Deferoxamine synergistically enhances iron-mediated AP-1 activation: a showcase of the interplay between extracellular-signal-regulated kinase and tyrosine phosphatase

Deferoxamine (DFO) is a drug widely used for iron overload treatment to reduce body iron burden. In the present study, it was shown in mouse epidermal JB6 cells that all iron compounds transiently induced extracellular signal-regulated kinases (ERK) phosphorylation, whereas DFO further enhanced ERK phosphorylation over long periods. The ERK phosphorylation by DFO treatment appears to be due to the inhibition of MAPK phosphatases (MKP) by DFO. The combined effects of iron-initiated MAPK activation and DFO-mediated MKP inhibition resulted in a synergistic enhancement on AP-1 activities. The results indicate that the interplay between MAPK and MKP is important in regulating the extent of AP-1 activation. It is known that administration of DFO in iron overload patients often results in allergic responses at the injection sites. The results suggest that this synergistic AP-1 activation might play a role in DFO-induced skin immune responses of iron overload patients.     

Negative regulation of Ros receptor tyrosine kinase signaling. An epithelial function of the SH2 domain protein tyrosine phosphatase SHP-1

Male "viable motheaten" (me(v)) mice, with a naturally occurring mutation in the gene of the SH2 domain protein tyrosine phosphatase SHP-1, are sterile. Known defects in sperm maturation in these mice correlate with an impaired differentiation of the epididymis, which has similarities to the phenotype of mice with a targeted inactivation of the Ros receptor tyrosine kinase. Ros and SHP-1 are coexpressed in epididymal epithelium, and elevated phosphorylation of Ros in the epididymis of me(v) mice suggests that Ros signaling is under control of SHP-1 in vivo. Phosphorylated Ros strongly and directly associates with SHP-1 in yeast two-hybrid, glutathione S-transferase pull-down, and coimmunoprecipitation experiments. Strong binding of SHP-1 to Ros is selective compared to six other receptor tyrosine kinases. The interaction is mediated by the SHP-1 NH(2)-terminal SH2 domain and Ros phosphotyrosine 2267. Overexpression of SHP-1 results in Ros dephosphorylation and effectively downregulates Ros-dependent proliferation and transformation. We propose that SHP-1 is an important downstream regulator of Ros signaling.     

Phosphorylation of the M3/6 dual-specificity phosphatase enhances the activation of JNK by arsenite

Specific outcomes upon activation of the c-Jun N-terminal kinase (JNK) pathway critically depend on the intensity and duration of signal transmission. Dual-specificity phosphatases (DUSPs) play a very important role in these events by modulating the extent of JNK phosphorylation and activation and thus regulating cellular responses to stress. M3/6 (DUSP8) is one of the dual-specificity protein phosphatases with distinct specificity towards JNK. It has been shown that M3/6 itself is phosphorylated by JNK upon stimulation with arsenite, but the role of this phosphorylation has not been investigated. In this study, we mapped JNK-induced phosphorylation sites on M3/6 using mass spectrometry. Phosphorylated residues Ser 515, Thr 518 and Ser 520 were identified and site-directed mutagenesis was employed to investigate their role. Upon arsenite stimulation, M3/6 mutated at these sites exhibited decreased phosphorylation compared to the wild-type protein. No difference was observed in terms of the enzyme's in vitro phosphatase activity, its substrate specificity towards JNK isoforms, its interactions with JNK and the scaffold family of JNK-interacting proteins (JIPs), its stability or its subcellular localization. Interestingly, expression of M3/6 phosphorylation mutants delayed the time-course of JNK phosphorylation and activation by arsenite. We propose that phosphorylation of the M3/6 phosphatase by JNK in response to stress stimuli results in attenuation of phosphatase activity and acceleration of JNK activation.     

