Dual-specificity protein kinases: will any hydroxyl do?

Protein kinases are classified by the target amino acid in their substrates. Those protein kinases that phosphorylate hydroxyamino acids comprise two groups, the protein-tyrosine and protein-serine/threonine kinases, which, until recently, had been thought to be mutually exclusive. However, several new protein kinases have been discovered that, by the criterion of primary structure, would be classified as protein-serine/threonine kinases but which, surprisingly, are able to phosphorylate tyrosine residues. Even more surprising, there are reports of protein kinases that are capable of phosphorylating both tyrosine and serine/threonine residues. We review and discuss recent developments concerning these 'dal-specificity' protein kinases.

Cell signalling. Receptor orphans find a family.

A family of ligands has been identified for the largest group of receptor protein-tyrosine kinases--the hitherto 'orphan' EPH receptor subfamily--and the functions of these receptors and ligands are starting to be elucidated.

cDNA cloning and characterization of eck, an epithelial cell receptor protein-tyrosine kinase in the eph/elk family of protein kinases.

A human epithelial (HeLa) cDNA library was screened with degenerate oligonucleotides designed to hybridize to highly conserved regions of protein-tyrosine kinases. One cDNA from this screen was shown to contain a putative protein-tyrosine kinase catalytic domain and subsequently used to isolate another cDNA from a human keratinocyte library that encompasses the entire coding region of a 976-amino-acid polypeptide. The predicted protein has an external domain of 534 amino acids with a presumptive N-terminal signal peptide, a transmembrane domain, and a cytoplasmic domain of 418 amino acids that includes a canonical protein-tyrosine kinase catalytic domain. Molecular phylogeny indicates that this protein kinase is closely related to eph and elk and that this receptor family is more closely related to the non-receptor protein-tyrosine kinase families than to other receptor protein-tyrosine kinases. Antibodies raised against a TrpE fusion protein immunoprecipitated a 130-kDa protein that became phosphorylated on tyrosine in immune complex kinase assays, indicating that this protein is a bona fide protein-tyrosine kinase. Analysis of RNA from 13 adult rat organs showed that the eck gene is expressed most highly in tissues that contain a high proportion of epithelial cells, e.g., skin, intestine, lung, and ovary. Several cell lines of epithelial origin were found to express the eck protein kinase at the protein and RNA levels. Immunohistochemical analysis of several rat organs also showed staining in epithelial cells. These observations prompted us to name this protein kinase eck, for epithelial cell kinase.

Synthesis of p36 and p35 is increased when U-937 cells differentiate in culture but expression is not inducible by glucocorticoids.

p36 and p35 are distinct but related proteins that share many structural and biochemical features which were first identified as major substrates for protein-tyrosine kinases. Subsequently, both proteins have been shown to be Ca2+-, phospholipid-, and F-actin-binding proteins that underlie the plasma membrane and are associated with the cortical cytoskeleton. Recent reports have claimed that these proteins function as lipocortins, i.e., phospholipase A2 inhibitors that mediate the anti-inflammatory action of glucocorticoids. To investigate this possibility and to learn more about the functions of p36 and p35, we used human-specific anti-p36 and anti-p35 monoclonal antibodies to determine whether the expression or secretion of either protein was inducible by dexamethasone in the human U-937 myeloid cell line and in other human cell types. Additionally, we examined the levels of mRNA for both proteins. No effect of dexamethasone was observed on p36 or p35 expression at either the mRNA or protein level, nor were these proteins secreted under any of the culture conditions investigated. However, it was observed that in these cells the rate of synthesis and accumulation of both proteins was increased when the U-937 cells were induced to differentiate in culture to adherent macrophagelike cells. This offers a model system with which to study the control of p36 and p35 expression.

trkB, a neural receptor protein-tyrosine kinase: evidence for a full-length and two truncated receptors.

We have screened an adult rat cerebellar cDNA library in search of novel protein tyrosine-kinase (PTK) cDNAs. A cDNA for a putative PTK, trkB, was cloned, and its sequence indicates that it is likely to be derived from a gene for a ligand-regulated receptor closely related to the human trk oncogene. Northern (RNA) analysis showed that the trkB gene is expressed predominantly in the brain and that trkB expresses multiple mRNAs, ranging from 0.7 to 9 kb. Hybridization of cerebral mRNAs with a variety of probes indicates that there are mRNAs encoding truncated trkB receptors. Two additional types of cDNA were isolated, and their sequences are predicted to encode two distinct C-terminally truncated receptors which have the complete extracellular region and transmembrane domain, but which differ in their short cytoplasmic tails.

