Search

The tyrosine-based activation motif is a 20- to 25-amino-acid sequence contained in the cytoplasmic domains of many hematopoietic receptors which is sufficient by itself to reconstitute signalling. This motif is characterized by two YXXL/I sequences separated by approximately 10 residues. The molecular basis of signalling by this motif is unknown. Here we demonstrate that the tyrosine-based activation motif is required and sufficient for association with the tyrosine kinases p59fyn and ZAP-70, suggesting that association with these kinases is a general feature of this motif. Focusing on the single activation motif present in epsilon, we analyzed which residues of the motif were critical for binding of p59fyn and ZAP-70. Surprisingly, we found that no single mutation of any residue of epsilon resulted in the loss of p59fyn association. In contrast, single mutations at five residues of the epsilon activating motif abrogated ZAP-70 binding. Both of the tyrosines and the leucine or isoleucine residues that follow them were critical. The spacing between the tyrosines was also important, as deletion of two residues disrupted binding of ZAP-70, although p59fyn binding was not disrupted. Most of the defined features of the tyrosine activation motif are therefore requirements for ZAP-70 binding. Interestingly, the interaction of ZAP-70 with the motif was dependent on the presence of both ZAP-70 SH2 domains and both of the tyrosine residues in the motif, suggesting that ZAP-70 interacts with two phosphotyrosine residues and that the binding of the two SH2 domains is cooperative. In addition, we demonstrate that the interaction between the tyrosine activation motif is direct and requires prior tyrosine phosphorylation of the motif. We propose that the activation of cells by the tyrosine activating motif occurs in four discrete steps: binding of p59fyn, phosphorylation of the motif, binding of ZAP-70, and activation of ZAP-70 kinase activity.The tyrosine-based activation motif is a 20- to 25-amino-acid sequence contained in the cytoplasmic domains of many hematopoietic receptors which is sufficient by itself to reconstitute signalling. This motif is characterized by two YXXL/I sequences separated by approximately 10 residues. The molecular basis of signalling by this motif is unknown. Here we demonstrate that the tyrosine-based activation motif is required and sufficient for association with the tyrosine kinases p59fyn and ZAP-70, suggesting that association with these kinases is a general feature of this motif. Focusing on the single activation motif present in epsilon, we analyzed which residues of the motif were critical for binding of p59fyn and ZAP-70. Surprisingly, we found that no single mutation of any residue of epsilon resulted in the loss of p59fyn association. In contrast, single mutations at five residues of the epsilon activating motif abrogated ZAP-70 binding. Both of the tyrosines and the leucine or isoleucine residues that follow them were critical. The spacing between the tyrosines was also important, as deletion of two residues disrupted binding of ZAP-70, although p59fyn binding was not disrupted. Most of the defined features of the tyrosine activation motif are therefore requirements for ZAP-70 binding. Interestingly, the interaction of ZAP-70 with the motif was dependent on the presence of both ZAP-70 SH2 domains and both of the tyrosine residues in the motif, suggesting that ZAP-70 interacts with two phosphotyrosine residues and that the binding of the two SH2 domains is cooperative. In addition, we demonstrate that the interaction between the tyrosine activation motif is direct and requires prior tyrosine phosphorylation of the motif. We propose that the activation of cells by the tyrosine activating motif occurs in four discrete steps: binding of p59fyn, phosphorylation of the motif, binding of ZAP-70, and activation of ZAP-70 kinase activity.

The tyrosine-based activation motif is a 20- to 25-amino-acid sequence contained in the cytoplasmic domains of many hematopoietic receptors which is sufficient by itself to reconstitute signalling. This motif is characterized by two YXXL/I sequences separated by approximately 10 residues. The molecular basis of signalling by this motif is unknown. Here we demonstrate that the tyrosine-based activation motif is required and sufficient for association with the tyrosine kinases p59fyn and ZAP-70, suggesting that association with these kinases is a general feature of this motif. Focusing on the single activation motif present in epsilon, we analyzed which residues of the motif were critical for binding of p59fyn and ZAP-70. Surprisingly, we found that no single mutation of any residue of epsilon resulted in the loss of p59fyn association. In contrast, single mutations at five residues of the epsilon activating motif abrogated ZAP-70 binding. Both of the tyrosines and the leucine or isoleucine residues that follow them were critical. The spacing between the tyrosines was also important, as deletion of two residues disrupted binding of ZAP-70, although p59fyn binding was not disrupted. Most of the defined features of the tyrosine activation motif are therefore requirements for ZAP-70 binding. Interestingly, the interaction of ZAP-70 with the motif was dependent on the presence of both ZAP-70 SH2 domains and both of the tyrosine residues in the motif, suggesting that ZAP-70 interacts with two phosphotyrosine residues and that the binding of the two SH2 domains is cooperative. In addition, we demonstrate that the interaction between the tyrosine activation motif is direct and requires prior tyrosine phosphorylation of the motif. We propose that the activation of cells by the tyrosine activating motif occurs in four discrete steps: binding of p59fyn, phosphorylation of the motif, binding of ZAP-70, and activation of ZAP-70 kinase activity.

