Phosphorylation of Mei2 and Ste11 by Pat1 kinase inhibits sexual differentiation via ubiquitin proteolysis and 14-3-3 protein in fission yeast.

Fission yeast Pat1 kinase inhibits sexual differentiation by phosphorylating the meiotic inducer Mei2 and the transcription factor Ste11. Here, we show how Pat1 downregulates these proteins. Mei2 is degraded via a ubiquitin-proteasome pathway in a phosphorylation-dependent fashion. The E2 Ubc2 and the E3 Ubr1 are required for this proteolysis. In addition, Pat1 negatively regulates Ste11 via Rad24/14-3-3, thereby repressing mei2+ transcription. The Pat1 phosphorylation sites of Ste11 match the consensus recognition sequence for 14-3-3. Rad24 binds preferentially to phosphorylated Ste11, and this binding results in inhibition of the transcriptional activation capacity of Ste11. Overall, therefore, these results show that Pat1 coordinates concerted molecular mechanisms that govern the sexual differentiation developmental decision.

The MAPK kinase Pek1 acts as a phosphorylation-dependent molecular switch.

The mitogen-activated protein kinase (MAPK) pathway is a highly conserved eukaryotic signalling cascade that converts extracellular signals into various outputs, such as cell growth and differentiation. MAPK is phosphorylated and activated by a specific MAPK kinase (MAPKK): MAPKK is therefore considered to be an activating regulator of MAPK. Pmk1 is a MAPK that regulates cell integrity and which, with calcineurin phosphatase, antagonizes chloride homeostasis in fission yeast. We have now identified Pek1, a MAPKK for Pmk1 MAPK. We show here that Pek1, in its unphosphorylated form, acts as a potent negative regulator of Pmk1 MAPK signalling. Mkh1, an upstream MAPKK kinase (MAPKKK), converts Pek1 from being an inhibitor to an activator. Our results indicate that Pek1 has a dual stimulatory and inhibitory function which depends on its phosphorylation state. This switch-like mechanism could contribute to the all-or-none physiological response mediated by the MAPK signalling pathway.

The fission yeast pmk1+ gene encodes a novel mitogen-activated protein kinase homolog which regulates cell integrity and functions coordinately with the protein kinase C pathway.

We have isolated a gene, pmk1+, a third mitogen-activated protein kinase (MAPK) gene homolog from the fission yeast Schizosaccharomyces pombe. The predicted amino acid sequence shows the most homology (63 to 65% identity) to those of budding yeast Saccharomyces Mpk1 and Candida Mkc1. The Pmk1 protein contains phosphorylated tyrosines, and the level of tyrosine phosphorylation was increased in the dsp1 mutant which lacks an attenuating phosphatase for Pmk1. The level of tyrosine phosphorylation appears constant during hypotonic or heat shock treatment. The cells with pmk1 deleted (delta pmk1) are viable but show various defective phenotypes, including cell wall weakness, abnormal cell shape, a cytokinesis defect, and altered sensitivities to cations, such as hypersensitivity to potassium and resistance to sodium. Consistent with a high degree of conservation of amino acid sequence, multicopy plasmids containing the MPK1 gene rescued the defective phenotypes of the delta pmk1 mutant. The frog MAPK gene also suppressed the pmk1 disruptant. The results of genetic analysis indicated that Pmk1 lies on a novel MAPK pathway which does not overlap functionally with the other two MAPK pathways, the Spk1-dependent mating signal pathway and Sty1/Spc1/Phh1-dependent stress-sensing pathway. In Saccharomyces cerevisiae, Mpk1 is involved in cell wall integrity and functions downstream of the protein kinase C homolog. In contrast, in S. pombe, Pmk1 may not act in a linear manner with respect to fission yeast protein kinase C homologs. Interestingly, however, these two pathways are not independent; instead, they regulate cell integrity in a coordinate manner.

A new group of conserved coactivators that increase the specificity of AP-1 transcription factors.

The Jun proteins are nuclear proteins that combine with Fos proteins to form a gene-regulatory protein, AP-1. They have highly conserved DNA-binding and dimerization domains, resulting in almost identical sequence-recognition properties. Nevertheless, there are many indications that each Jun protein activates a distinct and only partially overlapping set of AP-1 target genes. Using the more variable activation domain of c-Jun as a bait, we identified a protein, JAB1, that interacts with c-Jun and JunD, but not with JunB or v-Jun. As a result, JAB1 selectively potentiates transactivation by only c-Jun or JunD. In vitro, JAB1 specifically stabilizes complexes of c-Jun or JunD with AP-1 sites and does not affect binding of either JunB or v-Jun. The amino-terminal half of JAB1 is very similar to the amino terminal region of Pad1 from fission yeast, which was identified genetically as a coactivator of a subset of AP-1 target genes. JAB1 and Pad1 are also functionally interchangeable. They define a new group of coactivators that increase the specificity of target gene activation by AP-1 proteins.

