The Protein Phosphatase 2A Regulatory Subunit B56γ Mediates Suppression of T Cell Receptor (TCR)-induced Nuclear Factor-κB (NF-κB) Activity*

Background: The trimeric phosphatase PP2A suppresses NF-κB. The regulatory subunit mediating specificity of PP2A action remains undefined. Results: The regulatory PP2A subunit B56γ mediates suppression of NF-κB resulting in increased NF-κB target gene expression in T cells. Conclusion: B56γ mediates suppression of NF-κB in T cells. Significance: Our goal was an understanding of the physiology of PP2A in T cells and the pathophysiology of diseases involving PP2A and NF-κB. NF-κB is an important transcription factor in the immune system, and aberrant NF-κB activity contributes to malignant diseases and autoimmunity. In T cells, NF-κB is activated upon TCR stimulation, and signal transduction to NF-κB activation is triggered by a cascade of phosphorylation events. However, fine-tuning and termination of TCR signaling are only partially understood. Phosphatases oppose the role of kinases by removing phosphate moieties. The catalytic activity of the protein phosphatase PP2A has been implicated in the regulation of NF-κB. PP2A acts in trimeric complexes in which the catalytic subunit is promiscuous and the regulatory subunit confers substrate specificity. To understand and eventually target NF-κB-specific PP2A functions it is essential to define the regulatory PP2A subunit involved. So far, the regulatory PP2A subunit that mediates NF-κB suppression in T cells remained undefined. By performing a siRNA screen in Jurkat T cells harboring a NF-κB-responsive luciferase reporter, we identified the PP2A regulatory subunit B56γ as negative regulator of NF-κB in TCR signaling. B56γ was strongly up-regulated upon primary human T cell activation, and B56γ silencing induced increased IκB kinase (IKK) and IκBα phosphorylation upon TCR stimulation. B56γ silencing enhanced NF-κB activity, resulting in increased NF-κB target gene expression including the T cell cytokine IL-2. In addition, T cell proliferation was increased upon B56γ silencing. These data help to understand the physiology of PP2A function in T cells and the pathophysiology of diseases involving PP2A and NF-κB.


NF-B is an important transcription factor in the immune system, and aberrant NF-B activity contributes to malignant diseases and autoimmunity. In T cells, NF-B is activated upon TCR stimulation, and signal transduction to NF-B activation is
triggered by a cascade of phosphorylation events. However, finetuning and termination of TCR signaling are only partially understood. Phosphatases oppose the role of kinases by removing phosphate moieties. The catalytic activity of the protein phosphatase PP2A has been implicated in the regulation of NF-B. PP2A acts in trimeric complexes in which the catalytic subunit is promiscuous and the regulatory subunit confers substrate specificity. To understand and eventually target NF-Bspecific PP2A functions it is essential to define the regulatory PP2A subunit involved. So far, the regulatory PP2A subunit that mediates NF-B suppression in T cells remained undefined. By performing a siRNA screen in Jurkat T cells harboring a NF-B-responsive luciferase reporter, we identified the PP2A regulatory subunit B56␥ as negative regulator of NF-B in TCR signaling. B56␥ was strongly up-regulated upon primary human T cell activation, and B56␥ silencing induced increased IB kinase (IKK) and IB␣ phosphorylation upon TCR stimulation. B56␥ silencing enhanced NF-B activity, resulting in increased NF-B target gene expression including the T cell cytokine IL-2. In addition, T cell proliferation was increased upon B56␥ silencing. These data help to understand the physiology of PP2A function in T cells and the pathophysiology of diseases involving PP2A and NF-B.
The transcription factor NF-B 3 plays a key role in the immune system by controlling lymphocyte survival and activation (1). Aberrant NF-B activity is implicated in lymphoid malignancies and contributes to a variety of autoimmune disorders (2)(3)(4). In T cells, NF-B is activated upon TCR stimulation (5). Proximal TCR signaling induces activation of the protein kinase C (PKC), which eventually leads to IKK activation. IKK in turn phosphorylates IB proteins, which sequester NF-B in the cytosol. Phosphorylation-induced degradation of IB proteins allows NF-B to translocate to the nucleus and activate transcription (6). Whereas activation of NF-B is well studied, events that terminate NF-B activity are only partially understood (7).
Prominent examples of negative regulators of TCR-induced NF-B activity are the two ubiquitin-editing enzymes A20 and cylindromatosis (CYLD). Defects in NF-B regulation by A20 or CYLD lead to increased NF-B activity and, consequently, hyperactivation of cells and promotion of lymphoid malignancies (8).
