Calcium/calmodulin-dependent kinase IV in immune and inflammatory ...

14 downloads 0 Views 854KB Size Report
Oct 17, 2008 - IV in immune and inflammatory responses: novel routes for an ancient traveller. Luigi Racioppi1,2,3 and Anthony R. Means3. 1 Department of ...
Review

Calcium/calmodulin-dependent kinase IV in immune and inflammatory responses: novel routes for an ancient traveller Luigi Racioppi1,2,3 and Anthony R. Means3 1

Department of Molecular and Cellular Biology and Pathology, University of Naples ‘‘Federico II’’, via Pansini 5, 80131 Naples, Italy Interdipartimental Center for Immunological Science, University of Naples ‘‘Federico II’’, 80131 Naples, Italy 3 Department of Pharmacology and Cancer Biology, Duke University, Durham, NC 27705, USA 2

Ca2+ is a pivotal second messenger controlling the activation of lymphocytes. Crucial events in the social life of immunocytes are regulated by the calcium/calmodulin complex (Ca2+/CaM), which controls the activation status of many enzymes, including the Ca2+/CaM-dependent Ser–Thr kinases (CaMK) I, II and IV. Although CaMKI and CaMKII are expressed ubiquitously, CaMKIV is found predominately in cells of the nervous and immune systems. To be active, CaMKIV requires binding of Ca2+/ CaM and phosphorylation by CaMKKa or b. The requirement of two CaM kinases in the same signalling pathway led to the concept of a CaM kinase cascade. In this review, we focus on the roles of CaMKK and CaMKIV cascades in immune and inflammatory responses. The Ca2+/CaM cascade Calcium is a phylogenetically conserved universal second messenger in all eukaryotic cells [1,2]. >40 years ago, the travels of this ‘ancient’ messenger along cellular biochemical pathways were described as being crucial for the proliferation of T lymphocytes after mitogen stimulation. Calcium regulates many functions by forming a complex with calmodulin (CaM; see Glossary), a 17-kDa acidic protein with four Ca2+ high-affinity binding motifs. Upon the binding of Ca2+, CaM increases its affinity for its targets, which include the Ca2+/CaM-dependent Ser–Thr kinase (CaMK) family [3,4] (Figure 1). Based on their ability to phosphorylate either a very restricted repertoire or a wide range of substrates, CaMKs are known as ‘dedicated’ (e.g. the phosphorylase kinase, CaMKIII, myosin light chain kinase [MLCK]) or ‘multifunctional’, respectively. The CaM-dependent protein kinase I (CaMKI) subfamily (a, b, g and d), CaMKII subfamily (a, b, g and d) and CaMKIV, as well as the CaMKK subfamily (a and b), are multifunctional kinases. The general architecture of the CaMKs includes an N-terminal kinase domain, an autoregulatory domain and an overlapping CaM-binding domain and, in phosphorylase kinase and CaMKII, a Cterminal association domain essential for multimerization and targeting. Bound Ca2+/CaM activates target enzymes by facilitating a conformational change that removes the

autoregulatory domain from the catalytic pocket to enable substrate access, a structural modification resembling that induced by CD45-mediated dephosphorylation of a tyrosine residue (Tyr505) in the regulatory domain of Lck [5]. For full activation, CaMKI and CaMKIV require phosphorylation of a crucial activation loop threonine by CaMKKa or b. The requirement of two CaM kinases in the same signalling pathway is reminiscent of the mitogen activated protein (MAP) kinase cascade and led to the concept of a CaM kinase cascade [4] (Figure 2). Phosphorylation of CaMKIV by CaMKK not only increases CaMKIV kinase activity by relieving intrasteric autoinhibition, but also generates Ca2+/CaM independent or autonomous activity. It is the autonomously active form of CaMKIV that is translocated to the nucleus, where it participates in the regulation of transcription [6] (Figure 3). The CaMKKs are themselves regulated by intrasteric autoinhibition and by Ca2+/CaM binding. Although CaMKKa and b contain residues within their activation loops that can be phosphorylated, no evidence currently exists to indicate that these residues are targets of upstream regulatory kinases. The CaMKK-CaMKIV cascade regulates the activity of several transcription factors, such as cyclic-AMP-responseelement-binding protein (CREB), AP-1, myocyte enhancer factor 2A (MEF2) and retinoid orphan receptor (ROR) Glossary Calmodulin (CaM): a key protein that transduces signals in response to increases in intracellular Ca2+. CaM is a 148–amino acid protein comprised of four helix–loop–helix protein folding motifs called EF hands, which are able to bind Ca2+ ions. Ca2+/CaM dependent kinases (CaMK): a family of structurally related Ser–Thr protein kinases that are activated by CaM binding. These kinases are grouped according to whether they are dedicated kinases having a single substrate (phosphorylase kinase, CaM KIII and MLCK) or whether they are multifunctional (CaM KI, II and IV) and have several substrates. Ca2+/CaM-dependent kinases kinases (CaMKK): a structurally related subfamily of CaMK that have a molecular weight between 54–68 kDa. Two distinct CaMKK isoforms have been identified, namely CaMKKa and CaMKKb. CaMKK can phosphorylate and activate CaMKI and CaMKIV. In addition, CaMKKb can phosphorylate other substrates such as AMPK. CaM kinase cascade: an enzymatic cascade consisting of a CaMKK, which phosphorylates and activates one of two CaM kinases, CaMKI or CaMKIV. Phosphorylation of Ca2+/CaM-bound CaMKIV on its activation loop threonine residue by CaMKK generates CaMKIV kinase activity that is independent of Ca2+/CaM (so-called autonomous activity).

