Epigenomic alterations define lethal CIMP-positive ependymomas of infancy.

Ependymomas are common childhood brain tumours that occur throughout the nervous system, but are most common in the paediatric hindbrain. Current standard therapy comprises surgery and radiation, but not cytotoxic chemotherapy as it does not further increase survival. Whole-genome and whole-exome sequencing of 47 hindbrain ependymomas reveals an extremely low mutation rate, and zero significant recurrent somatic single nucleotide variants. Although devoid of recurrent single nucleotide variants and focal copy number aberrations, poor-prognosis hindbrain ependymomas exhibit a CpG island methylator phenotype. Transcriptional silencing driven by CpG methylation converges exclusively on targets of the Polycomb repressive complex 2 which represses expression of differentiation genes through trimethylation of H3K27. CpG island methylator phenotype-positive hindbrain ependymomas are responsive to clinical drugs that target either DNA or H3K27 methylation both in vitro and in vivo. We conclude that epigenetic modifiers are the first rational therapeutic candidates for this deadly malignancy, which is epigenetically deregulated but genetically bland.

Structure-function relationships of the yeast cyclin-dependent kinase Pho85.

The PHO85 gene of Saccharomyces cerevisiae encodes a cyclin-dependent kinase involved in both transcriptional regulation and cell cycle progression. Although a great deal is known concerning the structure, function, and regulation of the highly homologous Cdc28 protein kinase, little is known concerning these relationships in regard to Pho85. In this study, we constructed a series of Pho85-Cdc28 chimeras to map the region(s) of the Pho85 molecule that is critical for function of Pho85 in repression of acid phosphatase (PHO5) expression. Using a combination of site-directed and ethyl methanesulfonate-induced mutagenesis, we have identified numerous residues critical for either activation of the Pho85 kinase, interaction of Pho85 with the cyclin-like molecule Pho80, or substrate recognition. Finally, analysis of mutations analogous to those previously identified in either Cdc28 or cdc2 of Schizosaccharomyces pombe suggested that the inhibition of Pho85-Pho80 activity in mechanistically different from that seen in the other cyclin-dependent kinases.

Cloning and characterization of a member of the MST subfamily of Ste20-like kinases.

We have identified a second human homology of the yeast Ste20 protein kinase family, which we designate MST2. MST2 is most similar to the previously identified MST1 protein kinase (78% identity, 88% similarity). Northern analysis indicates that MST2 mRNA is expressed at high levels in adult kidney, skeletal and placental tissues and at very low levels in adult heart, lung, liver and brain tissues. An in vitro kinase assay indicates that MST2 can phosphorylate an exogenous substrate, as well as itself, and phospho-amino-acid analysis indicates that it is a serine/threonine protein kinase. The identification of MST2 suggests that there may be subfamilies of Ste20-like protein kinases and that MST1 and MST2 may define one of these subfamilies.

Negative transcriptional regulation of PH081 expression in Saccharomyces cerevisiae.

The PH081 gene product functions as an inhibitor of the cyclin-cyclin-dependent kinase pair, Pho80-Pho85, and is required for derepression of acid phosphatase-encoding gene (PH05) expression. PH081 is the only known regulator of this system whose transcriptional expression is regulated by the level of inorganic phosphate. This effect is mediated by the gene products of the PH04 and PH02 (BAS2, GRF10) genes which act as transcription factors. Fine structural analysis of the PH081 promoter region has revealed the existence of a negative regulatory sequence (NRS). That is, removal of this element causes an approx. 4-fold increase in PH081 expression. The NRS functions in either orientation, but only when located downstream from activation sequences. Interestingly, this element shows significant homology to a sequence present in the promoter of the PH08 gene, encoding a phosphate-repressible alkaline phosphatase. An electrophoretic mobility shift assay (EMSA) using whole-cell extracts and a NRS-containing DNA fragment detects a protein which specifically binds to this element.

Oligonucleotide-mediated, PCR-independent cloning by homologous recombination.