Cloning and characterization of a specific receptor for mouse oncostatin M.

Oncostatin M (OSM) is a member of a family of cytokines that includes ciliary neurotrophic factor, interleukin-6, interleukin-11, cardiotrophin-1, and leukemia inhibitory factor (LIF). The receptors for these cytokines consist of a common signaling subunit, gp130, to which other subunits are added to modify ligand specificity. We report here the isolation and characterization of a cDNA encoding a subunit of the mouse OSM receptor. In NIH 3T3 cells (which endogenously express gp130, LIF receptor beta [LIFRbeta], and the protein product, c12, of the cDNA described here), mouse LIF, human LIF, and human OSM signaled through receptors containing the LIFRbeta and gp130 but not through the mouse OSM receptor. Mouse OSM, however, signaled only through a c12-gp130 complex; it did not use the LIF receptor. Binding studies demonstrated that mouse OSM associated directly with either the c12 protein or gp130. These data highlight the species-specific differences in receptor utilization and signal transduction between mouse and human OSM. In mouse cells, only mouse OSM is capable of activating the mouse OSM receptor; human OSM instead activates the LIF receptor. Therefore, these data suggest that all previous studies with human OSM in mouse systems did not elucidate the biology of OSM but, rather, reflected the biological actions of LIF.

Budding and fission yeast casein kinase I isoforms have dual-specificity protein kinase activity.

We have examined the activity and substrate specificity of the Saccharomyces cerevisiae Hrr25p and the Schizosaccharomyces pombe Hhp1, Hhp2, and Cki1 protein kinase isoforms. These four gene products are isotypes of casein kinase I (CKI), and the sequence of these protein kinases predicts that they are protein serine/threonine kinases. However, each of these four protein kinases, when expressed in Escherichia coli in an active form, was recognized by anti-phosphotyrosine antibodies. Phosphoamino acid analysis of 32P-labeled proteins showed phosphorylation on serine, threonine, and tyrosine residues. The E. coli produced forms of Hhp1, Hhp2, and Cki1 were autophosphorylated on tyrosine, and both Hhp1 and Hhp2 were capable of phosphorylating the tyrosine-protein kinase synthetic peptide substrate polymer poly-E4Y1. Immune complex protein kinases assays from S. pombe cells showed that Hhp1-containing precipitates were associated with a protein-tyrosine kinase activity, and the Hhp1 present in these immunoprecipitates was phosphorylated on tyrosine residues. Although dephosphorylation of Hhp1 and Hhp2 by Ser/Thr phosphatase had little effect on the specific activity, tyrosine dephosphorylation of Hhp1 and Hhp2 caused a 1.8-to 3.1-fold increase in the Km for poly-E4Y1 and casein. These data demonstrate that four different CKI isoforms from two different yeasts are capable of protein-tyrosine kinase activity and encode dual-specificity protein kinases.

Protein B61 as a new growth factor: expression of B61 and up-regulation of its receptor epithelial cell kinase during melanoma progression.

Epithelial cell kinase (ECK) is a receptor protein tyrosine kinase, the role of which in melanoma biology is unclear. Here we studied the role of ECK during melanoma progression. ECK mRNA was overexpressed in virtually all melanoma lines tested, and levels were significantly higher in cell lines from distant metastases than primary melanomas; melanocytes were negative. Gene amplification was not detected in melanomas. Levels of ECK protein corresponded well with mRNA levels. B61 or LERK-1, recently identified as an ECK ligand, stimulated the growth of ECK-expressing melanoma cell lines, its first identified biological activity. Melanoma chemotaxis and chemoinvasion were not affected by B61. Growth of normal melanocytes was not affected. mRNA for B61 was detected in both melanoma cell lines and normal melanocytes. B61 was also identified by Western blotting and ECK binding activity with the use of a BIAcore binding assay in melanoma cell-conditioned media. These results suggest that B61 is an autocrine growth factor for melanomas but not normal melanocytes.

Characterization of a human protein threonine kinase isolated by screening an expression library with antibodies to phosphotyrosine.