Corresponding author: Racioppi, L. ([email protected])

600

1471-4906/$ – see front matter ß 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.it.2008.08.005 Available online 17 October 2008

Review

Figure 1. Proteins of the CaMK cascade, aligned by functional domain. Colour code: light blue, N- and C-terminal regions; yellow, catalytic domain; green, proline-rich domain; red, autoregulatory domain.

family members, proteins which have pivotal roles in the immune response and inflammation by regulating such processes as T-cell development, cytokine secretion, differentiation of T regulatory (Treg) [7] and Th17 [8–10] cell subsets and signalling through Toll-like receptors (TLR) [11]. CaMKIV and T-cell receptor signalling The expression of CaMKIV has been demonstrated in circulating human lymphocytes, in which the highest expression occurs in CD4+ T cells [12]. T-cell receptor

Trends in Immunology

Vol.29 No.12

(TCR) triggering induces a rapid increase in Ca2+-dependent and autonomous activities of CaMKIV, followed by a rapid return to baseline levels, thereby indicating that the activation–inactivation cycle is brief and tightly regulated [12]. The binding of CaMKIV to the Ser–Thr protein phosphatase 2A (PP2A) in quiescent T cells is important in regulating the temporal kinetics of CaMKIV activation [13]. The Ca2+/CaM-binding autoinhibitory domain of CaMKIV also interacts with PP2A and binding of PP2A and Ca2+/CaM to CaMKIV are mutually exclusive [14]. Thus, in resting T cells in which Ca2+ levels are low, PP2A binds CaMKIV and helps maintain the enzyme in its autoinhibited state. Stimulation of the TCR increases Ca2+, thereby promoting accumulation of the Ca2+/CaM complex, which displaces PP2A from the auto-regulatory domain of CaMKIV. This activates the enzyme and facilitates phosphorylation by a CaMKK, thus, generating autonomous CaMKIV activity. Subsequently, Ca2+ levels decrease, enabling dissociation of Ca2+/CaM and rebinding of PP2A, which dephosphorylates CaMKIV and attenuates its autonomous activity. Thus, upon TCR stimulation, the duration of CaMKIV autonomous activity is dependent on the stoichiometry and subcellular concentration of Ca2+/ CaM and PP2A, parameters that vary based on strength of TCR stimulation, quality and quantity of co-stimulatory signals and differentiation stage of the lymphocytes (Figures 2,3). Full activation of untransformed T cells requires convergence of signals from the TCR and its co-stimulatory molecules, including CD28. In the murine EL4 thymoma cell line, CD3- and CD28-mediated signalling promotes CREB–CREB binding protein (CBP) interaction through a mechanism that involves CaMKIV [15]. In this model,

Figure 2. Schematic representation of the CaMKK–CaMKIV cascade. Receptor stimulation increases intracellular Ca2+ and leads to accumulation of Ca2+/CaM complexes. (a) Binding of Ca2+/CaM to the autoregulatory domains of CaMKK and CaMKIV results in an increase of kinase activity [b (i) and (ii)]. Ca2+/CaM also displaces PP2A bound to the inactive form of CaMKIV. (b)(i) The activated form of CaMKK phosphorylates CaMKIV on T200. (c) This change results in accumulation of phospho-CaMKIV displaying Ca2+/ CaM-independent (autonomous) kinase activity. (d,e) Termination of the cascade is dependent on Ca2+ depletion and rebinding of PP2A to CaMKIV. Colour code: light blue, N- and C-terminal regions; yellow, catalytic domain; green, proline rich domain; red, autoregulatory domain.

601

Review

Trends in Immunology Vol.29 No.12

Figure 3. Ca2+ signalling regulates IL-2 promoter activity in T cells. (a) Owing to the low level of Ca2+ in quiescent T cells, calcineurin is inactive as are proteins of the CaMK cascade. PP2A binds and keeps CaMKIV catalytically inactive in the cytoplasm. In the nucleus, a complex of MEF-2–Cabin1 and HDAC occupies the MEF-2 site, repressing IL2 transcription. (b) The engagement of TCR by MHC–peptide complexes stabilizes the immunological synapse at the antigen presenting cell–T-cell interface. Early signals triggered by this event include a rise in Ca2+, accumulation of Ca2+/CaM, activation of calcineurin (CN), displacement of PP2A from CaMKIV and CD28 and stimulation of the CaMK cascade. As a consequence of these changes, NFAT, pCREB and phospho-CaMKIV (pCaMKIV) migrate to the nucleus (yellow circles indicate phosphorylated residues). As a result of phosphorylation of Cabin1 by CaMKIV, the Cabin1–HDAC complex detaches from MEF-2, relieving repression of the IL-2 promoter. Finally, activation of CREB by CaMKIV and NFAT dephosphorylation by CN mediate AP1 and NFAT occupation of their binding sites on the IL-2 promoter to stimulate transcription.

