sperling R, Burstein Y, Steinberg IZ. Selective reduction and mercuration of cystine IV–V in bovine pancreatic ribonuclease. Biochemistry 1953; 8: 3810–20. 21 Senga T et al. Clustered cysteine residues in the kinase domain of v-Src: critical role for protein stability, cell transformation and sensitivity to herbimycin A. Oncogene 1969; 19: 273–9. 22 Hamaguchi M et al. p60v-src causes tyrosine phosphorylation and inactivation of the N-cadherin-catenin cell adhesion system. Embo J 2000; 12: 307–14. 23 Uehara Y, Fukazawa H, Murakami Y, Mizuno S. Irreversible inhibition of vsrc tyrosine kinase activity by herbimycin A and its abrogation by sulfhydryl compounds. Biochem Biophys Res Commun 1993; 163: 803–9. 24 Veillette A, Dumont S, Fournel M. Conserved cysteine residues are critical for the enzymatic function of the lymphocyte-specific tyrosine protein kinase p56lck. J Biol Chem 1989; 268: 17547–53. 25 Nriagu JO, Pacyna JM. Quantitative assessment of worldwide contamination of air, water and soils by trace metals. Nature 1993; 333: 134–9. 26 Pfaller W, Gstraunthaler G, Willinger CC. Morphology of renal tubular damage from nephrotoxins. Toxicol Lett 1988; 53: 39–43. 27 Kanfer A. Role of coagulation in glomerular injury. Toxicol Lett 1990; 46: 83–92. 28 Luzina IG, Handwerger BS. Lessons from animal models of vasculitis. Curr Rheumatol Rep 1989; 2: 369–75. 29 White KL, David DW, Butterworth LF, Klykken PC. Assessment of autoimmunity-inducing potential using the brown Norway rat challenge model. Toxicol Lett 2000; 112–113: 443–51. 30 Bagenstose LM, Salgame P, Monestier M. Murine mercury-induced autoimmunity: a model of chemically related autoimmunity in humans. Immunol Res 2000; 20: 67–78. 31 Oo ML et al. Cysteine residues in the C-terminal lobe of Src: their role in the suppression of the Src kinase. Oncogene 1999; 22: 1411–17.
Cancer Sci | March 2008 | vol. 99 | no. 3 | 575 © 2008 Japanese Cancer Association