 |
Cellular Signaling by the Cytokine Receptor Superfamily
Summary: James Ihle's research is focused on identifying the biochemical events in the response to cytokines and establishing their role in cellular responses.
Cytokines regulate a variety of physiological functions through interactions with the structurally and functionally conserved receptors of the cytokine receptor superfamily. Members of this family include receptors that regulate many of the stages of lymphoid and myeloid development. Our long-term goals have been to identify the biochemical events initiated by cytokine binding and establish their individual roles in the physiological responses. To assess function, we rely primarily on established murine genetic models.
All cytokine receptors associate with and activate members of the Jak family of tyrosine kinases. For example, Jak2-deficient mice demonstrated that Jak2 plays an essential, nonredundant role in erythropoietin (Epo) receptor signaling. Jak2 interacts with the receptor through its amino-terminal region, and recent studies have identified an amino-terminal tyrosine of Jak2, which, when phosphorylated, results in the dissociation of Jak2 from the receptor complex and termination of signal transduction. Many studies have indicated that Jak2 phosphorylation of receptor tyrosines is critical to recruit substrates to the receptor. To assess this, we derived two strains of mice with genetically modified receptors. These mice demonstrated that the distal half of the receptor is not required for any physiological responses to Epo and that a truncated receptor, devoid of tyrosine, functions nearly as well as the wild-type receptor. In another approach, we have done extensive scanning mutational analysis of the essential region of the Epo receptor. In more than 40 mutants, the mutations that significantly affect receptor signaling also affect Jak2 association; thus the primary, if not sole, function of the receptor is to activate Jak2.
With activation, at least 10 Jak2 tyrosines are phosphorylated. We are evaluating the roles of individual phosphorylation sites by identifying proteins that can associate with them. This approach led to the identification of phospholipase Cγ2 (PLCγ) and p110δ as potentially important signaling proteins. Mice lacking these genes have normal hematopoiesis, however. An essential, nonredundant role for PLCγ2 in signal transduction was demonstrated, however, through a variety of receptors of the Fc receptor superfamily, including the B cell antigen-specific receptor. The studies also demonstrated a more restricted role, primarily in B cell receptor (BCR) signaling, for p110δ. The approach also identified a novel 70-kDa protein, and we subsequently identified the highly related protein p70b. Both proteins contain SH3 domains, a domain related to the catalytic center of bisphosphoglycerate mutase and a ubiquitin-associated domain. Although a deficiency of either gene in mice had no consequences, a deficiency of both resulted in hyperactivation of T cell receptor signaling. Based on the functions identified, we now term the proteins Sts-1 and Sts-2 (suppressors of T cell signaling 1 and 2). The biochemical basis for increased TCR signaling is the more rapid and higher level of activation of Zap70, a critical tyrosine kinase for signaling. The results are consistent with a model in which Sts-1 and Sts-2 sequester Zap70 away from the TCR complex and thereby limit the amount available for activation. In a mouse model of the human disease of multiple sclerosis, the increased TCR signaling led to an increased susceptibility to autoimmunity.
The role that Jak2 activation has in controlling Epo-associated expansion of erythroid lineage cells is largely unknown. One critical element of this regulation is the induction of Bcl-X expression, as shown by the severe consequences on fetal erythropoiesis of deleting the Bcl-X gene. The induction of Bcl-X does not depend on known pathways including the Stat5 transcription factors. Consequently, we are attempting to identify the important Jak2-dependent pathway for Bcl-X regulation. We have also screened a retroviral library for genes that could rescue Epo-induced erythroid colony formation of Jak2-deficient cells. Four genes have been identified. Two genes, Hax-1 and Ars-2, were initially identified by their ability to suppress apoptosis in various settings. This result and the ability of Bcl-X to rescue colony formation support the concept that the primary function of cytokine signaling is to provide survival factors and that cell cycle progression and subsequent terminal differentiation do not depend on cytokine signaling. The other two genes, Zan75 and a previously unidentified gene we term clone 67, have not been functionally defined. The expression of all four genes is increased in response to cytokines. A deficiency of Ars-2 in mice is associated with an early embryonic lethal phenotype; consequently, lineage-specific deletions are now being done. Deficiency of clone 67 results in no alteration in phenotypes. Zan75-deficient mice have just been generated. The most striking deficiency is that of Hax-1. This protein contains a BH1 and BH2 domain, placing it in the Bcl family of survival factors. Like other members, it is a mitochondrial membrane protein that, when overexpressed, can protect against induction of apoptosis. Deletion of Hax-1 results in two phenotypes. First, T and B cells are lost between 2 and 3 months of age; in this regard the phenotype is similar to that seen with deletion of Bcl-2. The second phenotypic change is a progressive loss of motor coordination, with death ultimately caused by neurological degeneration that is associated with marked apoptosis in the region of the brain responsible for initiation of motor neuron activity.
Cytokine receptors that utilize Jak2 activate the related transcription factors Stat5a and Stat5b. One unexpected function of Stat5a/b is its requirement for the response of peripheral T cells to interleukin-2. We are focusing on the role that Stat5a and Stat5b play in peripheral T cells. Various approaches have resulted in the identification of several interesting Stat5-dependent genes in T cells. To assess their individual functions, we have derived strains of mice deficient in the genes. Deletion of a novel gene, the nine-membrane-spanning protein (NMSP), results in an early embryonic lethal phenotype and thus requires the derivation of conditional mutants. Deletion of GADD45γ results in no phenotypic changes, however. Similarly, when members of a novel family of nuclear proteins—which we have termed SG1, SG2, and SG3—are deleted, there are no phenotypic changes. Deletion of the highly Stat5a/b-induced genes CIS and oncostatin M (OSM) also resulted in no phenotypic changes. We continue to explore the function of additional Stat5a/b-induced genes through the derivation of mutant strains of mice.
Cytokine signaling is negatively regulated by members of the suppressors of cytokine signaling (SOCS) family of proteins. SOCS1 is highly expressed at the thymic stage of T cell development. We demonstrated that the deletion of SOCS1 results in a striking perinatal phenotype that depends on the generation of T cells and their activation and production of interferon-γ. Through the use of a variety of genetic approaches, we are defining the individual components that contribute to the lethality. In particular, initial evidence supports our hypothesis that the normal function of SOCS1 is to block cytokine signaling during T cell selection. When cytokine signaling is absent, autoreactive cells are generated in numbers consistent with the large numbers of peripheral T cells, with an activated phenotype and producing cytokines in SOCS1-deficient mice. The studies also demonstrate that this may only be the initiating event and that altered responses to IFN-γ in a variety of cells contribute to the lethality. We also explored SOCS3 function by deriving deficient mice and described an embryonic lethal phenotype. Subsequently we demonstrated that this is due to a placental insufficiency. Specifically, SOCS3 plays an essential role in suppressing signaling through the leukemia inhibitory factor (LIF) receptor and thereby controls the number of placental giant cells that are generated from trophoblast stem cells. In the absence of SOCS3, too many giant cells are generated; a deficiency in the LIF receptor results in too few giant cells. The studies provide a paradigm for the importance of balancing cytokine signaling.
We continue to seek ways to identify additional new components in cytokine signaling. Essential in our studies has been the establishment of appropriate mouse mutant models for the analysis of physiological roles of genes. Such models have been critical in identifying essential nonredundant functions for the genes—both for cytokine signaling and for roles in signaling through other receptor systems.
Last updated August 31, 2004
|
|
|
|
