Hsp90 phosphorylation,Wee1 and the cell cycle

Heat Shock Protein 90 (Hsp90) is an essential molecular chaperone in eukaryotic cells, and it maintains the functional conformation of a subset of proteins that are typically key components of multiple regulatory and signaling networks mediating cancer cell proliferation, survival, and metastasis. It is possible to selectively inhibit Hsp90 using natural products such as geldanamycin (GA) or radicicol (RD), which have served as prototypes for development of synthetic Hsp90 inhibitors. These compounds bind within the ADP/ATP-binding site of the Hsp90 N-terminal domain to inhibit its ATPase activity. As numerous N-terminal domain inhibitors are currently undergoing extensive clinical evaluation, it is important to understand the factors that may modulate in vivo susceptibility to these drugs. We recently reported that Wee1Swe1-mediated, cell cycle-dependent, tyrosine phosphorylation of Hsp90 affects GA binding and impacts cancer cell sensitivity to Hsp90 inhibition. This phosphorylation also affects Hsp90 ATPase activity and its ability to chaperone a selected group of clients, comprised primarily of protein kinases. Wee1 regulates the G2/M transition. Here we present additional data demonstrating that tyrosine phosphorylation of Hsp90 by Wee1Swe1 is important for Wee1Swe1 association with Hsp90 and for Wee1Swe1 stability. Yeast expressing non-phosphorylatable yHsp90-Y24F, like swe1? yeast, undergo premature nuclear division that is insensitive to G2/M checkpoint arrest. These findings demonstrate the importance of Hsp90 phosphorylation for proper cell cycle regulation.     

Structure and inhibition of the human cell cycle checkpoint kinase, Wee1A kinase: an atypical tyrosine kinase with a key role in CDK1 regulation

Phosphorylation is critical to regulation of the eukaryotic cell cycle. Entry to mitosis is triggered by the cyclin-dependent kinase CDK1 (Cdc2), which is inactivated during the preceding S and G2 phases by phosphorylation of T14 and Y15. Two homologous kinases, Wee1, which phosphorylates Y15, and Myt1, which phosphorylates both T14 and Y15, mediate this inactivation. We have determined the crystal structure of the catalytic domain of human somatic Wee1 (Wee1A) complexed with an active-site inhibitor at 1.8 A resolution. Although Wee1A is functionally a tyrosine kinase, in sequence and structure it most closely resembles serine/threonine kinases such as Chk1 and cAMP kinases. The crystal structure shows that although the catalytic site closely resembles that of other protein kinases, the activation segment contains Wee1-specific features that maintain it in an active conformation and, together with a key substitution in its glycine-rich loop, help determine its substrate specificity.     

Hsp90 and a tyrosine embedded in the Hsp90 recognition loop are required for the Fer tyrosine kinase activity

Hsp90 is a key regulator of tyrosine kinases activity and is therefore considered as a promising target for intervention with deregulated signaling pathways in malignant cells. Here we describe a novel Hsp90 client — the intracellular tyrosine kinase, Fer, which is subjected to a unique regulatory regime by this chaperone. Inhibition of Hsp90 activity led to proteasomal degradation of the Fer enzyme. However, circumventing the dependence of Fer accumulation on Hsp90, revealed the dependence of the Fer kinase activity and its ability to phosphorylate Stat3 on the chaperon, expressing the necessity of Hsp90 for its function. Mutation analysis unveiled a tyrosine (Tyr⁶¹⁶) embedded in the Hsp90 recognition loop, which is required for the kinase activity of Fer. Replacement of this tyrosine by phenylalanine (Y616F) disabled the auto-phosphorylation activity of Fer and abolished its ability to phosphorylate Stat3. Notably, surrounding the replaced Y616F with subtle mutations restored the auto and trans-phosphorylation activities of Fer suggesting that Y⁶¹⁶ is not itself an essential auto-phosphorylation site of the kinase. Taken together, our results portray Hsp90 and its recognition loop as novel positive regulators of the Fer tyrosine kinase stability and activity.     