We have studied the activity and substrate specificity of the catalytic domain of a protein kinase that was isolated in a screen of a human lambda gt11 fibroblast cDNA library with anti-phosphotyrosine antibodies. The sequence of this protein kinase would predict that it is a protein serine/threonine kinase, which at first seemed incongruent with the cloning method. However, recent reports indicate that some protein kinases can phosphorylate both tyrosine and serine/threonine residues. To determine whether this protein kinase, which we call PYT (for phosphotyrosine picked threonine kinase), was a dual-specificity protein kinase we investigated its substrate specificity when expressed in bacteria. The catalytic domain was active as a protein kinase when expressed from any of several promoters and when expressed as a TrpE fusion protein. All experiments that resulted in an active protein kinase, as judged by incorporation of 32P by metabolic labeling, also resulted in the generation of proteins that were recognized by anti-phosphotyrosine antibodies. Phosphoamino acid analyses of the metabolically labeled proteins that were recognized by the antibodies consistently yielded large amounts of phosphothreonine and only trace amounts of phosphotyrosine. We mapped the phosphorylation sites in the phosphorylated PYT protein and found only phosphothreonine; 90% of the radioactivity mapped to a threonine in the region autophosphorylated by many protein kinases. These data demonstrate that PYT is primarily a protein threonine kinase, but that it can phosphorylate tyrosine to a small extent, making it a potential dual-specificity protein kinase.

Identification of cDNA clones that code for protein-tyrosine kinases by screening expression libraries with antibodies against phosphotyrosine.

A phage lambda gt11 human fibroblast cDNA expression library was screened with antibodies against phosphotyrosine to determine the feasibility of this approach as a method for the identification of clones that code for protein-tyrosine kinases. Several antibody positive clones were isolated. One clone also scored positive with degenerate oligonucleotides designed to identify sequences coding for protein-tyrosine kinases. This cDNA clone was partially sequenced and proved to be identical to part of lyn, a recently reported putative protein-tyrosine kinase. A portion of the cDNA was cloned into an inducible plasmid expression vector. Phosphorylation of many bacterial proteins on tyrosine was observed upon addition of inducing agent. The results demonstrate that this method of screening can identify cDNAs that encode active protein-tyrosine kinases and could prove useful for the identification of novel ones.

The ephrin-A1 ligand and its receptor, EphA2, are expressed during tumor neovascularization.

Eph receptor tyrosine kinases and their ephrin ligands have been implicated in embryonic vascular development and in in vivo models of angiogenesis. Eph proteins may also regulate tumor neovascularization, but this role has not been previously investigated. To screen for Eph proteins expressed in tumor blood vessels, we used tumor xenografts grown in nude mice from MDA-MB-435 human breast cancer cells or KS1767 human Kaposi's sarcoma cells. By immunohistochemistry, the ephrin-A1 ligand and one of its receptors, EphA2, were detected throughout tumor vasculature. Double-labeling with anti-CD34 antibodies demonstrated that both ephrin-A1 and EphA2 were expressed in xenograft endothelial cells and also tumor cells. Furthermore, EphA2 was tyrosine-phosphorylated in the xenograft tumors, indicating that it was activated, presumably by interacting with ephrin-A1. Ephrin-A1 and EphA2 were also detected in both the vasculature and tumor cells of surgically removed human cancers. In an in vitro angiogenesis model, a dominant negative form of EphA2 inhibited capillary tube-like formation by human umbilical vein endothelial cells (HUVECs), demonstrating a requirement for EphA receptor signaling. These data suggest that ephrin-A1 and EphA2 play a role in human cancers, at least in part by influencing tumor neovascularization. Eph proteins may represent promising new targets for antiangiogenic cancer treatments.

cDNA cloning and tissue distribution of five human EPH-like receptor protein-tyrosine kinases.

We have isolated cDNA clones from a human fetal brain library that encode five members of the EPH sub-family of receptor protein tyrosine kinases (PTKs). Comparison of the DNA sequences of these receptors to the Genbank database reveals that two of our clones correspond to the previously identified HEK and ERK receptors, two are apparently human homologues of the mouse receptors Sek and Bsk and one is novel. With these additions, the number of known human EPH sub-family members is nine and the total in all vertebrate species is 13 making it the largest known sub-family of PTKs. Analysis of the expression pattern of EPH sub-family mRNAs reveals that some are expressed in a wide variety of adult tissues while others are quite restricted. Consistent with the amplification of these sequences from a fetal brain cDNA library, all five members which we have isolated are expressed in the brain. We have named these receptors HEK4, HEK5, HEK7, HEK8 and HEK11, following the nomenclature of Wicks et al. (1992) and the numbering convention set forth by Sajjadi et al. (1991). Analysis of these new EPH sub-family members will increase our understanding of the biology of this receptor family and their isolation will provide reagents for the identification of ligands for this large family of orphan receptors.