Phosphatases remove activating phosphate groups and, hence, are important regulators of TCR signaling (7,9). The serine/threonine protein phosphatase 2A (PP2A) is ubiquitously expressed and together with PP1 the most abundant cellular serine/threonine phosphatase (10). PP2A acts in trimeric complexes consisting of a catalytic C subunit, a scaffolding A subunit, and a regulatory B subunit (11). The catalytic activity of PP2A is involved in almost all cellular processes such as various signaling pathways, cell cycle regulation, and migration. Because the catalytic PP2A subunit acts promiscuously with regard to phosphoprotein recognition and dephosphorylation, the regulatory subunits confer substrate specificity and subcellular localization (11). Up until now, four regulatory PP2A B subunit families with up to five family members have been identified, resulting in a plethora of differ-ent PP2A holoenzymes (11). To understand specific PP2A function, it is essential to characterize the regulatory subunit involved (12,13). However, despite its importance in regulation of PP2A, the regulatory PP2A subunit that mediates NF-B suppression in T cells is not yet defined.
We performed a siRNA screen in Jurkat T cells harboring a NF-B-responsive luciferase reporter to identify phosphatases involved in TCR-mediated NF-B signaling (14). Here, we identified and validated the PP2A regulatory subunit B56␥ as suppressor of NF-B in TCR signaling. B56␥ was strongly upregulated upon primary human T cell activation, and B56␥ silencing increased IKK and IB␣ phosphorylation upon TCR stimulation. In addition, phorbol 12-myristate 13-acetate (PMA)-induced NF-B activity was suppressed by B56␥, indicating a role of B56␥ downstream of PKC. B56␥ silencing caused enhanced NF-B activity, which resulted in increased NF-B target gene expression in primary human T cells. Especially, expression of the important T cell growth factor IL-2 (15) was strongly enhanced on mRNA and protein level, when B56␥ expression was silenced. Moreover, T cell proliferation was increased upon B56␥ silencing. Thus, B56␥ mediates suppression of NF-B in TCR signaling. These findings contribute to a more detailed understanding of PP2A holoenzyme function in T cell signaling and help to understand the physiology of PP2A cellular function and the pathophysiology of diseases involving PP2A and NF-B.

EXPERIMENTAL PROCEDURES
siRNA Screen and Screen Analysis-The RNAi screening procedure and the screen analysis were performed as described previously (14), except that single replicate siRNA transfections were analyzed separately. This allows for the analysis of the consistency of siRNA phenotypes. The threshold to discriminate hits from background noise was set to a Z score of ϩ1.5 for at least two of three siRNAs transfected per gene in one replicate. To select for consistently scoring candidates, these siRNAs had to score with a Z score above ϩ1.0 in the other replicate.
Cell Culture and Cell Stimulation-Human peripheral T cells were prepared and cultured as described previously (16). Autologous monocytes were isolated from T cell-depleted peripheral blood mononuclear cells and differentiated to dendritic cells as described previously (17). Alternatively, T celldepleted peripheral blood mononuclear cells were cultured in RPMI 1640 medium with 1% AB-Serum for 1 h, and adherent monocytes were differentiated to dendritic cells accordingly. Jurkat T cells (JE6.1; Sigma) and GLuc-J16 Jurkat T cells (14) were cultured in Iscove's modified Dulbecco's medium supplemented with 10% FCS. To induce shRNA expression stably transduced cells were treated with 2 g/ml doxycycline for 3 days. Jurkat T cells and primary human T cells were stimulated with agonistic soluble ␣-CD3 (OKT3), ␣-CD28 (CD28.2; BD Biosciences) and cross-linking goat anti-mouse monoclonal antibodies (Southern Biotechnology) to mimic TCR stimulation. Alternatively, cells were stimulated with PMA (Sigma) or TNF␣ (D. Männel, University of Regensburg, Germany).
ELISA-Human IL-2 ELISA was purchased from BD Biosciences and performed according to the manufacturer's protocol. Data were statistically analyzed with the two-tailed unpaired t test with Welch's correction (*, p Ͻ 0.05; **, p Ͻ 0.01; ***, p Ͻ 0.001).