G protein-coupled receptor kinase interaction with Hsp90 mediates kinase maturation

G protein-coupled receptor kinase 2 (GRK2) is a serine/threonine-specific protein kinase that mediates agonist-dependent phosphorylation of numerous G protein-coupled receptors. In an effort to identify proteins that regulate GRK2 function, we searched for interacting proteins by immunoprecipitation of endogenous GRK2 from HL60 cells. Subsequent analysis by gel electrophoresis and mass spectrometry revealed that GRK2 associates with heat shock protein 90 (Hsp90). GRK2 interaction with Hsp90 was confirmed by co-immunoprecipitation and was effectively disrupted by geldanamycin, an Hsp90-specific inhibitor. Interestingly, geldanamycin treatment of HL60 cells decreased the expression of endogenous GRK2 in a dose- and time-dependent manner, and metabolic labeling demonstrated that geldanamycin rapidly accelerated the degradation of newly synthesized GRK2. The use of various protease inhibitors suggested that GRK2 degradation induced by geldanamycin was predominantly through the proteasome pathway. To test whether Hsp90 plays a general role in regulating GRK maturation, additional GRKs were studied by transient expression in COS-1 cells and subsequent treatment with geldanamycin. These studies demonstrate that GRK3, GRK5, and GRK6 are also stabilized by interaction with Hsp90. Taken together, our work revealed that GRK interaction with heat shock proteins plays an important role in regulating GRK maturation.     

Hsp90 chaperone activity requires the full-length protein and interaction among its multiple domains

Hsp90 is an abundant and ubiquitous protein involved in a diverse array of cellular processes. Mechanistically we understand little of the apparently complex interactions of this molecular chaperone. Recently, progress has been made in assigning some of the known functions of hsp90, such as nucleotide binding and peptide binding, to particular domains within the protein. We used fragments of hsp90 and chimeric proteins containing functional domains from hsp90 or its mitochondrial homolog, TRAP1, to study the requirements for this protein in the folding of firefly luciferase as well as in the prevention of citrate synthase aggregation. In agreement with others who have found peptide binding and limited chaperone ability in fragments of hsp90, we see that multiple fragments from hsp90 can prevent the aggregation of thermally denatured citrate synthase, a measure of passive chaperoning activity. However, in contrast to these results, the luciferase folding assay was found to be much more demanding. Here, folding is mediated by hsp70 and hsp40, requires ATP, and thus is a measure of active chaperoning. Hsp90 and the co-chaperone, Hop, enhance this process. This hsp90 activity was only observed using full-length hsp90 indicating that the cooperation of multiple functional domains is essential for active, chaperone-mediated folding.     

Hsp90 chaperone complexes are required for the activity and stability of yeast protein kinases Mik1, Wee1 and Swe1.

The Wee1 protein kinase negatively regulates entry into mitosis by mediating the inhibitory tyrosine phosphorylation of Cdc2-cyclin B kinase. The stability and activity of Wee1 from the fission yeast Schizosaccharomyces pombe is critically dependent on functional Hsp90 chaperones. Here we identify two related tyrosine protein kinases, Mik1 from fission yeast and its Saccharomyces cerevisiae homolog Swe1, as Hsp90 substrates and show that the kinase domain is sufficient to mediate this interaction. Morphological and biochemical defects arising from overexpression of the kinases in fission yeast are suppressed in the conditional Hsp90 mutant swo1-26. A subset of all three kinases is associated with the Hsp90 cochaperones cyclophilin 40 and p23. Under conditions of impaired chaperone function or treatment with the Hsp90 inhibitory drug geldanamycin, intracellular levels of the kinases are reduced and the proteins become rapidly degraded by the proteasome machinery, indicating that Wee1, Mik1 and Swe1 require Hsp90 heterocomplexes for their stability and maintenance of function.     

Cell cycle regulation of a Xenopus Wee1-like kinase.

Using a polymerase chain reaction-based strategy, we have isolated a gene encoding a Wee1-like kinase from Xenopus eggs. The recombinant Xenopus Wee1 protein efficiently phosphorylates Cdc2 exclusively on Tyr-15 in a cyclin-dependent manner. The addition of exogenous Wee1 protein to Xenopus cell cycle extracts results in a dose-dependent delay of mitotic initiation that is accompanied by enhanced tyrosine phosphorylation of Cdc2. The activity of the Wee1 protein is highly regulated during the cell cycle: the interphase, underphosphorylated form of Wee1 (68 kDa) phosphorylates Cdc2 very efficiently, whereas the mitotic, hyperphosphorylated version (75 kDa) is weakly active as a Cdc2-specific tyrosine kinase. The down-modulation of Wee1 at mitosis is directly attributable to phosphorylation, since dephosphorylation with protein phosphatase 2A restores its kinase activity. During interphase, the activity of this Wee1 homolog does not vary in response to the presence of unreplicated DNA. The mitosis-specific phosphorylation of Wee1 is due to at least two distinct kinases: the Cdc2 protein and another activity (kinase X) that may correspond to an MPM-2 epitope kinase. These studies indicate that the down-regulation of Wee1-like kinase activity at mitosis is a multistep process that occurs after other biochemical reactions have signaled the successful completion of S phase.     

