Zap70 Signaling Pathway Mediates Glucocorticoid Receptor-Dependent Transcriptional Activation: Role in the Regulation of Annexin 1 Expression in T Cells¹ Free

Engagement of TCR during the adaptive immune response leads to the activation of Src family of protein tyrosine kinases Lck and/or Fyn. These enzymes phosphorylate ITAM of the TCR and create docking sites for tandem Src homology 2-domains of the Zap70 tyrosine kinase. Zap70 binds to two phosphotyrosine residues of individual ITAM and is subsequently phosphorylated and activated by Lck and/or Fyn (1, 2, 3, 4, 5, 6). These signaling events allow recruitment and activation of a number of other signaling proteins at the TCR followed by the activation of downstream signaling pathways. Targets of Zap70 include linker for activation of T cells, the adaptor protein SLP-76, and Vav, an exchange factor for small GTP-binding proteins. Zap70 plays a critical role in the normal T cell signaling and positive and negative selection of thymocytes. Zap70 knockout mice and patients with mutations in Zap70 have profound severe combined immunodeficiency syndrome phenotypes and other abnormalities (7, 8, 9). Recently, altered thymic T cell selection due to a mutation of the Zap70 gene, has been found to cause autoimmune arthritis in mice (10).

Glucocorticoid steroids are potent anti-inflammatory agents that exhibit their action by binding to glucocorticoid receptor (GR),³ a member of nuclear receptor superfamily of transcription factors. GR modulates the transcription by binding to a glucocorticoid response element (GRE) found in the promoters of a number of target genes or through protein-protein interaction with a number of other transcription factors and cofactors (11, 12, 13). T cells have been shown to be critical cellular targets of GR signaling as T cell-specific GR knockout mice show significant mortality following immune activation (14).

Annexin (ANXA)1 is a 37-kDa anti-inflammatory protein, induced and released by the glucocorticoid action, that has been implicated in cell growth and differentiation, membrane trafficking, mobilization of leukocytes, ischemic damage, fever, pain, and sepsis (15, 16, 17, 18). ANXA1 null mice exhibit elevation of cytokines and plasma markers of organ injury indicating a critical protective role of this protein in inflammation and sepsis (16, 19, 20). ANXA1 has also been identified as an endogenous ligand involved in the engulfment of apoptotic T cells (21).

We have recently reported that T cells defective in Zap70 signaling show loss of retinoid receptor dependent transcription (22). In this study we show that Zap70 signaling is also essential in dexamethasone (DEX)-inducible glucocorticoid receptor GR-mediated transactivation in T lymphocytes. We also provide evidence identifying Zap70 and its upstream kinase Lck as essential enzymes in maintaining normal levels of ANXA1, a protein known to be induced and released by the glucocorticoid action. Zap70 and Lck were also found to regulate ANXA1 promoter function in part through the modulation of GR-dependent transcription.

T lymphocyte leukemia Jurkat cell line (clone E6-1) and Lck-deficient Jurkat cell line JCaM1.6 were obtained from American Type Culture Collection. A. Weiss (University of California, San Francisco, CA) provided JCaM1.6 cells stably transfected with Lck (JCaM1.6/Lck) or vector (JCaM1.6/Vector). Zap70-negative Jurkat cell line P116, P116 cells stably expressing c-myc-tagged wild-type (WT) Zap70 (P116/pWT), and P116 cells stably expressing c-myc-tagged kinase-dead Zap70 mutant (K369R) (P116/pDK) were provided by R. L. Wange (National Institutes of Health, Baltimore, MD). These cells were maintained in RPMI 1640 medium (BioWhittaker) supplemented with 10 mM HEPES buffer, 2 mM l-glutamine, 60 μg/ml gentamicin, and 10% FBS (HyClone Laboratories). Human PBMC, obtained by lymphapheresis of healthy donors, were purified by Ficoll density gradient centrifugation. Purified PBMC were treated with PHA and IL-2 for 2 days in AIM-V medium (Invitrogen Life Technologies) supplemented with 10% FBS. The cells were washed to remove PHA and maintained in IL-2 as described (22). In this study, these cells will be referred to as proliferating peripheral blood T (PBT) cells. PBT cells were 98% CD3-positive as monitored by flow cytometry. ANXA1 Ab was from Santa Cruz Biotechnology. Abs to Zap70, Lck, ANXAVI, and ANXAVII were from BD Biosciences. Ab to GAPDH and β-actin were from Abcam. DEX and trichostatin A (TSA) were from Sigma-Aldrich.