cross-linking CD28 activates CaMKIV, which thus phosphorylates CREB and recruits CBP. More recently, a mechanism whereby CD28 signalling influences new gene expression during antigen recognition by controlling the co-activator function of p300 and CBP has been confirmed in untransformed murine T-cell clones [16]. In resting T cells, the intracellular tail of CD28 binds PP2A, which, as discussed earlier, can bind and inhibit CaMKIV. Upon TCR engagement, the CD28 tail is phosphorylated by the src kinase Lck, which releases PP2A [17]. Thus, engagement of CD28 theoretically favours sustained, autonomous activation of CaMKIV, which could then be recruited to the immunological synapse. It has been proposed that maximal activation of CaMKIV upon TCR stimulation requires an enzymatic cascade that includes a CaMKK [18]. Direct evidence for this has been provided by cloning human and rat CaMKKb and 602

documenting the ability of this enzyme to phosphorylate CaMKIV on amino acid T200 in Jurkat cells, leading to a marked increase in the enzymatic activity of CaMKIV [19]. In addition, CaMKKb activity is also enhanced by elevation of intracellular Ca2+. Thus, two consecutive steps in a signalling cascade are controlled by Ca2+/CaM. To explain this apparent conundrum, whereby the activity of both upstream and downstream kinases (CaMKK and CaMKIV, respectively) seem to be under the control of the same factor (Ca2+/CaM), it has been proposed that activation of CaMKK might require either higher concentrations or a different pool of Ca2+ than CaMKIV. Such a differential requirement for Ca2+ would enable selective and transient partial activation of CaMKIV by Ca2+/CaM alone under some circumstances and stronger activation through phosphorylation by CaMKK in response to other stimulatory events [18].

Review CaMKIV and T-cell development In vivo and in vitro studies in mouse models show that CaMKIV is developmentally regulated during thymus embryogenesis. Analysis of rat embryos reveals CaMKIV to be present throughout the whole thymus at embryonic day (E)16 and a dramatic increase is observed at E18 when CaMKIV reaches a level of expression comparable to that in pups at postnatal day (P)1 [20]. Similarly, in situ hybridization studies on mouse embryos indicated that by E12.5, CaMKIV mRNA is expressed strongly in the periphery of the thymus primordium. From E13.5, the localization pattern becomes diffuse and CaMKIV mRNA is observed within thymocytes. CaMKIV mRNA expression persists at high levels throughout the rest of thymic development and the signal observed is comparable to that in sections from P1 [21]. CaMKIV is differentially expressed in adult thymocyte subsets. Whereas a progressive accumulation is observed during thymocyte differentiation from DN1 (CD4-CD8-double negative) to DN3, in DP (CD4+CD8+ double positive) cells, CaMKIV levels inversely correlate with the expression of TCRab on the cell surface and progressively declines in the most mature DP cells and SP (CD4+ or CD8+ single positive) thymocytes [20]. A role for CaMKIV in thymocyte development was proposed, based on studies in transgenic mice that expressed a catalytically inactive form of human CaMKIV regulated by the murine proximal lck-promoter, which drives expression of the transgene selectively in DN and DP cells [22]. The expression of the inactive kinase in thymocytes resulted in a marked decrease in thymus size and cellularity, in addition to a significant decrease in the survival rate of isolated thymocytes cultured in vitro. Moreover, transgenic thymocytes fail to secrete IL-2 and upregulate CD25 upon stimulation with ionomycin and phorbol ester, owing to defects in the early signalling pathways that are characterized by markedly decreased CREB phosphorylation and immediate early gene expression. Interestingly, effects on thymus size and cellularity were not reproduced in Camk4 / mice [23]. However, in these animals and in transgenic mice expressing the inactive form of human CaMKIV, a mild decrease in the percentage of SP cells accompanied by a parallel increase in DP thymocytes was observed, indicating a role for CaMKIV in the transition from DP to SP cells. Studies in transgenic mice expressing TCR with defined specificity confirmed that the major defect in Camk4 / thymocytes involves a block in the transition from DP into SP. Notably, by analysing the effects exerted by agonists and altered peptide ligand, further evidence was obtained supporting the hypothesis that CaMKIV signalling contributes to setting the threshold for selection in DP cells. Thus, CaMKIV might be an essential component of the transduction machinery in DP cells undergoing selection. Discrepancy between the effects resulting from genetic ablation of Camk4 and those induced by transgenic expression of inactive CaMKIV might occur as a result of several factors. For example, expression of the inactive enzyme in the early DN stage could sequester proximal enzymes of the CaMK cascade (i.e. CaMKKa or b) and prevent the