Spontaneously hypertensive rats are resistant to the development of hypercholesterolemia.

The arterioles of young spontaneously hypertensive rats (SHR) are purported to have an enhanced sensitivity to nitric oxide (NO)-dependent vasodilators, relative to normotensive animals, while NO-related arteriolar responses are diminished in both mature SHR as well as hypercholesterolemic normotensive rats. Because endothelial production of NO relaxes vascular smooth muscle and inhibits platelet adhesion and aggregation, hypercholesterolemia may synergistically affect the development of genetic hypertension. The NO-mediated baseline vascular tone, acetylcholine-induced dilation, and inhibition of platelet thrombus formation were studied over time (10 weeks) in SHR and hypercholesterolemic SHR (HC-SHR). The in vivo microcirculation of the cremaster muscle was used to quantitate all observations. The HC-SHR became significantly hypercholesterolemic after 1 week on the cholesterol-supplemented diet, with serum cholesterol concentrations remaining elevated for the 10 weeks studied. However, the serum cholesterol concentrations of HC-SHR were significantly less than those of Sprague-Dawley and Wistar-Kyoto rats fed the same diet. Dietary hypercholesterolemia did not exacerbate the development of genetic hypertension. Second- and third-order arterioles of SHR and age-matched HC-SHR constricted to the same extent when the NO synthase inhibitor N omega-nitro-L-arginine methyl ester (L-NAME) was applied. The third order arterioles of both groups also dilated the same amount to acetylcholine and sodium nitroprusside. Platelet thrombus formation induced by light/dye photochemistry was not different between the SHR and HC-SHR groups either at 1 or 10 weeks of diet, and L-NAME decreased the time to thrombus occlusion of blood flow equally in both groups. This is in marked contrast to the previously reported hypercholesterolemia-induced decreases in vascular reactivity in Sprague-Dawley rats. These current findings demonstrate that SHR are resistant to the development of hypercholesterolemia and that NO-mediated vascular responses in SHR are not attenuated by hypercholesterolemia.

Comparison of coarse mesh gauze with biologic dressings on granulating wounds.

Twenty patients with large area granulating wounds were selected for the study. On each patient, a small area of uniform appearance was divided into four subareas, and each subarea was randomly assigned treatment with a single 24 hour application of cadaver allograft, fresh porcine xenograft, formalinized xenograft, or "wet-to-dry" applications of coarse mesh gauze changed three times daily. At 24 hours, all four areas were uncovered. The subareas were ranked in terms of appearance on a best, second best, third best, and worst scale by experienced paramedical personnel who were not told which area received which treatment. Contact sponge quantitative microbiology was performed on each subarea before and after treatment on seventeen of the twenty patients. The results of these rankings suggest that coarse mesh gauze, changed three times daily, was significantly better at improving wound appearance than any of the biologic dressings. Among the biologic dressings, formalinized xenograft was significantly the worst. There was no significant difference between cadaver allograft and porcine xenograft. Analysis of quantitative cultures was limited by the problems of applying statistical methods to series of paired cultures in which initial values are quite different. Within the framework of such limitations and our 24 hour study, there was not significant change in surface colonization when either coarse mesh gauze or biologic dressings were used.

Mice overexpressing murine oncostatin M (OSM) exhibit changes in hematopoietic and other organs that are distinct from those of mice overexpressing human OSM or bovine OSM.