Samples were hybridized on the human U133A chips from Affymetrix using their recommended protocol. The final gene list was selected using one-way ANOVA analysis with the value p < 0.05, fold change >2, and mean difference >30 between compared groups.

Total RNA was isolated using Absolutely RNA Miniprep kit from Stratagene. Total RNA (250 ng) was reverse transcribed with 200 U of Superscript II RNase H- reverse transcriptase (Invitrogen Life Technologies) in the presence of 2.5 μM random hexamers. The expression levels of ANXA1 and endogenous control gene GAPDH were measured by PCR. PCR was performed in the presence of [α-³²P]dCTP and the products were electrophoresed in 6% precast polyacrylamide gels (Invitrogen Life Technologies). The gels were dried and scanned for quantitation of the PCR products using a bio-imaging analyzer (Bas 1000; Fuji). The analysis of the PCR products was done when the reactions were in the linear range and the amount of product was directly proportional to the amount of input cDNA. In some of the experiments the products of reverse transcription were also subject to real-time PCR using 7900HT Fast Real-Time PCR System and TaqMan gene expression assay kits from Applied Biosystems. The comparative threshold cycle method was used to calculate the relative gene expression.

Protein extracts were electrophoresed in a 10% NuPAGE Bis Tris Gel using NuPAGE MES-SDS running buffer (Invitrogen Life Technologies), and transferred to a polyvinylidene difluoride membrane using XCell Blot Module (Invitrogen Life Technologies). The protein was detected using fluorophore-labeled secondary Abs and Odyssey Infrared Imaging System (LI-COR Biotechnology).

DEX-mediated activation of GR was studied by transient transfection using GRE-TK-Luc plasmid, which was a gift from Dr. K. Ozato (National Institutes of Health, Bethesda, MD). This plasmid contained two copies of GRE (TGTACAGGATGTTCT) from glucocorticoid-responsive tyrosine aminotransferase gene (23). ANXA1 promoter-driven transcriptional activation was studied using pGL3/ANXA1-Luc. This plasmid was constructed by cloning PCR amplified human ANXA1 promoter fragment (containing −987 to +72 sequences from the transcriptional start site of human ANXA1 promoter, Accession no. U25414) into KpnI and BglII sites of pGL3-Luc basic vector (Promega). pGL3/ANXA1/GRE-Luc was constructed by cloning PCR amplified human ANXA1 promoter fragment (containing −987 to +258 sequences from the transcriptional start site of human ANXA1 promoter). There is a GRE starting at +237 bp (in the first intron region) in this sequence (24). This plasmid also contained a splice acceptor site CCTGAAA introduced at the 5′ end of the reverse primer used for PCR amplification of the ANXA1 promoter. The acceptor site was introduced for proper splicing of ANXA1 intron from the RNA transcripts. GR expressing plasmid pcDNA4/TOPO/GR was constructed by cloning PCR amplified human GR in pcDNA4/HisMaxTOPO vector (Invitrogen Life Technologies). pGL3/GAPDH-Luc plasmid containing human GAPDH promoter was used as control to normalize the luciferase activity. This plasmid was constructed by cloning an 1137 bp PCR-generated GAPDH promoter fragment (25) into KpnI and BglII sites of pGL3-Luc basic vector. Zap70 and Lck SMARTpool short interfering RNAs (siRNAs) were purchased from Upstate Biotechnology. Jurkat and Jurkat-derived cells were transfected by electroporation using a Gene Pulser II (Bio-Rad) at 0.250 kV and 975 μF as described (26). PBT cells were electroporated using a T cell Nucleofection kit from Amaxa Biosystems, according to the manufacturer’s instructions as described (22). Luciferase activity was normalized to protein concentrations in the extracts. Data from the experiments represent the mean of three independent experiments with SE calculated for each value.