Regulation of cell size and Wee1 kinase by elevated levels of the cell cycle regulatory protein kinase Cdr2

Many cell cycle regulatory proteins catalyze cell cycle progression in a concentration-dependent manner. In the fission yeast Schizosaccharomyces pombe, the protein kinase Cdr2 promotes mitotic entry by organizing cortical oligomeric nodes that lead to inhibition of Wee1, which itself inhibits the cyclin-dependent kinase Cdk1. cdr2Δ cells lack nodes and divide at increased size due to overactive Wee1, but it has not been known how increased Cdr2 levels might impact Wee1 and cell size. It also has not been clear if and how Cdr2 might regulate Wee1 in the absence of the related kinase Cdr1/Nim1. Using a tetracycline-inducible expression system, we found that a 6× increase in Cdr2 expression caused hyperphosphorylation of Wee1 and reduction in cell size even in the absence of Cdr1/Nim1. This overexpressed Cdr2 formed clusters that sequestered Wee1 adjacent to the nuclear envelope. Cdr2 mutants that disrupt either kinase activity or clustering ability failed to sequester Wee1 and to reduce cell size. We propose that Cdr2 acts as a dosage-dependent regulator of cell size by sequestering its substrate Wee1 in cytoplasmic clusters, away from Cdk1 in the nucleus. This mechanism has implications for other clustered kinases, which may act similarly by sequestering substrates.     

A role for Ebi in neuronal cell cycle control

Mutations in ebi were isolated as enhancers of an over-proliferation phenotype generated by elevated E2F/DP activity in the DROSOPHILA: eye. ebi alleles also strongly suppress a phenotype caused by the cyclin-dependent kinase inhibitor p21, restoring S phases in the second mitotic wave of the developing eye disk. ebi mutant embryos display ectopic S phases within the peripheral nervous system and central nervous system at a time in development when neuronal precursor cells would normally begin to differentiate. Consistent with this, we find that ebi mutants have a reduced capacity to undergo neuronal differentiation, that Ebi physically interacts with Sina and phyllopod, and that Ebi promotes Ttk88 degradation in vitro and in S2 cells. Ectopic expression of Ttk88 inhibited differentiation in embryos and eye discs; however, this block to differentiation was insufficient to promote S phase entry in either of the situations where ebi mutations gave this effect. We conclude that Ebi has two distinct functions; it promotes the degradation of a repressor of neuronal differentiation (Ttk88), and has a second independent function that limits S phase entry.     

UBR1 promotes protein kinase quality control and sensitizes cells to Hsp90 inhibition

UBR1 and UBR2 are N-recognin ubiquitin ligases that function in the N-end rule degradation pathway. In yeast, the UBR1 homologue also functions by N-end rule independent means to promote degradation of misfolded proteins generated by treatment of cells with geldanamycin, a small molecule inhibitor of Hsp90. Based on these studies we examined the role of mammalian UBR1 and UBR2 in the degradation of protein kinase clients upon Hsp90 inhibition. Our findings show that protein kinase clients Akt and Cdk4 are still degraded in mouse Ubr1(-)/(-) cells treated with geldanamycin, but that their levels recover much more rapidly than is found in wild type cells. These findings correlate with increased induction of Hsp90 expression in the Ubr1(-)/(-) cells compared with wild type cells. We also observed a reduction of UBR1 protein levels in geldanamycin-treated mouse embryonic fibroblasts and human breast cancer cells, suggesting that UBR1 is an Hsp90 client. Further studies revealed a functional overlap between UBR1 and the quality control ubiquitin ligase, CHIP. Our findings show that UBR1 function is conserved in controlling the levels of Hsp90-dependent protein kinases upon geldanamycin treatment, and suggest that it plays a role in determining the sensitivity of cancer cells to the chemotherapeutic effects of Hsp90 inhibitors.     

The role of Hsp90 in cell response to hyperthermia

(1) Preincubation of SKOV3 human ovarian carcinoma cells with a non-toxic dose of Geldanamycin resulted in exacerbation of hyperthermia-induced cytotoxicity and re-distribution of dying cells toward necrosis. (2) Exposure of primary human ovarian carcinoma cells to mild hyperthermia (42 °C for 2 h) led, after a recovery period of 16 h, to several-fold increase in the levels of Hsp90 and ErbB2, to a moderate decrease in the levels of phospho-Erk1/2, whereas the level of β-catenin appeared to be unchanged. (3) The inhibitors of Hsp90 (Geldanamycin and Novobiocin) significantly affected the cell signaling of heat-pretreated cultures. (4) The results suggest that Hsp90 plays a pivotal role in cell response to hyperthermia. (5) A combination of Hsp90 inhibitors with hyperthermia may considerably increase the efficacy of thermotherapy.     