We have developed an oligonucleotide-mediated cloning technique based on homologous recombination in Saccharomyces cerevisiae that allows precise DNA sequences to be transferred independent of restriction enzymes and PCR. In this procedure, linear DNA sequences are targeted to a chosen site in a yeast vector by DNA linkers, which consist of two partially overlapping oligonucleotides. The linkers contain relatively short regions of both yeast vector sequences and insert sequences, which stimulate homologous recombination between the vector and the insert. The linkers can also contain sequences not found in either the vector or the insert (e.g., sequences that encode ribosome binding sites, epitope tags, preferred codons, etc.), thus allowing modification of the transferred DNA. Linkers can be designed such that DNA sequences can be transferred with just two reusable universal oligonucleotides and two gene-specific oligonucleotides. This cloning method, which is performed by co-transforming yeast with linear vector, substrate DNA, and unannealed oligonucleotides, has been termed the yeast-based, oligonucleotide-mediated gap repair technique (YOGRT).

Cloning and characterization of a human protein kinase with homology to Ste20.

A human protein kinase (termed MST1) has been cloned and characterized. The MST1 catalytic domain is most homologous to Ste20 and other Ste20-like kinases (62-65% similar). MST1 is expressed ubiquitously, and the MST1 protein is present in all human cell lines examined. Biochemical characterization of MST1 catalytic activity demonstrates that it is a serine/threonine kinase, and that it can phosphorylate an exogenous substrate as well as itself in an in vitro kinase assay. Further characterization of the protein indicates MST1 activity increases approximately 3-4-fold upon treatment with PP2A, suggesting that MST1 is negatively regulated by phosphorylation. MST1 activity decreases approximately 2-fold upon treatment with epidermal growth factor; however, overexpression of MST1 does not affect extracellular signal-regulated kinase-1 and -2 activation. MST1 is unaffected by heat shock or high osmolarity, indicating that it is not involved in the stress-activated or high osmolarity glycerol signal transduction pathways. Thus MST1, although homologous to a member of a yeast MAPK cascade, is not involved in the regulation of a known mammalian MAPK pathway and potentially regulates a novel signaling cascade.

The Ste20-like protein kinase, Mst1, dimerizes and contains an inhibitory domain.

The human serine/threonine protein kinases, Mst1 and Mst2, share considerable homology to Ste20 and p21-activated kinase (Pak) throughout their catalytic domains. However, outside the catalytic domains there are no significant homologies to previously described Ste20-like kinases or other proteins. To understand the role of the nonhomologous regions, we performed a structure/function analysis of Mst1. A series of COOH-terminal and internal deletions indicates that there is an element within a central 63-amino acid region of the molecule that inhibits kinase activity. Removal of this domain increases kinase activity approximately 9-fold. Coimmunoprecipitation assays, the yeast two-hybrid procedure, and in vitro cross-linking analysis indicate that Mst1 homodimerizes and that the extreme COOH-terminal 57 amino acids are required for self-association. Size exclusion chromatography indicates that Mst1 is associated with a high molecular weight complex in cells, suggesting that other proteins may also oligomerize with this kinase. While loss of dimerization alone does not affect kinase activity, a molecule lacking both the dimerization and inhibitory domains is not as active as one which lacks only the inhibitory domain. Comparison of Mst1 and Mst2 indicates that both functional domains lie in regions conserved between the two molecules.

Interaction of Saccharomyces cerevisiae Pho2 with Pho4 increases the accessibility of the activation domain of Pho4.

In Saccharomyces cerevisiae, expression of acid phosphatase, encoded by the PHO5 gene, requires two positive regulatory factors, Pho4 and Pho2 (also called Bas2 or Grf10). Using GAL4-PHO4 fusions, we demonstrate that a functional interaction between these two proteins is necessary for transcriptional activation to occur. This functional interaction between Pho4 and Pho2 is independent of the presence of the negative regulatory factor, Pho80, which also interacts with Pho4. Interestingly, truncations of Pho4 missing amino acids 252-265, which encompass the basic region of the basic helix-loop-helix (bHLH) DNA binding motif, exhibit high transcriptional activation that is independent of the Pho2 molecule. Single amino acid mutations of highly conserved residues within this area all display this Pho2-independent phenotype. A region near the C-terminus of Pho2 appears to be critical for this interaction with Pho4. A model to account for the requirement for Pho2 in Pho4-dependent transcriptional activation is proposed.

Molecular analysis of the PHO81 gene of Saccharomyces cerevisiae.