The results obtained from three independent experiments were expressed as mean ± SD. Statistical analysis was assessed by Student’s t test. A value of p < 0.05 was considered significant.

We recently reported that T lymphocytes lacking Zap70 tyrosine kinase show loss of retinoid receptor-dependent transcription, whereas signaling mediated by vitamin D and thyroid hormone receptors is not affected (22). To extend our studies of Zap70 signaling to other members of the nuclear receptor superfamily, we investigated the role of Zap70 signaling pathway in the modulation of GR-mediated transcription in Jurkat T cells. When WT, P116 (Zap70-negative), P116/pWT (P116 cells stably expressing c-myc-tagged WT Zap70), and P116/pDK (P116 cells stably expressing c-myc-tagged kinase-dead Zap70 mutant) cells were transfected with GRE-TK-Luc (GR responsive reporter vector) in the presence of pcDNA4/TOPO/GR (GR expressing plasmid) followed by treatment with DEX, P116, and P116/pDK, cells showed markedly reduced luciferase activity as compared with WT and P116/pWT cells (Fig. 1 A). Under similar experimental conditions GAPDH promoter activity did not differ significantly between the cell lines (data not shown). These data indicate that enzymatically active Zap70 is required for normal GR-mediated activation in Jurkat cells.

Lck deficiency is known to compromise Zap70 signaling in T cells. In addition to being important for binding of Zap70 to ITAM, Lck is required for the phosphorylation of tyrosine residues of Zap70 and its subsequent activation (1, 2, 3, 4). We considered the possibility that if active Zap70 is essential for normal GR function, then the loss of Lck could compromise this function in T cells due to the lack of active Zap70 in these cells. To address this possibility, we transfected WT, JCaM1.6, JCaM1.6/vector (JCaM1.6 cells stably transfected with vector), and JCaM1.6/Lck (JCaM1.6 cells stably transfected with Lck) cells with GRE-TK-Luc in the presence of pcDNA4/TOPO/GR plasmid followed by treatment with DEX. The results show that JCaM1.6 and JCaM1.6/vector cells exhibited significantly lower levels of GR activity (Fig. 1 B) as compared with WT and JCaM1.6 cells reconstituted with Lck, indicating that Lck is needed for normal GR activation in these cells.

Together, these results indicate that Zap70 plays an important role in GR signaling by modulating ligand-induced transcriptional activation and that Zap70 activation by Lck is an essential step to accomplish this mechanism.

To investigate the importance of Zap70 signaling in the regulation of GR-dependent gene expression, we screened P116 Jurkat T cells by gene array analysis (data not shown) and found that these cells expressed significantly reduced levels of ANXA1 mRNA as compared with expression in WT Jurkat cells. To validate the gene array data, we estimated the level of ANXA1 mRNA in WT and P116 Jurkat cells by RT-PCR. Fig. 2,A shows that there is a marked reduction in the levels of ANXA1 mRNA in P116 cells as compared with the WT cells. There was no significant difference in the expression of GAPDH mRNA between the two cell lines. Western blot analysis using Abs to human ANXA1 also showed significant reduction in ANXA1 protein expression in P116 as compared with WT cells (Fig. 2,B). No significant differences were observed between the two cell lines in the expression of ANXA VI and VII proteins, two other members of the annexin family of proteins expressed in Jurkat cells. In addition, GAPDH protein levels were also similar in the two cell lines. These data indicate that presence of Zap70 is essential for the normal ANXA1 expression in Jurkat cells. To investigate whether P116 cells that stably express WT Zap70 or catalytically inactive mutant (K369R) of Zap70 were able to express normal levels of ANXA1, we screened these cells for the expression of ANXA1 mRNA and protein. The data show that P116/pWT cells expressed ANXA1 mRNA (Fig. 2,C) and protein (Fig. 2,D) levels that closely approached the levels seen in WT Jurkat cells. In contrast, P116/pDK cells expressed ANXA1 mRNA and protein levels that were similar to the levels obtained with P116 cells. The expression of GAPDH, ANXA VI, and ANXA VII proteins was similar in all the cell lines (Fig. 2,D). These results demonstrate that protein tyrosine kinase activity of Zap70 can restore normal ANXA1 expression in P116 cells unlike the catalytically inactive Zap70 that fails to do so. In addition, Zap70 effect is specific to ANXA1 expression as the levels of two other members of annexin family expressed in Jurkat cells do not respond to the loss of Zap70 or the expression of catalytically inactive Zap70. To rule out the possibility that the low level expression of ANXA1 may be a nonspecific phenomenon in P116 cells, we knocked down Zap70 in WT Jurkat cells using RNAi. Fig. 2,E shows that the transfection of WT Jurkat cells with Zap70-specific siRNA duplexes resulted in a dramatic reduction in the levels of ANXA 1 expression. Under similar experimental conditions nonspecific siRNA duplexes had no effect on ANXA1 expression, indicating that Zap70 expression is indeed essential for the normal expression of ANXA1 in Jurkat cells. Fig. 2 F shows the fold loss in ANXA1 mRNA and/or protein expression in various cell lines and Zap70 siRNA-transfected WT Jurkat cells.