Trends in Immunology

Vol.29 No.12

activation of downstream kinases other than CaMKIV (e.g. CaMKI). This mechanism would result in a block in proliferation and survival of DN thymocytes. If this hypothesis is correct, then the phenotype observed in human-CaMKIV transgenic mice might shed light on the global effect exerted by the CaMK cascade on thymocyte development. In this context, the targeted ablation of the Camk4 gene reveals a specific role for the CaMKK–CaMKIV cascade in the mechanism setting the bandwidth for discrimination of signals during the selection of DP thymocytes and, in turn, in shaping the TCR repertoire expressed by mature thymocytes. CaMKIV and the inflammatory response: the osteoimmunology model Several reports have revealed interplay between the skeleton and immune system by identifying a variety of cytokines and transcription factors acting on cells of both systems that conspire to mutually influence them. The emerging field of osteoimmunology attempts to unravel these relationships [24]. Besides the bone marrow cavity in which bone cells and hematopoietic cells exist in proximity, these interactions occur also in inflammatory bone diseases, whereby osteoclasts (OCs), dendritic cells (DCs) and effector T cells accumulate and influence each other. OCs and DCs arise from the monocyte-macrophage lineage and are derived from common hematopoietic progenitors [25]. Whereas DCs are bone-marrow-derived antigen-presenting cells that have pivotal role in the immune response, OCs are important to oversee bone resorption during physiological skeletal remodelling or during the inflammatory process. Recent evidence on the ability of DCs to transdifferentiate into cells displaying a phenotype indistinguishable from OCs unveils an intriguing aspect of the DC–OC interplay [26–28] (Figure 4). Seminal to osteoimmunology is the discovery of the tumour necrosis factor (TNF)-related activation-induced cytokine (TRANCE), also referred to as receptor activator of NF-kB ligand (RANKL). Originally, RANKL was characterized by its role in regulating survival and the ability of DCs to support proliferation of allogenic naı¨ve T cells [29,30]. Subsequently, RANKL was identified as a pivotal factor driving the differentiation of bone-resorbing cells by modulating Ca2+-mediated activation of nuclear factor of activated T cells (NFAT)c1 through a pathway that requires co-stimulatory signalling from the immuno-receptor tyrosine-based activation motif (ITAM)-harbouring adaptors, Fc receptor common g subunit (FcRg) and DNAX-activating protein (DAP12) [31]. Whereas several reports show the relevance of Ca2+dependent calcineurin–NFAT signalling in the RANKL pathway, one in particular has identified CaMKIV to have a crucial role in connecting RANKL–ITAM signals with osteoclastogenesis [32]. Bone-marrow-derived OC precursors express high levels of Camk4 mRNA and accumulate phospho-CaMKIV early after stimulation with RANKL. Camk4 mRNA exists in OC precursors and this level does not change during the terminal differentiation process. In fact, a progressive accumulation of CaMKIV is observed in the more differentiated OCs, indicating that a post-transcriptional mechanism controls CaMKIV expression 603

Review

Trends in Immunology Vol.29 No.12

Figure 4. Schematic representation of the CaMK–CaMKIV cascade and inflammation: the osteoimmunology model. In bone, the inflammatory process involves migration of osteoclast progenitors (OCp), effector T cells and DC progenitors (DCp). Activated T cells release a variety of cytokines, including RANKL and IL-17. RANKL drives differentiation of OCp and IL-17 favours the trans-differentiation of DCp towards the OC lineage. The presence of microbial products, such as LPS, triggers activation and survival programs of OCs and DCs. In addition, stimulation of TLRs induces terminal differentiation of DCp, leading to accumulation of mature DCs in the inflamed bone. As a consequence of these events, OCs derived from OCp and DCp accumulate in the inflamed bone and drive bone resorption. The RANKL–ITAM signalling module triggers the CaMK cascade and regulates differentiation and survival of OCp and OCs through a mechanism that involves CREB and AP1. The CaMK–CaMKIV cascade links TLR4 signalling with the survival program of DC; CREB, BCL-2 and BCL-XL are involved in this process. Yellow circles indicate phosphorylated residues.

during osteoclastogenesis. The targeted ablation of Camk4 results in impairment of OC differentiation and attenuates bone resorption induced by exposure to bacterial lipopolysaccahride (LPS) [32]. Recognition of pathogen derivatives is a specialized function of DCs, which sense microbial products using a variety of pattern-recognition molecules expressed on their surface [33]. Among these molecules, the TLRs bind pathogen-derived molecules to trigger activation of DC, thus, inducing the release of cytokines and driving DC migration to the T-cell zone of secondary lymphoid organs. In addition to these effects, agonists of TLRs, including LPS, control survival of DC and, as recently reported, of OCs as well [34,35]. LPS signals to DCs and OCs via TLR4 require the lipopolysaccharide-binding protein and MD2, a TLR4-associated molecule. Two distinct biochemical pathways are activated by this interaction [36]. The ‘MyD88dependent’ cascade, involving Toll-interleukin-1 receptor 604

domain adaptors MyD88 and Mal, regulates activation of NF-kB and drives the synthesis of cytokines and terminal differentiation. The triggering of the ‘MyD88-independent’ pathway requires TRIF and TRAM (a second set of Tollinterleukin-1 receptor domain adaptors) and stimulates phosphorylation and dimerization of IRF-3, a key event in regulating synthesis of type-I interferon. Several studies support roles for Ca2+ in TLR signalling. TLR4 and TLR2 agonists induce an increase in Ca2+ in cell types including murine peritoneal macrophages, bonemarrow-derived macrophages and bone-marrow-derived DCs [37–39]. The proposed cascade linking proximal signalling triggered by TLR and Ca2+ includes the sequential activation of Src tyrosine kinase and phopholipase C g. Moreover, it has been revealed that the expression of CaMKIV is developmentally regulated in human monocyte-derived DCs [11] with Camk4 mRNA being identified in freshly isolated monocytes. Despite this, and as pre-