Coordination of cell polarization and migration by the Rho family GTPases requires Src tyrosine kinase activity.

BACKGROUND: The ability of a cell to polarize and move is governed by remodeling of the cellular adhesion/cytoskeletal network that is in turn controlled by the Rho family of small GTPases. However, it is not known what signals lie downstream of Rac1 and Cdc42 during peripheral actin and adhesion remodeling that is required for directional migration. RESULTS: We show here that individual members of the Rho family, RhoA, Rac1, and Cdc42, direct the specific intracellular targeting of c-Src tyrosine kinase to focal adhesions, lamellipodia, or filopodia, respectively, and that the adaptor function of c-Src (the combined SH3/SH2 domains coupled to green fluorescent protein) is sufficient for targeting. Furthermore, Src's catalytic activity is absolutely required at these peripheral cell-matrix attachment sites for remodeling that converts RhoA-dependent focal adhesions into smaller focal complexes along Rac1-induced lamellipodia (or Cdc42-induced filopodia). Consequently, cells in which kinase-deficient c-Src occupies peripheral adhesion sites exhibit impaired polarization toward migratory stimuli and reduced motility. Furthermore, phosphorylation of FAK, an Src adhesion substrate, is suppressed under these conditions. CONCLUSIONS: Our findings demonstrate that individual Rho GTPases specify Src's exact peripheral localization and that Rac1- and Cdc42-induced adhesion remodeling and directed cell migration require Src activity at peripheral adhesion sites.     

Farnesylation of Ydj1 is required for in vivo interaction with Hsp90 client proteins

Ydj1 of Saccharomyces cerevisiae is an abundant cytosolic Hsp40, or J-type, molecular chaperone. Ydj1 cooperates with Hsp70 of the Ssa family in the translocation of preproteins to the ER and mitochondria and in the maturation of Hsp90 client proteins. The substrate-binding domain of Ydj1 directly interacts with steroid receptors and is required for the activity of diverse Hsp90-dependent client proteins. However, the effect of Ydj1 alteration on client interaction was unknown. We analyzed the in vivo interaction of Ydj1 with the protein kinase Ste11 and the glucocorticoid receptor. Amino acid alterations in the proposed client-binding domain or zinc-binding domain had minor effects on the physical interaction of Ydj1 with both clients. However, alteration of the carboxy-terminal farnesylation signal disrupted the functional and physical interaction of Ydj1 and Hsp90 with both clients. Similar effects were observed upon deletion of RAM1, which encodes one of the subunits of yeast farnesyltransferase. Our results indicate that farnesylation is a major factor contributing to the specific requirement for Ydj1 in promoting proper regulation and activation of diverse Hsp90 clients.     

Receptor for bombesin with associated tyrosine kinase activity.

The neuropeptide bombesin is known for its potent mitogenic activity on murine 3T3 fibroblasts and other cells. Recently it has been implicated in the pathogenesis of small cell lung carcinoma, in which it acts through an autocrine loop of growth stimulation. Phosphotyrosine (P-Tyr) antibodies have been successfully used to recognize the autophosphorylated receptors for known growth factors. In Swiss 3T3 fibroblasts, phosphotyrosine antibodies identified a 115,000-Mr cell surface protein (p115) that became phosphorylated on tyrosine as a specific response to bombesin stimulation of quiescent cells. The extent of phosphorylation was dose dependent and correlated with the mitogenic effect induced by bombesin, measured by [3H]thymidine incorporation. Tyrosine phosphorylation of p115 was detectable minutes after the addition of bombesin, and its time course paralleled that described for the binding of bombesin to its receptor. Immunocomplexes of phosphorylated p115 and phosphotyrosine antibodies bound 125I-labeled [Tyr4]bombesin in a specific and saturable manner and displayed an associated tyrosine kinase activity enhanced by bombesin. Furthermore, the 125I-labeled bombesin analog gastrin-releasing peptide, bound to intact live cells, was coprecipitated with p115. These data strongly suggest that p115 participates in the structure and function of the surface receptor for bombesin, a new member of the family of growth factor receptors with associated tyrosine kinase activity.