The PHO81 gene product is a positive regulatory factor required for the synthesis of the phosphate repressible acid phosphatase (encoded by the PHO5 gene) in Saccharomyces cerevisiae. Genetic analysis has suggested that PHO81 may be the signal acceptor molecule; however, the biochemical function of the PHO81 gene product is not known. We have cloned the PHO81 gene and sequenced the promoter. A PHO81-LacZ fusion was shown to be a valid reporter since its expression is regulated by the level of inorganic phosphate and is controlled by the same regulatory factors that regulate PHO5 expression. To elucidate the mechanism by which PHO81 functions, we have isolated and cloned dominant mutations in the PHO81 gene which confer constitutive synthesis of acid phosphatase. We have demonstrated that overexpression of the negative regulatory factor, PHO80, but not the negative regulatory factor PHO85, partially blocks the constitutive acid phosphatase synthesis in a strain containing a dominant constitutive allele of PHO81. This suggests that PHO81 may function by interacting with PHO80 or that these molecules compete for the same target.

A cysteine residue in helixII of the bHLH domain is essential for homodimerization of the yeast transcription factor Pho4p.

The yeast transcription factor Pho4p is required for expression of the phosphate-repressible acid phosphatase encoded by the PHO5 gene. Functional studies have shown that the molecule is composed of an N-terminal acidic activation domain, a central region which is necessary for interaction with a negative regulatory factor (the cyclin Pho80) and a C-terminal basic helix-loop-helix domain, which mediates DNA binding and homodimerization. In this study the homodimerization domain maps specifically to helixII of this region and a cysteine residue within this region is essential for this function. Experiments support the role of an intermolecular disulfide bond in stabilization of homodimerization, which is critical for DNA binding.

Structure and expression of the PHO80 gene of Saccharomyces cerevisiae.

In yeast, the repression of acid phosphatase under high phosphate growth conditions requires the trans-acting factor PHO80. We have determined the DNA sequence of the PHO80 gene and found that it encodes a protein of 293 amino acids. The expression of the PHO80 gene, as measured by Northern analysis and level of a PHO80-LacZ fusion protein is independent of the level of phosphate in the growth medium. Disruption of the PHO80 gene is a non-lethal event and causes a derepressed phenotype, with acid phosphatase levels which are 3-4 fold higher than the level found in derepressed wild type cells. Furthermore, over-expression of the PHO80 gene causes a reduction in the level of acid phosphatase produced under derepressed growth conditions. Finally, we have cloned, localized and sequenced a temperature-sensitive allele of PHO80 and found the phenotype to be due to T to C transition causing a substitution of a Ser for a Leu at amino acid 163 in the protein product.

Identification of a mouse p21Cdc42/Rac activated kinase.

We have isolated a novel member of the mammalian PAK (p21 activated kinase) and yeast Ste20 serine/threonine kinase family from a mouse fibroblast cDNA library, designated mPAK-3. Expression of mPAK-3 in Saccharomyces cerevisiae partially restores mating function in ste20 null cells. Like other PAKs, mPAK-3 contains a putative Cdc42Hs/Rac binding sequence and when transiently expressed in COS cells, full-length mPAK-3 binds activated (GTP gamma S (guanosine 5'-3-O-(thio-triphosphate)-bound) glutathione S-transferase (GST)-Cdc42Hs and GST-Rac1 but not GST-RhoA. As expected for a putative target molecule, mPAK-3 does not bind to an effector domain mutant of Cdc42Hs. Furthermore, activated His-tagged Cdc42Hs and His-tagged Rac stimulate mPAK-3 autophosphorylation and phosphorylation of myelin basic protein by mPAK-3 in vitro. Interestingly, the amino-terminal region of mPAK-3 contains potential SH3-binding sites and we find that mPAK-3, expressed in vitro and in vivo, shows highly specific binding to the SH3 domain of phospholipase C-gamma and at least one SH3 domain in the adapter protein Nck. These results raise the possibility of an additional level of regulation of the PAK family in vivo.

REDK, a novel human regulatory erythroid kinase.