Review viously reported, CaMKIV protein is barely detectable in monocytes [12]. However, a progressive accumulation of CaMKIV occurs during the differentiation process of DCs. This change is not accompanied by a parallel increase in mRNA indicating that, similarly to what was observed in OCs, a post-transcriptional mechanism controls accumulation of CaMKIV [32]. In human monocyte-derived DCs, pharmacologic inhibition of CaMKs and ectopic expression of a catalytically inactive form of CaMKIV affects the ability of these cells to survive upon exposure to LPS [11]. Similarly, DCs derived from Camk4 / mice show a marked defect in their ability to survive in response to LPS in vitro. Moreover, DCs expressing a high level of MHC class II (I-A) fail to accumulate in spleens of LPS-treated mice. Overall, these findings reveal a role for CaMKIV in the molecular cascade of events linking TLR signalling to the survival program of activated DC. It should be pointed out that, on the basis of the in vivo results, it is not possible to exclude an additional involvement of this pathway in the mechanisms regulating trafficking of activated DC in secondary lymphoid tissues. The inflammatory response of bone involves activation and recruitment of multiple cell types and the coordinate activation of their functions is required to respond appropriately to microbes and terminate the inflammatory process. Alterations in this process lead to bone disorders such as rheumatoid arthritis, periodontal disease and osteomyelitis. The ability of the CaMKK–CaMKIV cascade to control differentiation and survival of OCs and DCs and to regulate cytokine secretion in memory T cells reveals a novel role for this cascade and paves the way to further understand the molecular basis of the inflammatory process (Figure 4). Transcriptional regulation mediated by CaMKIV Most of the effects of the CaMKK–CaMKIV cascade in cells of the immune system involve modulation of the activity of a restricted number of factors such as CREB, its co-activator CBP and the myocyte enhancer factor-2 (MEF-2) [3]. CREB CREB is a ubiquitous transcription factor that binds the cAMP-response element (CRE) that was named owing to its original description as a target of cAMP-dependent protein kinase (PKA) [40]. The transcriptional activity of CREB is regulated by phosphorylation of S133, which promotes recruitment of the transcriptional co-activator CBP, a protein that displays acetyltransferase activity towards both histone and other transcriptional regulatory proteins. The phospho-CREB–CBP complex then interacts with the basal transcriptional apparatus to initiate RNA synthesis (Figure 3). Subsequent studies showed that CREB can be phosphorylated by a variety of kinases including PKA, RSK, RSK 1–3, MSK, MAPKAPK2 and p38MAPK. Relevant to this review, CaMKI, II and IV can also phosphorylate CREB on S133 [40]. In addition to S133, CaMKII also phosphorylates S142, a modification that inhibits the transcriptional activity of CREB. S301 of CBP has been also identified as a target of CaMKIV phosphorylation both in vitro and in vivo and CaMKs inhibitors attenuated phosphorylation on S301 and

Trends in Immunology

Vol.29 No.12

blocked CBP-dependent transcription [41]. Therefore, a combined effect on CREB and CBP might account for the ability of CaMKIV to regulate CREB transcriptional activity and this dual effect makes it likely that, among the various CaMKs, CaMKIV functions as a physiological inducer of CREB activation in response to increased intracellular Ca2+. Of course, other CaM kinases might also be involved in this process because one of the CaMKII isoforms, CaMKIIdB, is also localized to the nucleus [42]. CREB binding motifs are present in the promoters of several genes controlling development and/or function of T cells [43]. One such function that relies on activation of CREB is IL-2 synthesis, because IL-2 is crucial for the clonal expansion of naı¨ve T cells and for the generation and survival of Treg cells. Maximal induction of the Il-2 gene requires the inducible transcription factor AP-1, comprised of jun and fos immediate-early gene (IEG) [10] family members, in combination with constitutively expressed NFAT. c-fos was one of the first identified gene promoters to be regulated by phospho-CREB and accumulation of the phospho-CREB is required for IEG and IL-2 synthesis in resting T cells. Thus, full activation of the Il-2 promoter requires both NFAT and CREB. CaMKIV has been proposed to control the activity of the IL-2 promoter by a mechanism involving AP1 [44], which indicates that CREB could be a target of CaMKIV. Indeed, CaMKIV does have a role in controlling the activation status of CREB in mature CD4 T cells [22]. Although activated naı¨ve CD4 T cells from Camk4 / mice are able to produce IL-2 to an extent comparable to lymphocytes derived from normal littermates, a marked defect in the synthesis of IL-2, IL-4 and IFN-g was observed in CD4 memory T cells. The failure of memory cells to produce cytokines seems secondary to the inability to phosphorylate CREB and induce the IEG required for cytokine gene induction in response to TCR stimulation. Thus, although in naı¨ve T cells and thymocytes CaMKIV seems dispensable for phosphorylation of CREB and induction of IL-2 secretion, this kinase does have a pivotal role in the activation of the CD4 memory subset of T cells. A common mechanism including CaMKIV-dependent phosphorylation of CREB is involved in the regulation of differentiation and survival of monocyte-derived cells [11,32]. Activation of CaMKIV signalling by RANKL requires ITAM-mediated co-stimulatory signals, promotes phosphorylation of CREB on S133 and induces the transcription of c-fos, which is crucial for regulating the transcriptional activity of NFATc1. Thus, as hypothesized for TCR–CD28 signalling in T cells, OC stimulatory and costimulatory signals triggered by the RANKL–ITAM module also converge on the CaMKIV–CREB pathway [15,32]. DCs isolated from Camk4 / mice fail to survive upon exposure to LPS, a phenomenon that seems to depend on the inability of TLR4 signalling to prevent the temporal decline in the survival factors B-cell lymphoma-2 (Bcl-2) and basal cell lymphoma-extra large (Bcl-xL) [11]. Remarkably, this is associated with failure of Camk4 / DCs to accumulate phospho-CREB (pCREB) in response to LPS. Thus, in T lymphocytes and monocyte-derived cells, the CaMKK–CaMKIV–CREB cascade is an essential com605