Oncostatin M (OSM) and leukemia inhibitory factor (LIF) belong to the interleukin-6 family of cytokines. The authors' previous in vitro work demonstrated that in mouse cells mouse OSM (mOSM) signals through a heterodimeric receptor complex incorporating the mOSM-specific receptor mOSMRbeta while human OSM (hOSM) and bovine OSM (bOSM) use the mouse LIF receptor mLIFRbeta rather than mOSMRbeta. These in vitro data suggest that prior studies in mouse systems with hOSM or bOSM (the usual molecules used in early studies) reflect LIF rather than OSM biology. The current work assessed whether or not this divergence in actions among these three OSMs also occurs in vivo in mouse models. Adult female (C57BL/6J x DBA/2J) F(1) mice were engineered to stably overexpress mOSM, hOSM, or bOSM by retrovirus-mediated gene transfer (n = 10 or more per group). After 4 weeks, molecular and hematologic profiles and anatomic phenotypes in multiple organs were assessed by standard techniques. Animals overexpressing either hOSM or bOSM had an identical phenotype resembling that associated with LIF activation, including significant hematologic abnormalities (anemia, neutrophilia, lymphopenia, eosinopenia, and thrombocytosis); weight loss; profound enlargement (lymph node, spleen) and/or structural reorganization (lymph node, spleen, thymus) of lymphoid organs; and severe osteosclerosis. In contrast, mice overexpressing mOSM did not develop hematologic changes, weight loss, or osteosclerosis and exhibited more modest and anatomically distinct restructuring of lymphoid organs. These data indicate that activities imputed to OSM and the mOSMRbeta signaling pathway using in vitro and in vivo mouse experimental systems are unique to mOSM.

Characterization and comparison of a Neurospora crassa RNase purified from cultures undergoing each of three different states of derepression.

Extracellular RNase N4 from Neurospora crassa is derepressible by limitation of any of the three nutrient elements obtainable from RNA. We have purified and characterized the enzyme from cultures grown under each of the three states of derepression. The purification procedure consisted of an ultrafiltration step, cation-exchange chromatography, and gel filtration. We found only one enzyme (N4) that hydrolyzed RNA at pH 7.5 in the presence of EDTA in culture filtrates from nitrogen-, phosphorus-, or carbon-limited cells. In all three cases, the enzymes were identical by polyacrylamide gel electrophoresis (Mr approximately 9,500) and by gel filtration (Mr approximately 10,000). There were no differences in thermal stability or pH optimum; all three cross-reacted with antibody to the nitrogen-depressed enzyme in interfacial ring and in Ouchterlony tests. Digestion of homopolyribonucleotides indicated that N4 preferentially cleaved phosphodiester bonds adjacent to guanine residues. Results indicate that the enzymes are very similar or identical and are probably products of the same gene. N4 appears to be homologous to guanine-specific RNases from other fungal sources.

Regulation of a Neurospora crassa extracellular RNase by phosphorus, nitrogen, and carbon derepressions.

A new extracellular RNase, designated N4, was detected in culture filtrates from Neurospora crassa and its regulation was studied. Limitation of a nutrient obtainable from RNA alone was not sufficient to cause enzyme derepression. The addition of RNA to the medium had no inductive effect, but the addition of exogenous protein caused enzyme production. With protein in the medium, N4 was derepressible for all three elemental nutrients obtainable from RNA: carbon, nitrogen, and phosphorus. Successful carbon derepression required the addition of a small amount of proteolytic activity to the cultures, as has been reported for the carbon-derepressible proteases of N. crassa. Exogenous protein affected RNase production before translation. Effects of the exogenous protein appeared similar to those previously reported for N. crassa protease induction. N4 was under the control of the nit-2 and nuc-1 gene products. nit-2 and nuc-1 mutants were unable to derepress enzyme synthesis for nitrogen and phosphorus limitation, respectively; however, these mutants responded like wild types to the other two states of derepression. Enzyme synthesis was constitutive in the preg mutant. Results indicate that the transcription of the N4 structural gene responds to multiple regulatory gene products from different regulatory circuits and that external protein affects the synthesis of classes of hydrolases other than proteases.

Purification and characterization of an extracellular acid protease from Neurospora crassa.

An extracellular acid protease was purified 1420-fold from sulfur-starved protein-induced cultures of Neurospora crassa. The enzyme was homogeneous as determined by polyacrylamide electrophoresis. The purification procedure consisted of an ultrafiltration step, cation-exchange chromatography, and affinity chromatography on Sepharose-linked pepstatin. The enzyme is homologous to aspartyl proteases that are characterized by pepstatin inhibition and trypsinogen activation. It is extremely autolytic, especially under denaturing conditions. The protease is stable between pH 3 and 7, showing optimal activity near pH 4.0 for both trypsinogen activation and hydrolysis of bovine serum albumin. The molecular weight of the enzyme was 34,500 by gel electrophoresis and gel filtration, and 34,975 by amino acid analysis.