DRAK2 is a serine/threonine kinase of the death associate protein kinase (DAPK) family. Of this family, DRAK2 is most similar to DRAK1, and these two kinases may represent a unique family. DRAK1 and DRAK2 were originally identified using a polymerase chain reaction (PCR) screen to identify additional DAPK members, and was first thought to be primarily involved in promoting apoptosis (Sanjo et al. 1998). While humans have genes for both DRAK1 and DRAK2, mice lack a DRAK1 gene. Although ectopic expression of DRAK2 in cell lines does induce apoptosis (Sanjo et al. 1998; Matsumoto et al. 2001), it is unlikely that apoptotic induction is its key physiologic function since DRAK2-deficient mice demonstrate no obvious defects in apoptotic signaling (McGargill et al. 2004; Friedrich et al. 2005). Instead, DRAK2 has been shown to negatively regulate calcium signaling in primary T cells. Since its catalytic activity is itself induced by calcium influx, DRAK2 may serve to maintain calcium homeostasis (Friedrich et al. 2007; Newton et al. 2011). DRAK2 (and likely its ortholog DRAK1) has been found to be an important immunomodulatory serine/threonine kinase, serving to a) set the initial threshold for thymic and peripheral T cell activation and later, to maintain the survival of effector T cells. In this capacity, mice lacking DRAK2 are resistant to organ-specific autoimmunity (see below). Thus, the development of small molecule antagonists will be of significant value to efforts aimed at combating such immune system diseases.
DRAK2 mRNA expression in humans and mice is fairly limited in adults. DRAK2 is highly enriched in lymphoid tissues including bone marrow, thymus, lymph nodes, and spleen. In particular, T cells and B cells express very high levels of DRAK2 mRNA and protein, and it has been demonstrated to be a critical regulator of T and B cell biology (Gatzka and Walsh 2008). During T cell development in the thymus, DRAK2 expression initiates at the single positive stage and is maintained in the periphery. Stimulation of the T cell receptor (TCR) and co-receptor CD4 leads to acute expression of DRAK2 mRNA and protein in double positive thymocytes (Friedrich et al. 2005). Similar to T cells, B cells also begin to express DRAK2 at the mature B cell stage of development, where it regulates cellular functions in these immune cells also (McGargill et al. 2004; Al-Qahtani et al. 2008).
DRAK2’s Role in Apoptosis
As a member of the DAP kinase family, DRAK2 was thought to also be important for initiation of apoptosis. Following isolation from human placental tissue, studies demonstrated that DRAK2 could, in some instances, induce apoptosis. Initial reports suggested that overexpression of DRAK2 in cells lines such as 3T3 fibroblasts, Cos7 cells, rat NRK cells, and human Caco-2 cells all led to enhanced apoptosis (Sanjo et al. 1998; Kuwahara et al. 2006). The requirements for induction of apoptosis by DRAK2 in these various cell lines is not well understood, although in some instances DRAK2 kinase activity and localization to the nucleus seems to be required (Kuwahara et al. 2006). Besides overexpression studies, DRAK2 siRNA treatment in ACL-15 and NRK cells led to diminished apoptosis following exposure to UV-irradiation, further suggesting a role for DRAK2 in apoptosis in cell lines (Kuwahara et al. 2006).
Although apoptosis can be enhanced and prevented in cell lines overexpressing or lacking DRAK2 respectively, several DRAK2 mouse models have been constructed to further elucidate the role of DRAK2 in cellular processes. First, a DRAK2 knockout mouse was created in which DRAK2 expression was abolished from all tissues. Surprisingly, DRAK2 knockout mice did not have any defects in apoptosis induction or embryonic development (McGargill et al. 2004). In particular, mice did not develop any signs of autoimmunity, cancer, or lymphadenopathy, which would be expected upon deletion of proteins involved in apoptosis. On the contrary, deletion of DRAK2 resulted in survival defects particularly in T cells and B cells. In T cells, DRAK2 has important non-apoptotic functions regulating signals transduced through the T cell receptor, and more detail as to how these processes are regulated will be covered in depth later in this entry.
Besides a DRAK2 knockout mouse, two DRAK2 transgenic mice have been studied. First, DRAK2 expression linked to a human beta actin promoter was constructed, which enhanced expression of DRAK2 roughly fivefold in all adult mouse tissues. As noted before, DRAK2 is not normally expressed in all tissues, and is generally enriched in the immune system. Nonetheless, DRAK2 transgenic mice were shown to have enhanced T cell apoptosis following stimulation, which was dependent on high levels of cytokine exposure. Spleen size was enhanced in these mice, but no other defects were noted (Mao et al. 2006).