Review ponent of the network linking signals from receptors (TCR, TLR4 and RANK), playing a crucial part in the induction of the immune response, with activation programs governing differentiation, survival and cytokine release in OCs, DCs and CD4 memory T cells. MEF2 The MEF2 proteins have DNA-binding domains located at their N termini, which are characterized by the presence of a MADS (Mini chromosome maintenance-1, Agamous, Deficiency, Serum response factor) box motif [45]. MEF2 sites are present in promoters of several muscle-specific genes and MEF2 was originally identified as a major regulator of myogenesis. MEF2 sites are also present in the promoters of several other genes including the orphan nuclear receptor Nur77 and IL-2. A molecular link between MEF2 and the apoptotic program in T cells has been proposed [46]. In thymocytes, Nur77 is transcribed in a Ca2+-dependent manner in response to TCR stimulation as part of a process that induces apoptosis. The Nur77 promoter contains two Ca2+-regulated DNA elements controlled by MEF2, which is constitutively bound to its cognate DNA-binding elements in the nucleus independently of the intracellular Ca2+ concentration [47]. In quiescent thymocytes, MEF2 recruits a family of functionally redundant transcriptional repressors including Cabin1 and histone deacetylase (HDAC)-4, -5 and -7. These MEF2-specific co-repressors function as histone deacetylases to remodel chromatin structure and silence promoter activity. Upon Ca2+ influx, these MEF2 repressors are removed from MEF2 by Ca2+/ CaM, enabling MEF2 to bind to co-activators such as p300 and to thereby turn on transcription of target genes [46]. It has been proposed that the convergence of calcineurin and CaMKIV signalling to MEF2 is required to fully activate the Nur77 promoter in T cells [48] (Figure 3). MEF2 regulates transcription of the IL2 gene in Jurkat cells and a murine T cell hybridoma [49]. A MEF2-binding site exists in the IL2 promoter close to the TATA box and, similar to that observed in the Nur77 promoter, this element is occupied by MEF2 in resting T cells, which functions to mediate recruitment of Cabin1 and HDAC4. Removing the MEF2 site from the IL-2 promoter or reducing MEF2 expression by siRNA results in a marked decrease of IL-2 transcription. Thus, MEF2 integrates two independent Ca2+-signaling modules namely, CaM– calcineurin–NFAT and CaM–Cabin1–HDAC (Figure 3). CaMKIV regulates the transcriptional activity of MEF2 on the IL-2 promoter by relieving Cabin1-mediated inhibition of MEF2-mediated transcription. It achieves this by directly phosphorylating the C-terminal region of Cabin1 to generate a docking site for 14–3–3, an adaptor protein involved in the nuclear export process. The binding of 14– 3–3 to phosphorylated Cabin 1 results in the export of the Cabin1 from the nucleus, which enables MEF2 to exert its transcriptional activation function [50]. Concluding remarks and future perspectives Studies on the CAMKK–CaMKIV cascade in the immune system show this pathway to be relevant to the molecular mechanisms that control crucial processes in the immune 606

Trends in Immunology Vol.29 No.12

response. In the thymus, CaMKIV regulates the selection threshold of DP thymocytes. In CD4 memory T cells, this cascade is required to link TCR signalling to production of IL-2, IFN-g and IL-4 through regulating transcriptional activity of CREB and MEF-2 (Figure 3). Finally, the CaMKK–CaMKIV cascade is important for regulating the differentiation and survival programs of OCs and DCs (Figure 4). These latter findings indicate a more general role for this cascade in the inflammatory response. Future studies are required to fully understand the molecular mechanisms underlying the effects of CaMKIV in cells of the immune system. Adaptors that link proximal components of TCR, TLR and RANKL–ITAM modules with the CaMK–CaMKIV cascade need to be identified. A promising field of investigation is examination of the ability of the CaMK–CaMKIV cascade to intercept components of cytokine receptor signalling. In this context, novel perspectives have been revealed by recent data showing that CBP, a putative target of CaMKIV [3], controls signal transduction initiated by the type-I interferon receptor by acetylating the IFN-aR2 chain, IRF-9, STAT-1 and STAT2, all of which are components of the interferon-stimulated gene factor 3 complex. These results elucidate a novel role for CBP and indicate the possibility that the CaMKK– CaMKIV cascade might be involved in regulating posttranscriptional control of cytokine and ITAM-dependent receptor signalling [51]. Finally, the involvement of transcription factors regulated by CaMKIV (e.g. CREB, RORg and RORg [52]) in crucial steps of the immune response, such as regulation of the transcriptional activity of the Foxp3 promoter in Treg cells [7] and the differentiation program of Th17 cells [8–10], offers intriguing perspectives on the involvement of the CaMKK–CaMKIV cascade in mechanisms controlling initiation, amplification and termination of the immune response and inflammatory process. Acknowledgements The authors thank all members of the Means laboratory and Anna Maria Masci for helpful discussion. This work was supported by grants from the Italian National Program for AIDS research (40F.66), PRIN (2006051402) and NIH (DK074701).