Determinants for the interaction between Janus kinase 2 and protein phosphatase 2A

We have shown that the serine/threonine protein phosphatase 2A (PP2A) associates with the Jak2 tyrosine kinase in a myeloid progenitor line. In this study, we characterized the regions of Jak2 and PP2A responsible for association and evaluated the functional consequences of association. We demonstrate that PP2A interacts with truncated forms of Jak2 containing the JH1 catalytic domain. Using GST fusion proteins, we show that the isolated JH1 and JH3 domains of Jak2 bind directly to PP2A. Jak2 contains putative PP2A binding sequences (LXXLL) in the JH1 domain (residues 1078-1082) and in the JH3 domain (residues 474-478). Mutation of the LXXLL sequence in the JH1 domain decreased PP2A binding in vitro, while mutation of the similar JH3 sequence did not affect PP2A binding. We analyzed full-length Jak2 bearing the LXXLL mutation in Cos-7 cells for association with PP2A. The JH1 mutation impaired Jak2 activity and had a modest effect on PP2A binding. Finally, we show that a mutant form of the PP2A catalytic subunit lacking a site for phosphorylation (Y307F) binds more tightly to Jak2 than wild-type PP2A, consistent with a model where phosphorylation disrupts the Jak2-PP2A interaction.     

Polo-like kinase 1 (PLK1) and protein phosphatase 6 (PP6) regulate DNA-dependent protein kinase catalytic subunit (DNA-PKcs) phosphorylation in mitosis

The protein kinase activity of the DNA-PKcs (DNA-dependent protein kinase catalytic subunit) and its autophosphorylation are critical for DBS (DNA double-strand break) repair via NHEJ (non-homologous end-joining). Recent studies have shown that depletion or inactivation of DNA-PKcs kinase activity also results in mitotic defects. DNA-PKcs is autophosphorylated on Ser²⁰⁵⁶, Thr²⁶⁴⁷ and Thr²⁶⁰⁹ in mitosis and phosphorylated DNA-PKcs localize to centrosomes, mitotic spindles and the midbody. DNA-PKcs also interacts with PP6 (protein phosphatase 6), and PP6 has been shown to dephosphorylate Aurora A kinase in mitosis. Here we report that DNA-PKcs is phosphorylated on Ser³²⁰⁵ and Thr³⁹⁵⁰ in mitosis. Phosphorylation of Thr³⁹⁵⁰ is DNA-PK-dependent, whereas phosphorylation of Ser³²⁰⁵ requires PLK1 (polo-like kinase 1). Moreover, PLK1 phosphorylates DNA-PKcs on Ser³²⁰⁵ in vitro and interacts with DNA-PKcs in mitosis. In addition, PP6 dephosphorylates DNA-PKcs at Ser³²⁰⁵ in mitosis and after IR (ionizing radiation). DNA-PKcs also phosphorylates Chk2 on Thr⁶⁸ in mitosis and both phosphorylation of Chk2 and autophosphorylation of DNA-PKcs in mitosis occur in the apparent absence of Ku and DNA damage. Our findings provide mechanistic insight into the roles of DNA-PKcs and PP6 in mitosis and suggest that DNA-PKcs’ role in mitosis may be mechanistically distinct from its well-established role in NHEJ.     

Structure of the hematopoietic tyrosine phosphatase (HePTP) catalytic domain: structure of a KIM phosphatase with phosphate bound at the active site

Hematopoietic tyrosine phosphatase (HePTP) is a 38kDa class I non-receptor protein tyrosine phosphatase (PTP) that is strongly expressed in T cells. It is composed of a C-terminal classical PTP domain (residues 44-339) and a short N-terminal extension (residues 1-43) that functions to direct HePTP to its physiological substrates. Moreover, HePTP is a member of a recently identified family of PTPs that has a major role in regulating the activity and translocation of the MAP kinases Erk and p38. HePTP binds Erk and p38 via a short, highly conserved motif in its N terminus, termed the kinase interaction motif (KIM). Association of HePTP with Erk via the KIM results in an unusual, reciprocal interaction between the two proteins. First, Erk phosphorylates HePTP at residues Thr45 and Ser72. Second, HePTP dephosphorylates Erk at PTyr185. In order to gain further insight into the interaction of HePTP with Erk, we determined the structure of the PTP catalytic domain of HePTP, residues 44-339. The HePTP catalytic phosphatase domain displays the classical PTP1B fold and superimposes well with PTP-SL, the first KIM-containing phosphatase solved to high resolution. In contrast to the PTP-SL structure, however, HePTP crystallized with a well-ordered phosphate ion bound at the active site. This resulted in the closure of the catalytically important WPD loop, and thus, HePTP represents the first KIM-containing phosphatase solved in the closed conformation. Finally, using this structure of the HePTP catalytic domain, we show that both the phosphorylation of HePTP at Thr45 and Ser72 by Erk2 and the dephosphorylation of Erk2 at Tyr185 by HePTP require significant conformational changes in both proteins.     