Signal transduction by antigen receptors and some Fc receptors requires the activation of a family of receptor-associated transmembrane accessory proteins. One common feature of the cytoplasmic domains of these accessory molecules is the presence is at least two YXXA repeats that are potential sites for interaction with Src homology 2 domain-containing proteins. However, the degree of similarity between the different receptor-associated proteins varies from that of T-cell receptor (TCR) zeta and Fc receptor RIIIA gamma chains, which are homologous, to the distantly related Ig alpha and Ig beta proteins of the B-cell antigen receptor. To determine whether T- and B-cell antigen receptors are in fact functionally homologous, we have studied signal transduction by chimeric immunoglobulins bearing the Ig alpha or Ig beta cytoplasmic domain. We found that Ig alpha and Ig beta cytoplasmic domains were able to activate Ca2+ flux, interleukin-2 secretion, and phosphorylation of the same group of cellular substrates as the TCR in transfected T cells. Chimeric proteins were then used to examine the minimal requirements for activation of the Fyn, Lck, and ZAP kinases in T cells. Both Ig alpha and Ig beta were able to trigger Fyn, Lck, and ZAP directly without involvement of TCR components. Cytoplasmic tyrosine residues in Ig beta were required for recruitment and activation of ZAP-70, but these amino acids were not essential for the activation of Fyn and Lck. We conclude that Fyn and Lck are able to recognize a clustered nonphosphorylated immune recognition receptor, but activation of these kinases is not sufficient to induce cellular responses such as Ca2+ flux and interleukin-2 secretion. In addition, the molecular structures involved in antigen receptor signaling pathways are conserved between T and B cells.

cDNAs for the murine lyn protein tyrosine kinase gene were cloned from mouse bone marrow-derived monocytic cells. Comparison of the human and murine genes demonstrated a 94% homology in peptide sequence. Comparable to the human gene, murine lyn was found to be expressed in myeloid and B-lymphoid lineage cells. During the cloning, two types of cDNAs were obtained that differed by the presence (lynA) or absence (lynB) of 63 bp within the amino-terminal coding region of the gene. The genomic structure of the murine lyn gene demonstrates that the two types of lyn transcripts are derived from alternative splicing utilizing an internal splice donor site. Transcripts for both forms were found to be expressed in myeloid cells. lyn-specific antisera detected comparable levels of proteins of 56 and 53 kDa in hematopoietic cells. these 56- and 53-kDa proteins comigrated with proteins produced by in vitro translation or in vivo expression of the lynA and lynB cDNAs, respectively. The two forms had comparable in vitro kinase activities in immunoprecipitates and showed similar peptide patterns, with partial V8 digestion of the in vitro-phosphorylated proteins. The potential significance of the two lyn proteins is discussed.

Neuronal cells are known to express at least two different forms of the C-SRC proto-oncogene as a consequence of alternative splicing events which add an 18-nucleotide exon (the NI exon) between C-SRC exons 3 and 4. Here we report that a second neuronal exon of C-SRC is also present between C-SRC exons 3 and 4. This neuronal exon (the NII exon) of C-SRC was isolated from human adult and fetal brain-derived cDNAs and contains 33 nucleotides capable of encoding 11 amino acids (Gln-Thr-Trp-Phe-Thr-Phe-Arg-Trp-Leu-Gln-Arg). The human NI exon was located approximately 390 nucleotides from the end of C-SRC exon 3, whereas the NII exon was approximately 1,000 nucleotides from the beginning of C-SRC exon 4. Analysis of human brain RNA revealed that the NII exon is utilized primarily in conjunction with the NI exon to yield transcripts capable of encoding C-SRC products possessing 17 additional amino acids. These splicing events, which occur between the NI and NII exons, are predicted to alter the sixth amino acid encoded by the NI exon from an arginine to a serine residue, producing a potentially novel phosphorylation site. Analysis of the different C-SRC RNA transcripts revealed that the level of C-SRC RNA containing both NI and NII exons is similar in adult and fetal brain tissue, whereas the level of C-SRC RNA containing only the NI exon or the nonneuronal form of C-SRC RNAs is significantly higher in fetal brain tissues. These results indicate that the expression and splicing pattern of the C-SRC gene are developmentally regulated in the human brain.