We have identified a novel regulatory erythroid kinase (REDK) that is homologous to a family of dual-specificity kinases. The yeast homolog of REDK negatively regulates cell division, suggesting a similar function for REDK, which is primarily localized in the nucleus. REDK is present in hematopoietic tissues, such as bone marrow and fetal liver, but the RNA is expressed at significant levels only in erythroid or erythropoietin (EPO)-responsive cells. Two novel forms of cDNA (long and short) for REDK have been isolated that appear to be alternative splice products and imply the presence of polypeptides with differing amino termini. The ratio of short-to-long forms of REDK increases dramatically in CD34(+) cells cultured with EPO, suggesting differing regulation and function for each form. REDK is predominantly found in nuclear, rather than cytoplasmic, protein extracts, and immunoprecipitated REDK is active in phosphorylating histones H2b, H3, myelin basic protein, and other coimmunoprecipitated proteins. Antisense REDK oligonucleotides promote erythroid colony formation by human bone marrow cells, without affecting colony-forming unit (CFU)-GM, CFU-G, or CFU-GEMM numbers. Maximal numbers of CFU-E and burst-forming unit-erythroid were increased, and CFU-E displayed increased sensitivity to suboptimal EPO concentrations. The data indicate that REDK acts as a brake to retard erythropoiesis. (Blood. 2000;95:2838-2846)

Saccharomyces cerevisiae Yak1p protein kinase autophosphorylates on tyrosine residues and phosphorylates myelin basic protein on a C-terminal serine residue.

The serine/threonine protein kinase, Yak1p, functions as a negative regulator of the cell cycle in Saccharomyces cerevisiae, acting downstream of the cAMP-dependent protein kinase. In the present work we report that overexpression of haemagglutinin-tagged full-lengthYak1p and an N-terminally truncated form (residues 148-807) lead to growth arrest in PKA compromised yak1 null yeast cells. Both forms of recombinant Yak1p kinase were catalytically active and preferred myelin basic protein (MBP) as a substrate over several other proteins. Phosphopeptide analysis of bovine MBP by tandem MS revealed two major Yak1p phosphorylation sites, Thr-97 and Ser-164. Peptides containing each site were obtained and tested as Yak1p substrates. Both forms of Yak1p phosphorylated a peptide containing the Ser-164 residue with far more efficient kinetics than MBP. The maximal velocity (V(max)) values of the full-length Yak1p reaction were 110+/-21 (Ser-164) and 8.7+/-1.7 (MBP), and those of N-terminally truncated Yak1p were 560.7+/-74.8 (Ser-164) and 34. 4+/-2.2 (MBP) pmol/min per mg of protein. Although neither form of Yak1p was able to phosphorylate two generic protein tyrosine kinase substrates, both were phosphorylated on tyrosine residues in vivo and underwent tyrosine autophosphorylation when reacted with ATP in vitro. Tandem MS showed that Tyr-530 was phosphorylated both in vivo and in vitro after reaction with ATP. Pre-treatment with protein tyrosine phosphatase 1B removed all of Yak1p phosphotyrosine content and drastically reduced Yak1p activity against exogenous substrates, suggesting that the phosphotyrosine content of the enzyme is essential for its catalytic activity. Although the N-terminally truncated Yak1p was expressed at a lower level than the full-length protein, its catalytic activity and phosphotyrosine content were significantly higher than those of the full-length enzyme. Taken together, our results suggest that Yak1p is a dual specificity protein kinase which autophosphorylates on Tyr-530 and phosphorylates exogenous substrates on Ser/Thr residues.

Expression, purification, and functional analysis of the human serine protease HtrA2.

HumHtrA2 or Omi is a recently described member of a novel family of mammalian serine proteases homologous to the Escherichia coli htrA gene product. Although the physiological function of members of this new family is unclear, the current understanding is that as well as being involved with the degradation aberrantly folded proteins during conditions of cellular stress, they may possess a chaperone-like role under normal conditions. In this report we describe the overexpression of humHtrA2 in two heterologous systems comparing the merits of each. We found that molecular analysis of processing events in Sf9 cells allowed us to revisit E. coli expression systems which were initially unsuccessful. Using E. coli we were able to produce milligram amounts of >90% pure recombinant enzyme as determined by SDS-PAGE gels. By means of fluorescently labeled substrates alpha- and beta-casein and zymography, the proteolytic activity of recombinant HumHtrA2 was also demonstrated.