Crystal structure of human protein tyrosine phosphatase SHP-1 in the open conformation

SHP‐1 belongs to the family of non‐receptor protein tyrosine phosphatases (PTPs) and generally acts as a negative regulator in a variety of cellular signaling pathways. Previously, the crystal structures of the tail‐truncated SHP‐1 and SHP‐2 revealed an autoinhibitory conformation. To understand the regulatory mechanism of SHP‐1, we have determined the crystal structure of the full‐length SHP‐1 at 3.1 Å. Although the tail was disordered in current structure, the huge conformational rearrangement of the N‐SH2 domain and the incorporation of sulfate ions into the ligand‐binding site of each domain indicate that the SHP‐1 is in the open conformation. The N‐SH2 domain in current structure is shifted away from the active site of the PTP domain to the other side of the C‐SH2 domain, resulting in exposure of the active site. Meanwhile, the C‐SH2 domain is twisted anticlockwise by about 110°. In addition, a set of new interactions between two SH2 domains and between the N‐SH2 and the catalytic domains is identified, which could be responsible for the stabilization of SHP‐1 in the open conformation. Based on the structural comparison, a model for the activation of SHP‐1 is proposed. J. Cell. Biochem. 112: 2062–2071, 2011. © 2011 Wiley‐Liss, Inc.     

Phosphorylation of protein phosphatase type-1 inhibitory proteins by integrin-linked kinase and cyclic nucleotide-dependent protein kinases

Protein phosphatases play key roles in cellular regulation and are subjected to control by protein inhibitors whose activity is in turn regulated by phosphorylation. Here we investigated the possible regulation of phosphorylation-dependent type-1 protein phosphatase (PP1) inhibitors, CPI-17, PHI-1, and KEPI, by various kinases. Protein kinases A (PKA) and G (PKG) phosphorylated CPI-17 at the inhibitory site (T38), but not PHI-1 (T57). Phosphorylated CPI-17 inhibited the activity of both the PP1 catalytic subunit (PP1c) and the myosin phosphatase holoenzyme (MPH) with IC(50) values of 1-8 nM. PKA predominantly phosphorylated a site distinct from the inhibitory T73 in KEPI, whereas PKG was ineffective. Integrin-linked kinase phosphorylated KEPI (T73) and this dramatically increased inhibition of PP1c (IC(50)=0.1 nM) and MPH (IC(50)=8 nM). These results suggest that the regulatory phosphorylation of CPI-17 and KEPI may involve distinct kinases and signaling pathways.     

Translocation and activation of sphingosine kinase 1 by ceramide-1-phosphate

Sphingosine kinases (SphKs) and ceramide kinase (CerK) phosphorylate sphingosine to sphingosine-1-phosphate (S1P) and ceramide to ceramide-1-phosphate (C1P), respectively. S1P and C1P are bioactive lipids that regulate cell fate/function and human health/diseases. The translocation and activity of SphK1 are regulated by its phosphorylation of Ser ²²⁵ and by anionic lipids such as phosphatidic acid and phosphatidylserine. However, the roles of another anionic lipid C1P on SphK1 functions have not yet been elucidated, thus, we here investigated the regulation of SphK1 by CerK/C1P. C1P concentration dependently bound with and activated recombinant human SphK1. The inhibition of CerK reduced the phorbol 12-myristate 13-acetate-induced translocation of SphK1 to the plasma membrane (PM) and activation of the enzyme in membrane fractions of cells. A treatment with C1P translocated wild-type SphK1, but not the SphK1-S225A mutant, to the PM without affecting phosphorylation signaling. A cationic RxRH sequence is proposed to be a C1P-binding motif in α-type cytosolic phospholipase A ₂ and tumor necrosis factor α-converting enzyme. The mutation of four cationic amino acids to Ala in the 56-RRNHAR-61 domain in SphK1 reduced the phorbol 12-myristate 13-acetate- and C1P-induced translocation of SphK1 to the PM, however, the capacity of C1P to bind with and activate SphK1 was not affected by this mutation. In conclusion, C1P modulates SphK1 functions by interacting with multiple sites in SphK1.     