A second DRAK2 transgenic mouse was generated in which overexpression occurred only in the T cell compartment. Specifically, DRAK2 was linked to the lck promoter, which begins to be expressed at the double negative stage of thymocyte development. DRAK2 expression was very high in the thymus during development, and subsequently returned to wild-type levels in the periphery. In this mouse, thymocytes did not exhibit enhanced spontaneous or stimuli-induced apoptosis, and negative selection was diminished. Additionally, apoptosis of peripheral T cells was not enhanced following stimulation, although these cells were hypersensitive to stimuli (Gatzka et al. 2009).
Overall, although DRAK2 is a member of the DAP kinase family, it has controversial roles in apoptosis. Cell lines become sensitized to apoptosis when DRAK2 is overexpressed, yet DRAK2 deletion in mice does not confer resistance to apoptosis or manifest as disease pathology. Additionally, overexpression of DRAK2 in the thymus has no effect on apoptosis, while overexpression in the periphery enhanced apoptosis of stimulated T cells. It is important to note that DRAK2 seems to only enhance apoptosis in cells when overexpressed to very high levels, thus we do not expect DRAK2 to have a specific role in apoptosis under normal physiological conditions.
Structure/Function of DRAK2
Within the DAPK family of Ca2+/calmodulin-regulated serine/threonine kinases, DAPK, DRP-1, and Zip Kinase (ZIPK) comprise a highly homologous subfamily, whereas DRAK1 and DRAK2 represent a more distantly related group (Bialik and Kimchi 2006). DRAK2 has been the focus of intense study not only because orthologs for DRAK1 are present only in higher order primates, but because of its unique role in T cell activation. While DRP-1 and ZIPK share 80% and 83% homology, respectively, to the founding member of this family, DAPK, DRAK1 and DRAK2 share only 48% and 51%, respectively. Another important attribute that distinguishes the DRAKs from other members of this family is the complete lack of C-terminal features that command the regulation and apoptosis-promoting function of these kinases. Outside of its kinase domain, DRAK2 lacks homology with all other proteins, containing a short N-terminal region subject to autophosphorylation and a C-terminus important for its subcellular localization and its ability to induce apoptosis upon ectopic expression in various carcinoma cell lines.
As immature thymocytes transit through developmental stages in the thymus to become mature T cells, DRAK2 upregulation occurs and directly affects the degree of calcium mobilization elicited through antigen receptor stimulation. In the mature T cell compartment, where DRAK2 protein levels are highest, loss of DRAK2 leads to substantially enhanced calcium responses (McGargill et al. 2004). Reconstitution of DRAK2-deficient T cells with wild-type DRAK2 restored negative regulation of calcium mobilization, whereas expression of a Ser12Ala DRAK2 mutant was not sufficient to reestablish this negative regulation, indicating the importance of autophosphorylation on Ser12 for DRAK2 biological function (Friedrich et al. 2007). The generation of DRAK2-transgenic mice in which DRAK2 transgene levels are driven specifically within the immature T cell population also lends support to the role of DRAK2 in directly modulating calcium responses. Whereas immature thymocytes ectopically expressing DRAK2 exhibited dampened calcium responses to antigen receptor stimulation, restoration of normal DRAK2 levels within the peripheral T cell compartment in these mice resulted in normal calcium responses (Gatzka et al. 2009). Finally, knockdown of DRAK2 within the clonal T cell line D10 recapitulated the exacerbated calcium response phenotype seen in DRAK2-deficient T cells, arguing against a developmental defect upon loss of DRAK2, and for a role in DRAK2 signaling to regulate this aspect of T cell activation (Newton et al. 2011).
Autophosphorylation on Ser12 is itself elicited by calcium mobilization, implicating DRAK2 in a negative feedback loop whereby calcium influx is necessary for autocatalytic activity on Ser12 and is required for DRAK2 to limit calcium influx, and not ER calcium store release. This process has been shown to be dependent on protein kinase D (PKD), potentially through direct transphosphorylation of DRAK2 by PKD, suggested by in vitro kinase assays. Association of DRAK2 with PKD has been demonstrated in T cells in response to stimuli that activate PKD, and is enriched within mitochondria. Immunofluorescence images of endogenous DRAK2 in primary T cells have revealed DRAK2 punctae formation localized to staining of mitochondria in response to thapsigargin to provoke calcium mobilization, and autophosphorylation on Ser12 was induced directly through generation of mitochondrial reactive oxygen species. Together with data showing that association of PKD and DRAK2 is disrupted by molecules that scavenge reactive oxygen species, activation of DRAK2 by PKD is thought to be dependent on calcium-induced mitochondrial reactive oxygen generation in response to antigen receptor stimulation (Newton et al. 2011).