Fission yeast pak1+ encodes a protein kinase that interacts with Cdc42p and is involved in the control of cell polarity and mating.

A STE20/p65pak homolog was isolated from fission yeast by PCR. The pak1+ gene encodes a 72 kDa protein containing a putative p21-binding domain near its amino-terminus and a serine/threonine kinase domain near its carboxyl-terminus. The Pak1 protein autophosphorylates on serine residues and preferentially binds to activated Cdc42p both in vitro and in vivo. This binding is mediated through the p21 binding domain on Pak1p and the effector domain on Cdc42p. Overexpression of an inactive mutant form of pak1 gives rise to cells with markedly abnormal shape with mislocalized actin staining. Pak1 overexpression does not, however, suppress lethality associated with cdc42-null cells or the morphologic defeat caused by overexpression of mutant cdc42 alleles. Gene disruption of pak1+ establishes that, like cdc42+, pak1+ function is required for cell viability. In budding yeast, pak1+ expression restores mating function to STE20-null cells and, in fission yeast, overexpression of an inactive form of Pak inhibits mating. These results indicate that the Pak1 protein is likely to be an effector for Cdc42p or a related GTPase, and suggest that Pak1p is involved in the maintenance of cell polarity and in mating.

mDYRK3 kinase is expressed selectively in late erythroid progenitor cells and attenuates colony-forming unit-erythroid development.

DYRKs are a new subfamily of dual-specificity kinases that was originally discovered on the basis of homology to Yak1, an inhibitor of cell cycle progression in yeast. At present, mDYRK-3 and mDYRK-2 have been cloned, and mDYRK-3 has been characterized with respect to kinase activity, expression among tissues and hematopoietic cells, and possible function during erythropoiesis. In sequence, mDYRK-3 diverges markedly in noncatalytic domains from mDYRK-2 and mDYRK-1a, but is 91.3% identical overall to hDYRK-3. Catalytically, mDYRK-3 readily phosphorylated myelin basic protein (but not histone 2B) and also appeared to autophosphorylate in vitro. Expression of mDYRK-1a, mDYRK-2, and mDYRK-3 was high in testes, but unlike mDYRK1a and mDYRK 2, mDYRK-3 was not expressed at appreciable levels in other tissues examined. Among hematopoietic cells, however, mDYRK-3 expression was selectively elevated in erythroid cell lines and primary pro-erythroid cells. In developmentally synchronized erythroid progenitor cells, expression peaked sharply following exposure to erythropoietin plus stem cell factor (SCF) (but not SCF alone), and in situ hybridizations of sectioned embryos revealed selective expression of mDYRK-3 in fetal liver. Interestingly, antisense oligonucleotides to mDYRK-3 were shown to significantly and specifically enhance colony-forming unit-erythroid colony formation. Thus, it is proposed that mDYRK-3 kinase functions as a lineage-restricted, stage-specific suppressor of red cell development. (Blood. 2001;97:901-910)

Characterization of human HtrA2, a novel serine protease involved in the mammalian cellular stress response.

Human HtrA2 is a novel member of the HtrA serine protease family and shows extensive homology to the Escherichia coli HtrA genes that are essential for bacterial survival at high temperatures. HumHtrA2 is also homologous to human HtrA1, also known as L56/HtrA, which is differentially expressed in human osteoarthritic cartilage and after SV40 transformation of human fibroblasts. HumHtrA2 is upregulated in mammalian cells in response to stress induced by both heat shock and tunicamycin treatment. Biochemical characterization of humHtrA2 shows it to be predominantly a nuclear protease which undergoes autoproteolysis. This proteolysis is abolished when the predicted active site serine residue is altered to alanine by site-directed mutagenesis. In human cell lines, it is present as two polypeptides of 38 and 40 kDa. HumHtrA2 cleaves beta-casein with an inhibitor profile similar to that previously described for E. coli HtrA, in addition to an increase in beta-casein turnover when the assay temperature is raised from 37 to 45 degrees C. The biochemical and sequence similarities between humHtrA2 and its bacterial homologues, in conjunction with its nuclear location and upregulation in response to tunicamycin and heat shock suggest that it is involved in mammalian stress response pathways.