Secretion Assay-APC-IL-2 and PE-IFN␥ secretion assays were obtained from Miltenyi Biotec and performed according to the manufacturer's instructions. By catching secreted cytokines on the cell surface with a cytokine-specific antibody that can be coated on cells, the number of cells that secrete this cytokine were detected with a fluorescently labeled cytokinespecific antibody by FACS. This method is especially developed for detection of antigen-specific T cells. To gate on CD4 ϩ T cells, cells were costained with anti-CD4-FITC antibody (RPA-T4; BD Biosciences) T Cell Proliferation Assay-T cell proliferation was measured by [ 3 H]thymidine incorporation as described previously (19). Data were statistically analyzed with the two-tailed unpaired t test with Welch's correction (**, p Ͻ 0.01; ****, p Ͻ 0.0001).
Reverse Transcriptase PCR and Quantitative Real-time PCR-Reverse Transcriptase PCR and Quantitative Real-Time PCR were performed as described previously (14).

B56␥ Suppresses TCR-and PMA-induced NF-B Activation-
We have performed a siRNA screen in Jurkat T cells harboring a NF-B-responsive luciferase reporter (GLuc-J16 T cells) to identify phosphatases involved in TCR-mediated NF-B signaling (14). Because not all positive controls scored in the initial screen analysis, we reanalyzed our screen (Fig. 1A). Every gene analyzed in our screen was covered by three independent siRNAs transfected in duplicate. The Z score, a measure of the biological effect of a siRNA reflecting the number of standard deviations from the population mean (20), was applied to replicate transfections separately in our new analysis. To identify PP2A regulatory subunits that mediate suppression of NF-B in TCR signaling, we specifically analyzed the effects of these subunits included in our new siRNA screen analysis (Fig. 1B). PPP2R5C, the gene encoding B56␥, was the only PP2A regulatory subunit meeting our refined threshold criteria, whereas all other PP2A regulatory subunits did not score at this threshold, indicating a specific role of B56␥ in TCR-mediated NF-B signaling.
Subsequently, we confirmed the results of the reanalysis with NF-B luciferase reporter gene assays. Knockdown of B56␥ by two independent siRNAs increased NF-B activity upon TCR stimulation compared with control siRNA-treated cells (Fig. 2,  A and B). Conversely, overexpression of B56␥ decreased NF-B activity upon TCR stimulation compared with vector control (Fig. 2, C and D). In addition, NF-B activation by TNF␣ was decreased by ectopic expression of B56␥ compared with vector control (Fig. 2, C and E). Moreover, activation of NF-B by PMA, a pharmacological diacylglycerol mimetic, was suppressed upon overexpression of B56␥ (Fig. 2, C and F). PMA circumvents proximal TCR signaling and directly activates PKC to induce NF-B activation (21). Hence, B56␥ must act on or downstream of PKC.
B56␥ Suppresses NF-B Activation between PKC and the IKK Complex-To further investigate the point of interference of B56␥ with NF-B signaling we analyzed IKK and IB␣ phosphorylation upon TCR stimulation in B56␥ knockdown cells. Knockdown of B56␥ led to increased phosphorylation of the IKK complex 10 min after stimulation compared with control, and this increase lasted up to 45 min of stimulation (Fig. 3). In addition, phosphorylation and degradation of IB␣ were elevated in B56␥ knockdown cells compared with control cells. Moreover, TCR-induced activation of the MAPK pathway was analyzed. Phosphorylation of ERK was similar in B56␥ knockdown cells compared with control cells (Fig. 3). In conclusion, B56␥ regulates NF-B signaling on the level or between PKC and IKK complex phosphorylation.
B56␥ Is Up-regulated upon TCR Stimulation-To address the function of B56␥ in primary human peripheral blood T cells we tested B56␥ expression in resting cells and upon T cell activation. Primary human T cells were activated with the mitogen phytohemagglutinin (PHA) and expanded in the presence of IL-2 for up to 6 days. B56␥ was expressed at basal levels in the resting state (Fig. 4A)  further kinetics were applied using TCR-agonistic antibodies. B56␥ expression was up-regulated after 11 h of TCR stimulation (Fig. 4B). Collectively, B56␥ expression is up-regulated upon TCR stimulation, which might point to a role of B56␥ in the control of NF-B activation in activated T cells. B56␥ Suppresses NF-B Target Gene Expression-To further assess the physiological function of B56␥-mediated NF-B suppression in primary human T cells, TCR-induced NF-B target gene transcription was analyzed upon knockdown of B56␥. siRNA-transfected cells were stimulated via the TCR, and NF-B target gene transcription was analyzed using a qPCR array (Fig. 4, A and B). B56␥-deficient cells exhibited a specific Values were normalized to cell viability using CellTiter-Glo. C-F, Jurkat T cells were transfected with an expression plasmid encoding HA-tagged B56␥ or empty vector control. In addition, a NF-B reporter system was cotransfected to determine NF-B activity. 48 h after transfection B56␥ overexpression was confirmed by Western blotting (C), and cells were stimulated as indicated for 5 h (D-F). NF-B activity was measured using a reporter gene assay. **, p Ͻ 0.01; ***, p Ͻ 0.001; ****, p Ͻ 0.0001. Error bars, S.D.  Bands were quantified, and the B56␥ signal was normalized to the respective actin signal. The 0 time point was set to 1. Numbers below the panels indicate relative quantification of signal intensity. One representative of five donors is shown. B, primary human T cells were stimulated via the TCR with ␣-CD3 ϩ ␣-CD28 antibodies for the indicated time points. Lysates were analyzed by Western blotting. Bands were quantified, and the B56␥ signal was normalized to the respective actin signal. The 0 time point was set to 1. Numbers below the panels indicate relative quantification of signal intensity. One representative of three donors is shown.