The fission yeast sts5+ gene is required for maintenance of growth polarity and functionally interacts with protein kinase C and an osmosensing MAP-kinase pathway.

Cell morphogenesis is a fundamental phenomenon that involves understanding a number of biological processes including the developmental program, polarity and cell division. Fission yeast sts5 mutant cells are round rather than cylindrical with cortical actin randomly dispersed. Genetic analyses demonstrate that the sts5+ gene is required for maintenance of cell shape during interphase when the cell normally exhibits polarised growth. The sts5 mutant is not defective in cell wall integrity. Deletion of ppe1+, which encodes a type 2A-like protein phosphatase, shows similar phenotypes to the sts5 mutant and these two mutations are synthetically lethal. Multicopy plasmids containing either the protein kinase C-like gene pck1+ or the protein tyrosine phosphatase pyp1+, an inhibitor of an osmosensing Sty1/Spc1 MAP-kinase, are capable of suppressing the sts5 mutation. Consistent with this, we have found that the wis1 mutation, which is defective in a MAP-kinase kinase of the pathway, suppresses the sts5 mutation. The predicted sts5+ gene product exhibits sequence similarity to two yeast proteins, Dis3 and Ssd1 and a nematode protein, F46E8.6, where the former two yeast proteins have been shown to be involved in cell cycle control and cell morphogenesis. The sts5+ gene is not essential for cell viability, but is absolutely required for polarised growth as the gene disruption showed the same phenotypes as those of the original mutants. Overexpression of the sts5+ gene resulted in altered cell morphology and, cortical actin in these overproducing cells was also abnormal, fainter and often dispersed. Anti-Sts5 antibody specifically detected a 130 kDa protein by western blotting. A green fluorescent protein-Sts5 fusion protein localised in the cytoplasm with a discrete punctate pattern, suggesting that the Sts5 protein is a component of a novel structure. These results have indicated that the Sts5 protein is a crucial determinant of polarised growth and that it functionally interacts with the serine/threonine phosphatase, protein kinase C, and an osmosensing MAP-kinase to maintain cell morphology.

Molecular cloning of a novel 11q23 breakpoint associated with non-Hodgkin's lymphoma.

Chromosomal analysis of a non-Hodgkin's lymphoma revealed a t(11;14)(q23;q32) translocation amongst other abnormalities. To investigate the molecular basis of this translocation, a cosmid library was constructed from the tumour DNA and the rearranged IGH locus was isolated in a single cosmid. Fluorescence in situ hybridization confirmed that the cloned region contained sequences from chromosome 11q23 fused to chromosome 14q32. Sequence analysis identified the breakpoint as a fusion between a region from the switch segment of the C gamma 4 gene of the IGH locus and an unknown sequence on chromosome 11. The chromosome 11 sequence maps proximal to the CD3 gene cluster and is therefore distinct from both the HTRX1 gene (rearranged in acute leukaemias) and the RCK gene (rearranged in a cell line derived from a histiocytic B-cell lymphoma). This newly identified region contains a cluster of rare cutting restriction enzyme sites located within 200 bases of the breakpoint, suggestive of a CpG island. Although this t(11;14)(q23;q32) translocation and that in the RC-K8 cell line affect different regions on chromosome 11, the breakpoints on chromosome 14 were found to have occurred at equivalent positions of S gamma 2 and S gamma 4 segments.

Detection and significance of bcr-abl mRNA transcripts and fusion proteins in Philadelphia-positive adult acute lymphoblastic leukemia.

Blast cells from an unselected consecutive series of 84 adults presenting with acute lymphoblastic leukemia (ALL) to St Bartholomew's Hospital over a seven year period were tested prospectively by cytogenetic and retrospectively by RT-PCR analysis for the presence of the Ph translocation and bcr-abl mRNA. This combination gave an overall figure of 20.3% for bcr-abl-positive and/or Ph-positive ALL. The incidence of bcr-abl-positive/Ph-positive ALL was most common between the ages of 31 and 50 years, becoming less common after the age of 50. Eight out of ten bcr-abl-positive patients expressed the e1a2 mRNA transcript, the other two expressed the b3a2 and b2a3 transcripts respectively. Cells from all patients with bcr-abl mRNA transcripts expressed the appropriate p190 or p210 bcr-abl protein and all were Ph-positive.