References 1 Berridge, M. et al. (1999) Calcium signalling. Curr. Biol. 9, R157–R159 2 Berridge, M.J. et al. (2000) The versatility and universality of calcium signalling. Nat. Rev. Mol. Cell Biol. 1, 11–21 3 Hook, S.S. and Means, A.R. (2001) Ca2+/CaM-dependent kinases: from activation to function. Annu. Rev. Pharmacol. Toxicol. 41, 471–505 4 Soderling, T.R. (1999) The Ca-calmodulin-dependent protein kinase cascade. Trends Biochem. Sci. 24, 232–236 5 Yamaguchi, H. and Hendrickson, W.A. (1996) Structural basis for activation of human lymphocyte kinase Lck upon tyrosine phosphorylation. Nature 384, 484–489 6 Chow, F.A. et al. (2005) The autonomous activity of calcium/ calmodulin-dependent protein kinase IV is required for its role in transcription. J. Biol. Chem. 280, 20530–20538 7 Kim, H.P. and Leonard, W.J. (2007) CREB/ATF-dependent T cell receptor-induced FoxP3 gene expression: a role for DNA methylation. J. Exp. Med. 204, 1543–1551 8 Ivanov, I.I. et al. (2006) The orphan nuclear receptor RORgat directs the differentiation program of proinflammatory IL-17+ T helper cells. Cell 126, 1121–1133 9 Yang, X.O. et al. (2008) T helper 17 lineage differentiation is programmed by orphan nuclear receptors ROR a and ROR g. Immunity 28, 29–39

Review 10 Du, J. et al. (2008) Isoform-specific inhibition of ROR a-mediated transcriptional activation by Human FOXP3. J. Immunol. 180, 4785–4792 11 Illario, M. et al. (2008) Calmodulin-dependent kinase IV links Toll-like receptor 4 signaling with survival pathway of activated dendritic cells. Blood 111, 723–731 12 Hanissian, S.H. et al. (1993) Expression of a Ca2+/calmodulindependent protein kinase, CaM kinase-Gr, in human T lymphocytes. Regulation of kinase activity by T cell receptor signaling. J. Biol. Chem. 268, 20055–20063 13 Westphal, R.S. et al. (1998) A signaling complex of Ca2+-calmodulindependent protein kinase IV and protein phosphatase 2A. Science. 280, 1258–1261 14 Anderson, K.A. et al. (2004) Regulation and function of the calcium/ calmodulin-dependent protein kinase IV/protein serine/threonine phosphatase 2A signaling complex. J. Biol. Chem. 279, 31708–31716 15 Yu, C.T. et al. (2001) Multiple signals required for cyclic AMPresponsive element binding protein (CREB) binding protein interaction induced by CD3/CD28 costimulation. J. Immunol. 166, 284–292 16 Nandiwada, S.L. et al. (2006) p300/Cyclic AMP-responsive element binding-binding protein mediates transcriptional coactivation by the CD28 T cell costimulatory receptor. J. Immunol. 177, 401–413 17 Chuang, E. et al. (2000) The CD28 and CTLA-4 receptors associate with the serine/threonine phosphatase PP2A. Immunity 13, 313–322 18 Park, I.K. and Soderling, T.R. (1995) Activation of Ca2+/calmodulindependent protein kinase (CaM-kinase) IV by CaM-kinase kinase in Jurkat T lymphocytes. J. Biol. Chem. 270, 30464–30469 19 Anderson, K.A. and Kane, C.D. (1998) Ca2+/calmodulin-dependent protein kinase IV and calcium signaling. Biometals 11, 331–343 20 Krebs, J. et al. (1997) Calmodulin-dependent protein kinase IV during T-cell development. Biochem. Biophys. Res. Commun. 241, 383–389 21 Wang, S.L. et al. (2001) Expression of Ca2+/calmodulin-dependent protein kinase IV (caMKIV) messenger RNA during murine embryogenesis. Cell Growth Differ. 12, 351–361 22 Anderson, K.A. and Means, A.R. (2002) Defective signaling in a subpopulation of CD4+ T cells in the absence of Ca2+/calmodulindependent protein kinase IV. Mol. Cell. Biol. 22, 23–29 23 Raman, V. et al. (2001) Requirement for Ca2+/calmodulin-dependent kinase type IV/Gr in setting the thymocyte selection threshold. J. Immunol. 167, 6270–6278 24 Walsh, M.C. et al. (2006) Osteoimmunology: interplay between the immune system and bone metabolism. Annu. Rev. Immunol. 24, 33–63 25 Miyamoto, T. et al. (2001) Bifurcation of osteoclasts and dendritic cells from common progenitors. Blood 98, 2544–2554 26 Rivollier, A. et al. (2004) Immature dendritic cell transdifferentiation into osteoclasts: a novel pathway sustained by the rheumatoid arthritis microenvironment. Blood 104, 4029–4037 27 Speziani, C. et al. (2007) Murine dendritic cell transdifferentiation into osteoclasts is differentially regulated by innate and adaptive cytokines. Eur. J. Immunol. 37, 747–757 28 Coury, F. et al. (2008) Langerhans cell histiocytosis reveals a new IL17A-dependent pathway of dendritic cell fusion. Nat. Med. 14, 81–87 29 Wong, B.R. et al. (1997) TRANCE (tumor necrosis factor [TNF]-related activation-induced cytokine), a new TNF family member predominantly expressed in T cells, is a dendritic cell-specific survival factor. J. Exp. Med. 186, 2075–2080 30 Anderson, D.M. et al. (1997) A homologue of the TNF receptor and its ligand enhance T-cell growth and dendritic-cell function. Nature 390, 175–179 31 Takayanagi, H. (2007) The role of NFAT in osteoclast formation. Ann. N. Y. Acad. Sci. 1116, 227–237