We next considered the possibility that if active Zap70 is essential for normal ANXA1 expression, then the loss of Lck could decrease ANXA1 expression in T cells due to the lack of active Zap70 in these cells. To address this question, we compared the levels of ANXA1 mRNA and protein expression in Lck-negative JCaM1.6 cells with the levels in WT Jurkat cells. The results show that JCaM1.6 cells express significantly lower levels of ANXA1 mRNA (Fig. 3,A) and protein (Fig. 3,B) as compared with WT Jurkat cells, indicating that Lck is needed for normal ANXA1 expression. The expression of Zap70, ANXA VI, and ANXA VII proteins was similar in both cell lines. To investigate whether JCaM1.6 cell line that stably expresses Lck produce more ANXA1 than parental JCaM1.6 cells, we screened JCaM1.6/Lck and JCaM1.6/vector cell lines for the presence of ANXA1 mRNA and protein. The results show that JCaM1.6/Lck cells expressed significantly more ANXA1 mRNA (Fig. 3,C) and protein (Fig. 3,D) than JCaM1.6/Vector cell line. This reconstitution experiment further demonstrates that Lck is necessary for normal ANXA1 expression. Finally, knockdown of Lck in WT Jurkat cells using RNAi with Lck-specific siRNA greatly reduced ANXA1 levels in these cells (Fig. 3,E). Nonspecific siRNA had no effect on the ANXA1 levels and the levels of Zap70 remained unchanged in the transfected cells. Fig. 3 F shows the fold loss in ANXA1 mRNA and/or protein expression in various cell lines and Lck siRNA transfected WT Jurkat cells. Together, these data strongly point to an essential role of Lck-dependent signaling in maintaining normal ANXA1 expression in Jurkat cells.

When the half-life of ANXA1 mRNA was determined, there was no significant difference between WT Jurkat and P116 cells (data not shown), suggesting that Zap70 does not affect the stability of ANXA1 mRNA thus excluding the possibility that posttranscriptional mechanisms are involved in the maintenance of normal ANXA1 expression in these cells. Glucocorticoids are known to regulate ANXA1 levels. We found that the incubation of WT Jurkat and P116 cells with DEX resulted in a significant increase in the expression of ANXA1 protein only in WT Jurkat cells, indicating that Zap70 is essential in the DEX-induction of ANXA1 expression (Fig. 4,A). Analysis of the human ANXA1 promoter (Fig. 4,B) indicates that there is a GRE starting at +237 bp from the transcriptional start site in this promoter (24). We studied the effect of DEX on the induction of ANXA1 promoter with and without GRE. In addition to being more active than ANXA1-Luc (possibly due to the presence of additional transcription factor binding sites), only ANXA1-GRE-Luc was induced by DEX in Jurkat cells (Fig. 4,C). To study the levels of DEX-induced promoter activity in the absence of Zap70, we transfected cells with ANXA1-Luc-GRE and TOPO/GR plasmids and treated them with DEX. P116 and P116/pDK cells showed significantly lower levels of DEX-induced luciferase activity as compared with WT and P116/pWT cells (Fig. 4,D). This response level indicates that catalytically active Zap70 regulates DEX-induced ANXA1 promoter activity in Jurkat cells. DEX-induced ANXA1-Luc activity was also found significantly diminished in JCaM1.6 and JCaM1.6/vector cells as compared with WT Jurkat and JCaM1.6/Lck cells (Fig. 4 E) demonstrating that Lck is essential for normal DEX-induced ANXA1 promoter activity in Jurkat cells. To summarize, our data demonstrate the involvement of a putative GRE, located in the ANXA1 promoter, in DEX-induced activation and that this GR binding site may function suboptimally in the absence or partial loss of Zap70 signaling.