Expression of the ABL-BCR fusion gene in Philadelphia-positive acute lymphoblastic leukemia.

We have previously shown that the chimeric gene ABL-BCR, formed on the derivative chromosome 9q+ as a result of the t(9;22) translocation, is transcriptionally active in 65% of chronic myeloid leukemia patients. We have now used the same technique, reverse transcription/polymerase chain reaction amplification of ABL-BCR transcripts, to study nine patients with Philadelphia (Ph) chromosome-positive acute lymphoblastic leukemia (ALL); seven expressed the P190 and two the P210 type of BCR-ABL fusion protein. All seven patients with P190 had ABL-BCR transcripts containing a junction between ABL exon Ib and BCR exon 2 (Ib-e2); in two cases, ABL-BCR transcripts with the Ia-e2 junction type were also present. Of the two P210 ALL patients, one had a Ib-b4 ABL-BCR transcript and the other showed no detectable ABL-BCR expression. Although the BCR-ABL gene is probably fundamental in the pathogenesis of the Ph+ leukemias, differential expression of the ABL-BCR gene could contribute to the biologic heterogeneity of the disease.

Fluorescent in situ identification of human marker chromosomes using flow sorting and Alu element-mediated PCR.

A novel approach to the identification of human chromosomes has been developed. Chromosomal in situ hybridization (or "chromosome painting") has been performed using Alu element-mediated PCR products from small quantities (250-500) of flow-sorted normal and abnormal chromosomes. Chromosome paints for various normal chromosomes, including 5, 6, 7, 14, 18, 19, 21, and 22, were generated and shown to be effective in the identification of the appropriate chromosomes. In addition, certain abnormal chromosomes, including a mental retardation-associated deletion chromosome 11 (q22-q23), the products of the constitutional translocation t(11;22), and the CML-associated t(9;22), were used to generate region-specific paints. In each case, the appropriate regions of the chromosomes were highlighted and this strategy is, therefore, well suited to the identification of previously unidentified marker chromosomes. A further direct consequence of this work is that chromosome paints specific for the common aberrant chromosomes, such as the Philadelphia chromosome, can be generated and made widely available. These may find particular use in the analysis of complex or masked chromosomal translocations.

Normal c-abl gene protein--a nuclear component.

The subcellular distribution of the c-abl and bcr-abl gene products from KG1A and K562 cells has been studied by two different techniques. Firstly, physical disruption followed by subcellular fractionation was used to demonstrate that normal c-abl (p145) was recovered from the cytosol and the nuclear fractions of KG1A cells. In contrast, bcr-abl products were recovered exclusively from the cytosol fraction of K562 cells. Secondly, indirect immunofluorescence was used to localize c-abl protein to the cytoplasm, nuclear membrane and infrequently to the nucleus of KG1A cells and bcr-abl protein to only the cytoplasm of K562 cells. Thus both the approaches indicate that there is a component of normal c-abl products which appears to be nuclear and this is not reflected in the distribution of the bcr-abl 210 kDa protein, which remains cytosolic.

CD3G is within 200 kb of the leukemic t(4;11) translocation breakpoint.

The t(4;11)(q21;q23) has been associated with acute lymphocytic leukemia (ALL) especially in infants. The t(4;11) breakpoint on chromosome 11 is cytogenetically indistinguishable from breakpoints for other leukemia-associated translocations affecting 11q23. The molecular basis of the t(4;11) is unknown although a number of genes have been mapped to 11q23. The CD3D, G, and E genes have been positioned proximal to the 11q23 breakpoint of the 4;11 translocation while the THY1 and ETS1 genes have been mapped distal to this breakpoint. We report evidence that CD3G is within 200 kb of the 4;11 breakpoint as observed by pulsed field gel analysis. A rearrangement of the CD3G gene has been observed in a cell line derived from a patient with the t(4;11) translocation and in a hybrid cell line containing the derivative 11q chromosome derived from this cell line, using the restriction enzymes SacII and ClaI. Similar rearrangements using SacII were observed in 2 further patients with ALL and the t(4;11) translocation. No rearrangements in the same DNA were observed using ETS1, THY1, and D11S29 and a range of rare cutter restriction enzymes. CD3G thus provides a tool for the cloning and analysis of the 4;11 translocation, and poses a question of its possible involvement at long range with this translocation.

Establishment of a lymphoblastoid cell line, SD-1, expressing the p190 bcr-abl chimaeric protein.

