Signaling functions of the deubiquitinating enzyme CYLD in lymphocyte activation and osteoclastogenesis

Mammalian immune system is composed of innate and adaptive immune responses, which function cooperatively to combat microbial infections. The innate immune responses are largely mediated by phagocytic cells, including macrophages and neutrophils, whereas the adaptive immune responses rely on antigen-specific lymphocytes, the T and B cells. Defective immune cell function can cause diverse immunodeficiencies, while deregulated immune cell activation leads to chronic inflammation, autoimmunity and cancer. Therefore, activation of immune cells is subject to tight regulation by both positive and negative mechanisms. One increasingly recognized mechanism of immunoregulation is protein ubiquitination, a reversible process that is counter regulated by ubiquitin conjugating enzymes and deubiquitinating enzymes (DUB).;This thesis project concerns a recently identified deubiquitinating enzyme, CYLD. CYLD was originally identified as a tumor suppressor mutated in patients with cylindroma, benign tumors of skin appendages. In vitro studies suggest that CYLD deubiquitinates the signaling molecules TRAF2 and IKKgamma, thereby negatively regulates activation of the transcription factor NF-kappaB by innate immune stimuli, Toll-like receptors (TLRs) and tumor necrosis factor receptors (TNFRs). However, our recent studies using CYLD knockout mice fail to reveal an important role for CYLD in the regulation of TLR and TNFR signaling in innate immune cells. Instead, these genetic analyses reveal a critical role for CYLD in regulating adaptive immune function, particularly the development of T cells in the thymus. Studies to be presented in this thesis have further elucidated pleotropic functions of CYLD in the regulation of peripheral activation of T and B lymphocytes and the prevention of chronic inflammation. Additionally, CYLD also regulate a specific aspect of macrophage function, osteoclastogenesis. The major findings are summarized in the following: (1) CYLD negatively regulates ubiquitin-dependent kinase Tak1 and prevents abnormal T cell responses. Although CYLD deficiency causes impaired thymocyte development therefore reduced peripheral T cell number, T cells isolated from spleen and mesenteric lymph nodes of CYLD knockout mice are hyper-responsive upon TCR and CD28 stimulation. In vivo, CYLD-deficient T cells infiltrate into colon and mediate intestinal inflammation, which resembles human inflammatory bowel disease (IBD). At the molecular level, CYLD targets a ubiquitin-dependent kinase, Tak1, by inhibiting its ubiquitination and auto-activation. In the absence of CYLD, Tak1 is constantly ubiquitinated and activated, leading to activation of downstream kinases, IKK and JNK. These findings emphasize a critical role for CYLD in preventing the spontaneous activation of Tak1 axis of T cell signaling, thereby maintaining normal T cell function. (2) CYLD regulates the peripheral development and naive phenotype maintenance of B cells. CYLD knockout mice develop B cell hyperplasia in secondary lymphoid organs especially in mesenteric lymph nodes, which is dramatically enlarged at 12 week of age. B cells deficient with CYLD spontaneously display surface activation markers and are hyper-responsive to IgM and LPS stimulation both in vitro and in vivo, indicating their abnormal activation status. EMSA results obtained from freshly isolated spleen and mesenteric lymph node B cells reveal hyper activation of NF-kappaB due to loss of CYLD. The constitutive NF-kappaB activation in CYLD knockout B cells is due to constant phosphorylation and degradation of NF-kappaB inhibitory protein IkappaBalpha by its kinase IKKbeta. This finding suggests that CYLD is a crucial negative regulator of IKK activation in B cells. However, unlike in T cells, loss of CYLD results in hyper activation of Erk but not JNK and p38 MAPKs, which indicates different upstream molecular target(s) of CYLD in B cells. (3) CYLD regulates osteoclast function and osteoporosis. Macrophage, one of the major players in innate immunity, is also the precursor of bone-resorbing osteoclasts. Macrophage can differentiate into osteoclast in the presence of both M-CSF and RANKL. In contrast to our previous finding that macrophage activation by innate immune stimuli (LPS and TNF-alpha) is normal in the absence of CYLD, the function of macrophage as osteoclast precursor is regulated by CYLD. CYLD-deficient bone marrow derived macrophages differentiate into osteoclasts in vitro considerably more efficiently than wildtype cells. Moreover, the CYLD knockout mice display an osteoporotic phenotype, characterized by loss of trabecular bone mass. Mechanistically, CYLD physically interacts with and deubiquitinates TRAF6, the key signaling component of RANK signaling, therefore controlling downstream NF-kappaB and AP-1 activation. These findings establish a negative-regulatory role of CYLD in RANK pathway and osteoclast differentiation. In agreement with these functional data, the expression of CYLD protein is potently induced by RANKL but not by LPS or TNF-alpha, thus emphasizing the importance of CYLD protein level in its function in macrophages.