To extend the observations obtained with Jurkat cell lines to a more physiologic model for T cell function, we used normal human PBT cells in which Zap70 was partially knocked down by RNAi. When PBT cells were transfected with siRNA duplexes specific for human Zap70 or nonspecific siRNA control, there was nearly 62% loss of Zap70 protein in Zap70 siRNA-transfected cells as compared with cells transfected with nonspecific siRNA duplexes. This was accompanied by a statistically significant (p = 0.0075) reduction in the levels of ANXA1 protein in Zap70 siRNA-transfected cells as compared with control (Fig. 5,A). These results indicate that Zap70 is important to maintain normal ANXA1 levels in these cells. We next studied the role of Zap70 in DEX-mediated activation of ANXA1 promoter in PBT cells. This was accomplished by transfecting these cells with ANXA1-GRE-Luc plasmid in the presence of Zap70- specific siRNA or nonspecific siRNA followed by treatment with DEX. As shown in Fig. 5 B, DEX-induced luciferase activity was significantly inhibited by Zap70-specific siRNA, indicating that Zap70 signaling pathway is essential in the modulation of DEX-induced ANXA1 promoter activity in circulating human T cells.

The role of nuclear hormone receptor-mediated gene expression in T cell physiology is poorly understood. Our earlier studies with nuclear retinoid receptors have provided some insights in the functioning of these receptors and have revealed the importance of epigenetic mechanisms in modulating their function during T cell signaling (22, 26, 27, 28, 29). We have recently identified Zap70 as an important kinase in modulating retinoid receptor-dependent activation in T cells (22). In this study we have extended our study to the role of Zap70 and its upstream activating kinase Lck in regulating the GR transactivation and have identified an essential role for these kinases in mediating DEX-dependent GR signaling in T cells. Our studies indicate that the nuclear transport of GR after DEX treatment is not affected in the absence of Zap70 or Lck (data not shown) ruling out the role of Zap70 signaling in the ligand-induced nuclear transport of GR. GR contains many potential phosphorylation sites, including those for MAPKs such as ERK and p38 MAPK, in addition to sites for protein kinase C and protein kinase A (23). Zap70 plays a crucial role in the activation of a number of downstream protein kinases and phosphatases in addition to its essential role in the translocation of protein kinase Cθ to the lipid rafts during the formation of functional immune synapse (5, 6, 30, 31). Whether Zap70 regulates the GR activity by direct phosphorylation or via downstream kinases remains to be studied. Recent studies have found that DEX induces rapid phosphorylation of Zap70 and DEX-bound GR associates with Zap70 (32, 33), indicating cross-talk between the two signaling molecules. A role for the DEX-induced phosphorylation of Zap70 and its association with the receptor in mediating GR-dependent transactivation is unknown, but remains a possibility.