ATG16L1 phosphorylation is oppositely regulated by CSNK2/casein kinase 2 and PPP1/protein phosphatase 1 which determines the fate of cardiomyocytes during hypoxia/reoxygenation

Recent studies have shown that the phosphorylation and dephosphorylation of ULK1 and ATG13 are related to autophagy activity. Although ATG16L1 is absolutely required for autophagy induction by affecting the formation of autophagosomes, the post-translational modification of ATG16L1 remains elusive. Here, we explored the regulatory mechanism and role of ATG16L1 phosphorylation for autophagy induction in cardiomyocytes. We showed that ATG16L1 was a phosphoprotein, because phosphorylation of ATG16L1 was detected in rat cardiomyocytes during hypoxia/reoxygenation (H/R). We not only demonstrated that CSNK2 (casein kinase 2) phosphorylated ATG16L1, but also identified the highly conserved Ser139 as the critical phosphorylation residue for CSNK2. We further established that ATG16L1 associated with the ATG12-ATG5 complex in a Ser139 phosphorylation-dependent manner. In agreement with this finding, CSNK2 inhibitor disrupted the ATG12-ATG5-ATG16L1 complex. Importantly, phosphorylation of ATG16L1 on Ser139 was responsible for H/R-induced autophagy in cardiomyocytes, which protects cardiomyocytes from apoptosis. Conversely, we determined that wild-type PPP1 (protein phosphatase 1), but not the inactive mutant, associated with ATG16L1 and antagonized CSNK2-mediated phosphorylation of ATG16L1. Interestingly, one RVxF consensus site for PPP1 binding in the C-terminal tail of ATG16L1 was identified; mutation of this site disrupted its association with ATG16L1. Notably, CSNK2 also associated with PPP1, but ATG16L1 depletion impaired the interaction between CSNK2 and PPP1. Collectively, these data identify ATG16L1 as a bona fide physiological CSNK2 and PPP1 substrate, which reveals a novel molecular link from CSNK2 to activation of the autophagy-specific ATG12-ATG5-ATG16L1 complex and autophagy induction.     

Inhibition of protein tyrosine phosphatase 1B and alkaline phosphatase by bis(maltolato)oxovanadium (IV)

Vanadate has been recognized as a specific and potent phosphatase inhibitor since its structure is similar to that of phosphate. In this study, we measured the inhibition of glutathione S-transferase-tagged protein tyrosine phosphatase 1B (GST-PTP1B) and alkaline phosphatase (ALP) by the insulin enhancing compounds, bis(maltolato)oxovanadium(IV) (BMOV). The results showed that the activity of GST-PTP1B was reversibly inhibited by solutions of BMOV with an IC₅₀ value of 0.86 ± 0.02 μM. Steady state kinetic studies showed that inhibition of GST-PTP1B by BMOV was of a mixed competitive and noncompetitive type. In addition, incubation of GST-PTP1B with BMOV showed a time-dependent biphasic inactivation of the protein. On the other hand, the inhibitory behavior of BMOV on ALP activity was reversible and competitive with an IC₅₀ value of 32.1 ± 0.6 μM. Incubation with BMOV did not show biphasic inactivation of ALP. The reversible inhibition of GST-PTP1B by BMOV is more potent than that of ALP, but solutions of BMOV inhibited both enzymes. This data support the suggestion that mechanisms for the inhibitory effects of BMOV on GST-PTP1B and ALP are very different.