How DRAK2 down-modulates calcium influx is currently unknown, and few substrates have been identified. In vitro, DRAK2 targets myosin light chain (MLC), a result that was anticipated based on its high level of homology in its kinase domain with MLCK, and based on DAPK’s ability to target MLC in vivo. DRAK2 has been shown to interact with calcineurin homologous protein (CHP) in a manner that negatively regulates its autocatalytic activity and its activity toward MLC (Matsumoto et al. 2001). Although this interaction of CHP was not shown to be dependent on calcium, the suppression of DRAK2 catalytic activity by CHP was. S6K1 has been shown to be targeted by DRAK2 on the same residue (Thr389) as targeted by mTOR, leaving a role for DRAK2 in S6K1 signaling in T cells to be determined (Mao et al. 2009). An incredibly important tool for understanding not only what lies downstream of DRAK2, but for the selective targeting of T cell responses given the unique function of DRAK2 and its role in immune system, will be the development of specific inhibitors to target this kinase. The solved crystal structure with 2.8 Ǻ resolution will undoubtedly aid in the discovery of novel inhibitors that offer the possibility to disrupt a pathway central to the exquisite control of T cell activation and tolerance yet dispensable for key immunological events that maintain resistance toward a multitude of pathological threats.
Role in Immune System
Since DRAK2 is highly expressed in lymphoid tissues, its role in the immune system has been extensively studied. DRAK2 expression is developmentally regulated during thymocyte maturation and its expression is increased following activation of the T cell receptor (TCR) of double positive thymocytes (Friedrich et al. 2005). DRAK2 is involved in setting the threshold for TCR signaling during thymocyte selection as evidenced by the increased calcium flux of DRAK2−/− thymocytes following the double positive stage (Friedrich et al. 2005). To study if DRAK2 plays a role in positive or negative selection, DRAK2-deficient mice have been bred with various TCR transgenic mice (McGargill et al. 2004). DRAK2-deficient mice crossed to OT-II and AND mice have slight increases in CD4 single positive T cells and slight decreases in double positive T cells. Interestingly, there was no effect on CD8 single positive distribution when DRAK2-deficient mice were crossed to OT-I or P14 mice. Although DRAK2-deficiency led to enhanced positive selection, DRAK2 does not seem to play a role in negative selection as DRAK2 deficiency did not affect the loss of self-reactive T cells under the AND or H-Y TCR transgenic backgrounds. In addition, double positive DRAK2−/− T cells from OT-I and OT-II backgrounds had slight increases in activation markers. Therefore, DRAK2 is necessary for proper TCR activation during thymocyte selection.
In peripheral tissues, CD4+ and CD8+ T cells express similar levels of DRAK2 protein. Not surprisingly, DRAK2−/− peripheral T cells also have an increased calcium flux following TCR stimulation (McGargill et al. 2004; Friedrich et al. 2005). In addition, DRAK2−/− T cells have been observed to hyperproliferate to suboptimal stimulation and to a greater rate with weak agonist. The increased calcium flux and hypersensitivity of DRAK2−/− T cells is characteristic of T cells deficient in a negative regulator. In accord with the increase in proliferation, DRAK2−/− T cells produce higher amounts of IL-2. In addition, activated DRAK2−/− T cells produce higher amounts of IFN-γ and IL-4. Suboptimal stimulation results in higher surface expression levels of the costimulatory markers CD25, IL-7R, ICOS, CD27, OX40, and 41BB in DRAK2−/− T cells compared to wild-type T cells. Interestingly, the addition of exogenous anti-CD28 restores the levels of costimulatory markers and the observed hyper-proliferation of activated DRAK2−/− T cells back to wild-type levels (McGargill et al. 2004; Ramos et al. 2008). These studies support the hypothesis that DRAK2 is a negative regulator of T cell activation.
B cells express the highest level of DRAK2 protein and similar to T cells, DRAK2 expression increases following B cell maturation (McGargill et al. 2004; Friedrich et al. 2005). Following immunization of mice with a T-dependent antigen, the loss of DRAK2 in B cells results in up to a fivefold decrease in germinal centers and, consequently, a decrease in high affinity antibodies (Al-Qahtani et al. 2008). DRAK2−/− B cells proliferate similar to wild-type B cells and have no defects in somatic hypermutation and class switch DNA recombination. Further analysis indicates that the defects in DRAK2−/− B cells is a direct consequence of a loss of DRAK2 in T cells since a T-dependent antigen was used to cause the germinal center reaction. To study any B cell intrinsic defects due to the loss of DRAK2, T-independent antigen immunizations should be conducted.