We have also identified a novel role for Zap70 signaling pathway in maintaining normal levels of ANXA1, an anti-inflammatory protein known to be induced and released by the glucocorticoid action. Zap70 signaling was found essential in maintaining normal levels of both ANXA1 mRNA and protein. However, there was no tight correlation between the fold loss of ANXA1 mRNA and protein in Zap70- (Fig. 2,F) and Lck-negative (Fig. 3,F) cells that would suggest a differential impact of these kinases on protein and mRNA expression. The loss of ANXA1 levels induced by Zap70-specific siRNA was more pronounced in Jurkat as compared with PBT cells indicating that additional mechanisms are also operative in maintaining normal ANXA1 levels in PBT cells. In this study, Zap70 RNAi resulted in >90% loss in Zap70 expression in Jurkat cells as compared with only 62% in PBT cells. This difference in the levels of RNAi-induced Zap70 loss could also explain lower ANXA1 reduction in PBT cells as compared with Jurkat cells. Although this study does not elucidate the complete mechanism of loss of ANXA1 expression, it is clear that GRE-mediated ANXA1 promoter function is attenuated in the absence of Zap70 in both Jurkat and PBT cells. In addition to GRE, ANXA1 promoter has binding sites for various other transcription factors. Because the GR-independent activity of the ANXA1 promoter was also compromised in the kinase-negative cells (Fig. 4, D and E), this study does not exclude the possibilities that other promoter elements also play essential roles in the Zap70–dependent regulation of ANXA1 expression. In addition to regulating GR-mediated transcription, Zap70 may also mediate ANXA1 promoter activation through the modulation of various signaling intermediates that are essential for the activation of other transcriptional binding sites. Recent studies have shown that Src kinases like Lck are responsive to histone deacetylase inhibitors, indicating a cross-talk between acetylation and tyrosine kinase pathways (34). To consider the possibility that the loss of ANXA1 expression in Zap70-negative P116 cells may be due to the defect in the acetylation in these cells, we studied the effect of TSA, an inhibitor of histone deacetylases (35), on the expression of ANXA1 mRNA and protein in P116 cells. Treatment with TSA increased the levels of ANXA1 protein (Fig. 6,A) and mRNA (Fig. 6 B) in both WT and P116 cells, indicating that acetylation plays a role in the regulation of ANXA1 expression and that Zap70 may not be important for TSA-induced increase in ANXA1 expression in these cells. To further investigate whether TSA effect on the ANXA1 levels involved the induction of ANXA1 promoter and if so, was the level of induction any different in P116 cells as compared with WT cells, we performed transient transfection with ANXA-Luc reporter in the presence of TSA and found that the lack of Zap70 does not significantly alter the response of ANXA1 promoter to the effects of TSA (data not shown). These results indicate that ANXA1 promoter is tightly regulated in T cells and the level of histone acetylation associated with ANXA1 promoter may not play a major role in its activation. Furthermore, TSA induction of ANXA1 mRNA and protein levels reflects the probable role of nonhistone transcriptional regulators (not necessarily associated with the promoter) (34) in the regulation of ANXA1 gene expression.

ANXA1 is emerging as an essential protein with diverse biological functions including anti-inflammatory role in host inflammatory responses. In addition to T cells, ANXA1 is expressed in neutrophils, endothelial, and mast cells that are involved in the process of inflammation (15). Previous studies have shown that ANXA1 regulates engulfment of apoptotic cells (21, 36), suppresses activation of autoimmune T cell lines (37), and induces apoptosis in neutrophils (18). Zap70 is an essential signaling protein that has been shown to have a role in positive and negative selection of thymocytes (7, 8, 9). Altered thymic T cell selection due to a mutation of the Zap70 gene has been implicated in the autoimmune arthritis in mice (10). In addition to the lack of IL-2 expression following T cell activation, Zap70–negative P116 cells have been found to be defective in activation-induced apoptosis due to failure to induce the expression of Fas ligand (38). The downstream targets of Zap70 signaling remain poorly understood. Whether Zap70 mediates some of these important biological functions in T cells by regulating ANXA1 expression remains to be investigated.

Our data have important biological implications in the understanding of the role of Zap70 signaling in modulating GR-mediated transactivation in T cells and anti-inflammatory and other functions of ANXA1. Identification of cross-talk between GR and Zap70 signaling in regulating ANXA1 promoter function is of considerable interest that should lead to new insights into the mechanism of inflammation. A full understanding of the interrelationship between Zap70, GR, and ANXA1 pathways will aid in designing novel anti-inflammatory drugs.

We thank Drs. R. L. Wange, A. Weiss, and K. Ozato for their gifts of cell lines and plasmids. We also thank P. Patel for technical help.

The authors have no financial conflict of interest.