A lymphoblastoid cell line (SD-1) has been established by Epstein-Barr virus immortalisation of Philadelphia chromosome positive acute lymphoblastic leukaemia (ALL) cells. Using newly derived anti-bcr monoclonal and anti-abl polyclonal antibodies it was demonstrated that both the original leukaemic cells and the derived cell line expressed the p190 form of the bcr-abl protein found in a proportion of cases of Philadelphia chromosome positive ALL. Interestingly, the leukaemia and the derived cell line each displayed different, clonal patterns of immunoglobulin gene rearrangements providing direct evidence that the t(9;22) translocation which results in the expression of the p190 bcr-abl protein must occur before immunoglobulin heavy chain gene rearrangement. In contrast to the leukaemia, which had multiple chromosome abnormalities in addition to the t(9;22), the cell line had the t(9;22) translocation as its sole abnormality. Although SD-1 cells were demonstrated to express continuously the p190 bcr-abl protein, they were unable to form colonies in soft agar and did not cause tumours in splenectomised nude mice. This cell line therefore represents an appropriate target cell line in which to examine the cooperativity of the p190 bcr-abl protein with other activated oncogene products.

BCR-ABL and BCR proteins: biochemical characterization and localization.

The Philadelphia translocation results in the expression of a family of chimaeric proteins in which a portion of the bcr protein is fused to c-abl protein. Using antibodies which recognize different portions of the bcr gene and abl gene products we have compared the normal bcr products with their chimaeric counterparts. We first conclude that the enhanced kinase activity of the rearranged bcr-abl products (p210 and p190) is recovered almost exclusively from the cytosolic fraction. This methodology was confirmed by the demonstration that in cells transformed by the Abelson murine leukemia virus (A-MuLV) the gag-abl kinase activity was recovered equally from the membrane and cytosolic fractions, in agreement with previous studies. To determine whether the distribution of kinase activity reflected the bulk distribution of the bcr-abl proteins, in vivo labeling followed by subcellular fractionation was performed. Both normal bcr proteins and the p210 bcr-abl protein were recovered from the cytosolic fraction with little detectable amounts present in other fractions. In vivo labeling was also used to demonstrate that both normal bcr products and the p210 bcr-abl had a relatively long half-life. It is concluded that bcr-abl products, like normal bcr products are located in the cytosolic fraction.

Inhibition of P210BCR/ABL Expression in K562 Cells by Electroporation with an Antisense Oligonucleotide.

We designed experiments to study the effects on P210BCR/ABL expression of introducing antisense oligonucleotides into K562 cells. We used two antisense oligonucleotides: one (AS1) is complementary to the first coding codon of the BCR/ABL mRNA and the two 5' and three 3' codons, and the other (AS2) to BCR coding codons 5 to 11 inclusive. To facilitate entry of the oligonucleotides the K562 cells were subjected to electroporation on three occasions at 24 hr intervals (0, 24 and 48 hr). P210BCR/ABL expression was assayed by in vivo phosphorylation followed by immune precipitation with a BCR antibody. Introduction of AS1 inhibited P210BCR/ABL expression at 72 and 96 hrs, whereas AS2 and the control oligonucleotide had no effect. AS1 also killed K562 cells. We conclude that selected antisense oligonucleotides can modify leukaemia-specific protein expression in K562 cells. This approach could prove valuable for purging CML bone marrow cells in vitro.

Identification of two normal bcr gene products in the cytoplasm.

A monoclonal antibody (7C6) has been derived against a synthetic bcr peptide and used to study normal bcr gene products. The expression of a bcr phosphoprotein of 130 kd was demonstrated, in addition to the previously identified bcr phosphoprotein of 160 kd. Sequential immunoprecipitation demonstrated that both p160 and p130 had determinants from two separate regions of the putative bcr translated sequence. The synthesis of bcr products in Philadelphia positive and negative cells was examined by metabolic labelling and it was shown that the rate of synthesis of the p210 bcr-abl product was comparable with that of the normal bcr products. The in vivo phosphorylation of the p160 exceeded that of the p130 and both normal products were unaffected by the increased phosphorylation of the p210 bcr-abl. There was no evidence with the 7C6 antibody of any normal bcr products larger than 160 kilodaltons. Immunofluorescence analysis by conventional and confocal microscopy identified normal bcr products as cytoplasmic proteins with relatively high expression in the myeloid cell line KGl.

Novel chimaeric protein expressed in Philadelphia positive acute lymphoblastic leukaemia.