Trends in Immunology

Vol.29 No.12

32 Sato, K. et al. (2006) Regulation of osteoclast differentiation and function by the CaMK-CREB pathway. Nat. Med. 12, 1410– 1416 33 Steinman, R.M. and Hemmi, H. (2006) Dendritic cells: translating innate to adaptive immunity. Curr. Top. Microbiol. Immunol. 311, 17–58 34 Hou, W.S. and Van Parijs, L. (2004) A Bcl-2-dependent molecular timer regulates the lifespan and immunogenicity of dendritic cells. Nat. Immunol. 5, 583–589 35 Itoh, K. et al. (2003) Lipopolysaccharide promotes the survival of osteoclasts via Toll-like receptor 4, but cytokine production of osteoclasts in response to lipopolysaccharide is different from that of macrophages. J. Immunol. 170, 3688–3695 36 Kawai, T. and Akira, S. (2007) TLR signaling. Semin. Immunol. 19, 24–32 37 Chun, J. and Prince, A. (2006) Activation of Ca2+-dependent signaling by TLR2. J. Immunol. 177, 1330–1337 38 Aki, D. et al. (2008) Peptidoglycan and lipopolysaccharide activate PLCg2, leading to enhanced cytokine production in macrophages and dendritic cells. Genes Cells 13, 199–208 39 Zhou, X. et al. (2006) Ca2+- and protein kinase C-dependent signaling pathway for nuclear factor-kB activation, inducible nitric-oxide synthase expression, and tumor necrosis factor-a production in lipopolysaccharide-stimulated rat peritoneal macrophages. J. Biol. Chem. 281, 31337–31347 40 Shaywitz, A.J. and Greenberg, M.E. (1999) CREB: a stimulus-induced transcription factor activated by a diverse array of extracellular signals. Annu. Rev. Biochem. 68, 821–861 41 Impey, S. et al. (2002) Phosphorylation of CBP mediates transcriptional activation by neural activity and CaM kinase IV. Neuron 34, 235–244 42 Takeuchi, Y. et al. (1999) Nuclear localization of the d subunit of Ca2+/ calmodulin-dependent protein kinase II in rat cerebellar granule cells. J. Neurochem. 72, 815–825 43 Kuo, C.T. and Leiden, J.M. (1999) Transcriptional regulation of T lymphocyte development and function. Annu. Rev. Immunol. 17, 149–187 44 Ho, N. et al. (1996) Activation protein 1-dependent transcriptional activation of interleukin 2 gene by Ca2+/calmodulin kinase type IV/ Gr. J. Exp. Med. 184, 101–112 45 Black, B.L. and Olson, E.N. (1998) Transcriptional control of muscle development by myocyte enhancer factor-2 (MEF2) proteins. Annu. Rev. Cell Dev. Biol. 14, 167–196 46 Youn, H.D. et al. (1999) Apoptosis of T cells mediated by Ca2+-induced release of the transcription factor MEF2. Science. 286, 790–793 47 Hsu, H.C. et al. (2004) Nur77 family of nuclear hormone receptors. Curr. Drug Targets Inflamm Allergy 3, 413–423 48 Blaeser, F. et al. (2000) Ca2+-dependent gene expression mediated by MEF2 transcription factors. J. Biol. Chem. 275, 197–209 49 Pan, F. et al. (2004) Myocyte enhancer factor 2 mediates calciumdependent transcription of the interleukin-2 gene in T lymphocytes: a calcium signaling module that is distinct from but collaborates with the nuclear factor of activated T cells (NFAT). J. Biol. Chem. 279, 14477–14480 50 Pan, F. et al. (2005) Calmodulin-dependent protein kinase IV regulates nuclear export of Cabin1 during T-cell activation. EMBO J. 24, 2104– 2113 51 Tang, X. et al. (2007) Acetylation-dependent signal transduction for type I interferon receptor. Cell 131, 93–105 52 Kane, C.D. and Means, A.R. (2000) Activation of orphan receptormediated transcription by Ca2+/calmodulin-dependent protein kinase IV. EMBO J. 19, 691–701

607