NF-B target gene signature compared with control cells with most target genes up-regulated by at least 1.5-fold. After 60 min of stimulation, especially IL-2 and IFN␥ transcription were increased upon B56␥ knockdown by 2.8-and 2.9-fold, respectively. To substantiate the qPCR array results, we tested IL-2 transcription in more detailed kinetics by conventional qPCR (Fig. 5, A and C). IL-2 transcription was increased by approximately 2-fold. In conclusion, B56␥-mediated NF-B suppression regulates transcription of several NF-B target genes including IL-2.
To confirm these results, the induction of IL-2 was analyzed on the protein level. For this purpose, activated primary human T cells were transfected with two independent siRNA oligonucleotides targeting B56␥. siRNAs targeting the known NF-B suppressor CYLD (8,22) or nontargeting siRNA were used as controls. Knockdown of B56␥ led to an increase in IL-2 secretion for both B56␥ siRNA oligonucleotides compared with siRNA control (Fig. 6, A and B). Knockdown of CYLD increased IL-2 secretion to a similar extent. To evaluate the results obtained for all six donors, the -fold increase of IL-2 secretion for knockdown of B56␥ and CYLD compared with siRNA control was calculated for both concentrations of TCR-stimulating antibodies (Fig. 6, C and D). The median IL-2 induction was ϳ2-fold for knockdown of B56␥ with two independent siRNA oligonucleotides and for knockdown of CYLD, respectively. In addition, we analyzed IL-2 and IFN␥ secretion upon T cell stimulation with staphylococcal enterotoxin B in the presence of autologous dendritic cells using a secretion assay (Fig. 6, E and F). Upon B56␥ knockdown, IFN␥-and IL-2-secreting CD4 ϩ T cells were increased compared with control, further substantiating our results.
B56␥ Suppresses T Cell Proliferation-To assess whether T cell proliferation was also affected by B56␥-mediated suppression of NF-B, we performed T cell proliferation assays. siRNA-transfected primary human T cells were stimulated via the TCR, and after 5 days [ 3 H]thymidine incorporation was measured (Fig. 7). Knockdown of B56␥ led to an increase in proliferation for both B56␥ siRNA oligonucleotides compared with siRNA control (Fig. 7, A and B). The median -fold increase in proliferation of six independent donors comparing B56␥ knockdown and control cells was 7.3-fold (Fig. 7C). In conclusion, our data indicate a role of B56␥ as suppressor of NF-Bdriven activation in human T cells.

DISCUSSION
NF-B is an essential transcription factor in immunity, and aberrant regulation of NF-B contributes to disorders such as cancer and autoimmunity (1-4). Understanding of NF-B reg- ulation in physiological and pathophysiological conditions will enable the development of targeted therapies. NF-B is essential for activation of normal T cells (5). In lymphomas, aberrant NF-B activity is described to contribute to pathogenesis (23). Activation of NF-B upon TCR stimulation by phosphorylation is well studied (6). However, termination and fine-tuning of TCR signaling by dephosphorylation are only partially understood (7).
Phosphatases are important negative regulators in signal transduction (11). Serine/threonine phosphatases act in dimeric or trimeric complexes consisting of a promiscuous catalytic subunit, a scaffolding subunit, and a specificity-mediating regulatory subunit (11). The catalytic activity of the serine/thre-onine phosphatase PP2A is described to be involved in NF-B regulation (24 -28). However, the regulatory subunit that confers specificity to TCR-mediated NF-B regulation by PP2A was so far undefined.