Cytogenic changes are becoming increasingly important in understanding the pathogenesis of human malignancies. The t(9;22) (q34;q11) translocation is one of the most consistent and generates the Philadelphia chromosome (Ph1) (ref. 1) in chronic myeloid leukaemia (CML); it has also been observed in some acute lymphoblastic leukaemias (ALL) (ref. 2). In CML the breakpoints occur on chromosome 22 in the region designated bcr (ref. 3) and result in the expression of a bcr-abl fusion product of relative molecular mass (MT) 210,000 (210K) with associated in vitro tyrosine kinase activity (P210bcr-abl). In some cases of Ph1-positive ALL, a novel abl-related protein (P190all-abl) of 190K has been shown to have tyrosine kinase activity. In this report we demonstrate that the P190all-abl protein has a bcr determinant from the amino-terminal region, but is lacking a bcr determinant normally found in the P210bcr-abl near the bcr-abl junction. The chimaeric nature of the P190all-abl was confirmed by sequential immunoprecipitation with antisera against abl and bcr peptides.

Cross-reactive variable-region associated epitopes of human IgG1 lambda paraprotein detected by a monoclonal antibody panel.

This paper describes the production and characterization of a panel of 15 mouse monoclonal antibodies selected for putative activity against V-region related or allotypic determinants of a single IgG1 lambda paraprotein obtained from a patient with malignant lymphoplasmacytoid lymphoma. The specificity of each reagent for epitopes on the heavy (H) or light (L) chain or for conformational determinants (CD) of the immunogen was determined and the ability of one reagent to compete with another for these sites investigated. The fine specificity of the antibodies was assessed by screening on a large series of normal and paraprotein-containing sera. One monoclonal showed specificity for the Glm(f) allotype. The 14 other reagents identified a minimum of nine different epitopes in the V region of the immunogen, with four antibodies detecting private conformationally determined idiotypic specificities. Eight determinants were V-region markers also expressed on other paraproteins. A total of 30 out of 159 different paraproteins cross-reacted with one or more of the antibodies. Four of the shared epitopes were lambda-chain associated, three were H-chain associated and one was a conformational determinant. Homologies of lambda chain were identified more frequently among other paraproteins than those of H chain. The relationship between epitope expression and H-chain class of paraprotein was not random. The frequency of expression of cross-reactivities in association with IgG1 proteins was always exceeded by higher frequency of epitope expression in association with other classes of H-chain isotype. The potential therapeutic value of such panels of characterized monoclonal reagents is discussed.

Two monoclonal antibodies raised against a Burkitt lymphoma cell line recognise different cell types within lymphoid follicles.

Monoclonal antibodies were raised against a laboratory derived variant of the Raji Burkitt lymphoma cell line. We have characterized two of these antibodies by screening on haematopoietic cell lines, and frozen sections of both reactive and neoplastic lymphoid tissue, and found reactivity with separate cellular components of lymphoid follicles. Monoclonal FW37.4.D5 reacts in section specifically with follicular dendritic reticular cells. Monoclonal 35.1C5 reacts with a subpopulation of splenic marginal zone, and lymph node mantle zone, B lymphocytes, but stains only rare cells within germinal centres. This monoclonal is operationally B lymphocyte specific, distinguishing an 'intermediate' B cell subset, represented in lymphoma by B chronic lymphocytic leukaemia, B hairy cell leukaemia and centrocytic lymphoma.

Extracellular matrix antigen of human muscle defined by a monoclonal antibody.

A monoclonal antibody (7.1B3) has been produced in mice previously immunised with a glycoprotein fraction prepared from human myotube cells grown in tissue culture. The antibody precipitates an antigen complex of four glycoprotein components from detergent extracted human muscle myotube cultures. These are two chains of about 205K MW and two chains of about 130-140K MW. The antigen complex is trypsin-sensitive collagenase insensitive and bands as a complex of 10S on sucrose gradients in NP40. Indirect immunofluorescence studies on cell cultures of myoblasts and myotubes showed that the antigen existed in a number of locations: as a granular stain on the cell surface of myotubes, and as short streaks of immunoreactivity in the extracellular matrix. Clonal populations of myoblasts and myotubes gave a similar result showing that 7.1B3 is specifically synthesized by cells of the myogenic lineage. Double indirect immunofluorescence staining with 7.1B3 antibody and fibronectin antibody showed no overlap of immunoreactivity indicating that these two antigens are different. 7.1B3 antigen was found during human fetal muscle development in vivo at all time points studied from 15 weeks onwards. Immunoreactivity was localised to the loose extracellular matrix and in the endomysium. Perymysium staining was very prominent at 25 weeks and in adult muscle. Double staining of 7.1B3 antibody and alpha-bungarotoxin was carried out to determine whether 7.1B3 was present in synaptic junctions. Although 7.1B3 was found in the endomysium it is excluded from synaptic areas. We conclude that 7.1B3 antibody defines a new and novel antigen in the extracellular matrix of human muscle in vivo and in vitro.