Deletion of negative regulators of T cell activation often leads to increased sensitivity to autoimmune diseases (Pentcheva-Hoang et al. 2009). Based on studies on T cell negative regulators, it was predicted that DRAK2-deficient mice would also be more vulnerable to autoimmune disease. To the contrary, DRAK2-deficient mice were less susceptible to autoimmunity than wild-type mice. Aged DRAK2-deficient mice were not predisposed to spontaneous autoimmunity since there were no differences in the levels of cellular infiltrates in major organs and autoantibodies compared to wild-type (McGargill et al. 2004). DRAK2-deficient mice were also resistant to MOG-induced experimental autoimmune encephalomyelitis (EAE), a mouse model of multiple sclerosis (McGargill et al. 2004, 2008; Ramos et al. 2007, 2008). Overexpression of DRAK2 in mice via the LCK promoter results in spontaneous autoimmunity and increased susceptibility to EAE (Gatzka et al. 2009). However, this is most likely due to ectopically expressing DRAK2 in double positive thymocytes, which may result in altered thymic selection. In addition, DRAK2−/− mice were resistant to type 1-diabetes when bred to the NOD strain of mice that spontaneously develop autoimmune diabetes (McGargill et al. 2008). The resistance is not due to developmental defects in Th17 or antigen-specific effector T cells (Ramos et al. 2008; McGargill et al. 2008). DRAK2−/− mice were susceptible to collagen-induced arthritis and systemic lupus erythematosus, both of which are mediated by autoantibodies (McGargill et al. 2008). DRAK2−/− mice were also susceptible to autoimmune diseases that depend on mast cells and neutrophils (McGargill et al. 2008). Hence, DRAK2−/− mice remain susceptible to autoimmune diseases caused by autoantibodies or cells of the innate immune system but are resistant to autoimmune diseases where pathogenesis is primarily mediated by T cells.
The response to virus has also been studied in DRAK2-deficient mice. DRAK2-deficient mice had antiviral responses to Lymphocytic Choriomeningitis Virus (LCMV) (McGargill et al. 2004) and Murine Hepatitis Virus (MHV) (Ramos et al. 2007) that were indistinguishable from wild-type mice. Interestingly, DRAK2−/− mice were also capable of efficiently eliminating West Nile Virus, but did not succumb to the lethal encephalomyelitis, suggesting that while DRAK2 is not required for antiviral responses, it does promote entry of antiviral T cells into the brain. As stated above, activated DRAK2−/− T cells respond similar to activated wild-type T cells following addition of exogenous costimulation. During a viral infection the amount of costimulation is probably maximal and this may explain why DRAK2-deficient mice mount a normal immune response to viral infection. In addition, studies with MHV showed that DRAK2-deficient mice have enhanced memory T cell function on a per cell basis (Schaumburg et al. 2007). Hence, the absence of DRAK2 does not result in generalized suppression of the immune system and blockade of DRAK2 may be useful in treating T cell–dependent autoimmune diseases and enhance antiviral responses. The role DRAK2 plays in T cell survival likely explains the mechanism behind the restoration of DRAK2−/− T cells to a wild-type phenotype with the addition of costimulation (Ramos et al. 2008).
While the catalytic targets of DRAK2 (and DRAK1) remain to be fully clarified, this serine/threonine kinase offers a unique opportunity to control autoimmunity and potentially cancer. The elucidation of the targets of the kinase, as well as the structural features of the kinase, will be of great value for understanding how DRAK2 controls cellular physiology. While DRAK2 and DRAK1 clearly share significant homology with other members of the DAPK family, ongoing studies should help to determine the functional roles these serine/threonine kinases serve in distinct organ systems. Given the significant autoimmune resistance, but overtly normal antiviral immunity that DRAK2-deficient mice possess, it is likely that small molecule antagonists of the DRAK kinases will be valuable weapons in the arsenal to control autoimmune and other autoinflammatory diseases.
This work was supported by the National Institutes of Health (AI63419); the Arthritis National Research Foundation; the National Multiple Sclerosis Society; and the Juvenile Diabetes Research Foundation. R.H.N. was supported by National Institutes of Health Immunology Research Training Grant T32 AI-060573.
- Gatzka M, Newton RH, Walsh CM. Negative regulation of TCR signaling in immunological tolerance: taming good and evil. Curr Immunol Rev J Immunol. 2009a;183(1):285–97.Google Scholar
- Matsumoto M, Miyake Y, Nagita M, Inoue H, Shitakubo D, Takemoto K, Ohtsuka C, Murakami H, Nakamura N, Kanazawa H. A serine/threonine kinase which causes apoptosis-like cell death interacts with a calcineurin B-like protein capable of binding Na(+)/H(+) exchanger. J Biochem. 2001;130:217–25.PubMedCrossRefGoogle Scholar