To our knowledge, this study is the first description of the PP2A regulatory subunit B56␥ in TCR-mediated NF-B regulation. We performed a siRNA screen to identify phosphatases involved in TCR signaling (14). B56␥ was the only PP2A regulatory subunit that scored in this siRNA screen. Both B56␥ siRNAs scoring in our screen showed a similar strength in NF-B enhancement, indicating the robustness of our findings. Furthermore, we provide several lines of evidence that the PP2A regulatory subunit B56␥ mediates suppression of TCR- induced NF-B activity. (i) Knockdown of B56␥ enhanced TCR-mediated NF-B activity, and conversely, overexpression of B56␥ suppressed NF-B activity upon TCR triggering. (ii) In primary human T cells B56␥ knockdown enhanced TCR-induced NF-B target gene expression on mRNA and protein levels by ϳ2-fold. Notably, knockdown of the known negative NF-B regulator CYLD increased TCR-induced IL-2 secretion to a similar extent. (iii) T cell proliferation was enhanced upon B56␥ knockdown. (iv) TCR-induced IKK phosphorylation, IB␣ phosphorylation, and IB␣ degradation were enhanced upon B56␥ knockdown, whereas TCR-induced ERK phosphorylation remained unaffected.
PP2A is described to regulate a plethora of signaling pathways and cellular processes (11). However, PP2A is a trimeric complex consisting of a catalytic C subunit, a scaffolding A subunit, and a regulatory B subunit. The PP2A C and A subunits are represented by two homologous genes in humans, but for the regulatory B subunits a multitude of different unrelated protein families exists (11). The regulatory B subunits confer specificity to the dephosphorylation activity of PP2A and hence are essential to drug a specific PP2A function (12,13). PP2A regulatory subunits, however, have not been investigated in detail. Several studies link PP2A catalytic activity to regulation of IKK phosphorylation (26,29,30). Our results show that B56␥ suppresses IKK phosphorylation upon TCR stimulation, indi-cating a role for B56␥ on the level of IKK phosphorylation or upstream. In addition, the PP2A core enzyme, consisting of the catalytic PP2A C␣ and scaffolding PP2A A␣ subunit, is described to dephosphorylate CARMA1 (CARD-containing membrane-associated guanylate kinase protein 1), a mediator of NF-B activation in T cells, and to consequently suppress IL-2 secretion upon TCR triggering (24). It is likely that B56␥ is the missing regulatory subunit to form the PP2A holoenzyme. B56␥ may strengthen phosphatase substrate interaction or modulate phosphatase activity.
Moreover, B56␥ suppressed PMA-induced NF-B activity. PMA circumvents proximal TCR signaling and directly activates PKC (21). Hence, B56␥ must act downstream or on the level of PKC to suppress NF-B (Fig. 8). Besides TCR triggering and PMA treatment, TNF␣ stimulation also induces NF-B activation in T cells (6). TNF␣-induced NF-B activity was also suppressed by B56␥ in T cells. In line with these findings, B56␥ was reported to mediate dephosphorylation of TRAF2 (TNF receptor-associated factor 2) upon TNF␣ stimulation in an astrocyte cell line (26). TRAF2 is a mediator of TNF␣-induced NF-B activation that is not relevant in TCR signaling (6,31). This indicates a dual role of B56␥ in NF-B regulation, suppressing both TNF␣and TCR-mediated NF-B activation by distinct mechanisms. We have shown enhanced expression of B56␥ upon activation of primary human T cells. This may imply a potential role of B56␥ in differentiation of these cells. In line with our findings, two other B56 family members, B56␤ and B56␦, were shown to be up-regulated during differentiation of a neuron-like cell line (32). In addition, also other modulators of NF-B are reported to be up-regulated upon T cell activation (14,33,34).
NF-B is relevant in many malignancies. Especially in lymphomas and leukemias, regulators of NF-B are mutated or differentially expressed (14,(35)(36)(37). Besides a general connection between PP2A and cancer, B56␥ is implicated as a tumor suppressor in different settings (12). B56␥ is described to be down-regulated in human melanoma and in several lung cancer cell lines (38,39). Moreover, deregulation of several oncogenic signaling pathways upon depletion of B56␥, including Akt and Wnt signaling, have been reported (12). In addition, the tumor suppressor p53 was described to be activated by B56␥-containing PP2A. Importantly, NF-B and p53 fulfill opposing roles in tumorigenesis, acting as protooncogene and tumor suppressor, respectively (3). B56␥ not only activates p53 and regulates Akt and Wnt signaling; our findings, that B56␥, in addition, suppresses NF-B add further weight to the tumor suppressive role of B56␥ and make it an interesting target for cancer therapy.