key: cord-0000531-qzkpik3z
authors: Carvalho, Gabrielle; Poalas, Konstantinos; Demian, Catherine; Hatchi, Emeline; Vazquez, Aimé; Bidère, Nicolas
title: Participation of the Cell Polarity Protein PALS1 to T-Cell Receptor-Mediated NF-κB Activation
date: 2011-03-30
journal: PLoS One
DOI: 10.1371/journal.pone.0018159
sha: 8996c8a004bc82aca37ba1f1f4265868503ce69f
doc_id: 531
cord_uid: qzkpik3z

BACKGROUND: Beside their established function in shaping cell architecture, some cell polarity proteins were proposed to participate to lymphocyte migration, homing, scanning, as well as activation following antigen receptor stimulation. Although PALS1 is a central component of the cell polarity network, its expression and function in lymphocytes remains unknown. Here we investigated whether PALS1 is present in T cells and whether it contributes to T Cell-Receptor (TCR)-mediated activation. METHODOLOGY/PRINCIPAL FINDINGS: By combining RT-PCR and immunoblot assays, we found that PALS1 is constitutively expressed in human T lymphocytes as well as in Jurkat T cells. siRNA-based knockdown of PALS1 hampered TCR-induced activation and optimal proliferation of lymphocyte. We further provide evidence that PALS1 depletion selectively hindered TCR-driven activation of the transcription factor NF-κB. CONCLUSIONS: The cell polarity protein PALS1 is expressed in T lymphocytes and participates to the optimal activation of NF-κB following TCR stimulation.

Establishment and maintenance of cell polarity is chiefly orchestrated by a tightly regulated interplay between three multi-protein complexes: i) Scribble (SCRIB)/Discs Large (Dlgh1)/Lethal giant larvae (Lgl) complex, ii) partitioning-defective (PAR) 3 and PAR6/ atypical protein kinase C (aPKC) complex, and iii) Crumbs (CRB)/ Protein Associated with Lin Seven 1 (PALS1)/ PALS1-associated tight junction protein (PATJ) complex [1, 2] . However, each complex is not exclusive, as PAR6 links PALS1 to PAR3/PAR6/aPKC [3] . In T lymphocytes, cell polarity proteins were shown to partition the leading edge from the uropod at the cell rear, and therefore participate to cell migration, homing, and scanning [4, 5, 6] . In addition, SCRIB and Dlgh1 are transiently recruited to the nascent immunological synapse formed with an antigen-presenting-cell (APC) [4] . Their depletion in lymphocytes has been associated with an alteration of antigen receptor-mediated activation [7, 8, 9, 10] .

The adaptor PALS1 is crucial for cellular architecture as it maintains the apico-basal polarity in epithelial cells and authorizes indirect interactions between CRB and PATJ [11, 12] . Interestingly, Dlgh1 and PALS1 share a COOH-terminal part composed of a PSD-95/Dlg/ZO-1 (PDZ) domain followed by an SH3 domain adjacent to an inactive Guanylate kinase (GK) homology region [2] . This unique sequence of PDZ/SH3/GK defines the so-called membrane-associated guanylate kinase (MAGUK) proteins family, a group of molecules that serve as scaffolds to organize multi-protein signalosomes through their protein-protein interaction domains [13] . For example, the MAGUK-containing CARMA1 emerges as a central regulator of lymphocytes activation and proliferation downstream of antigen receptor stimulation [14] . Indeed, CARMA1 operates as scaffold to recruit the heterodimer BCL10/ MALT1 (CBM complex), a key step for conveying NF-kB signaling [14, 15, 16] . In addition to its established role in polarity, Dlgh1 was shown to modulate lymphocyte proliferation upon T-cell receptor ligation, possibly through p38 recruitment or via the transcription factor NF-AT [9, 10, 17, 18] .

Although the MAGUK PALS1 plays a central role in the establishment of cell polarity, its contribution to lymphocyte activation remains elusive [8] . Here we show that PALS1 mRNA and protein is expressed in human lymphocytes. Furthermore, knocking down of PALS1 with small interfering RNAs (siRNAs) led to a decreased proliferation of human T lymphocytes, resulting from a reduced activation of the transcription factor NF-kB.

Although several cell polarity proteins have been characterized in lymphocytes [4, 5] , PALS1 expression in T cells remains to be determined [8] . To address this question, we first performed RT-PCR analysis on resting human CD3 + T cells and Jurkat lymphocytes extracts, and detected mRNA for PALS1 ( Figure 1A ). These mRNA were efficiently translated into protein, as antibodies against PALS1 detected a band, which was absent from PALS1-siRNA transfected primary T lymphocytes lysates ( Figure 1B ). Similar results were obtained with Jurkat T cells ( Figure 1B ). Of note, PALS1 levels remained unchanged in cells stimulated with antibodies to CD3 and CD28, or with PMA and ionomycin ( Figure 1C ). We next investigated PALS1 subcellular location by confocal microscopy. In contrast to epithelial cells where it accumulate to tight junctions [12] , PALS1 did not reach membrane domains and remains essentially cytosolic with punctuate structures. Additional staining revealed that these structures coalesced with the Golgi apparatus ( Figure 1D and Figure S1 ). Accordingly, Brefeldin A-triggered disassembly of the Golgi apparatus also disrupted PALS1 punctuate structures ( Figure 1D ). This is reminiscent of PALS1 relocation to the Golgi apparatus in cells infected with SARS coronovirus [19] . Last, we observed that TCR-mediated stimulation only promoted a discrete redistribution of PALS1 within the cytosol of Jurkat cells ( Figure S1 ). Altogether, our results suggest that similarly to Dlgh1, SCRIB, CRB3, and PKCf [4, 5] , the cell polarity protein PALS1 is expressed in lymphocytes at both mRNA and protein level.

Because SCRIB and Dlgh1 were proposed to modulate lymphocyte proliferation [7, 8, 9, 10] , we evaluated whether PALS1 might also participate to T cell activation. To this end, peripheral blood lymphocytes (PBL) were purified on Ficoll-isopaque gradients. Primary human T cells were nucleofected for three days with siRNA targeting PALS1, prior stimulation with anti-CD3 and anti-CD28 antibodies. PALS1 knockdown led to a significant decrease in TCR-mediated induction of the activation markers CD69 and CD25 on cell surface (Figure 2A , B). This was accompanied by a reduction in Carboxyfluorescein Succinimydyl Ester (CFSE) dilution, which reflects cell proliferation ( Figure 2C ). Collectively, these data suggest that PALS1 participates to the optimal lymphocyte activation and subsequent proliferation upon TCR stimulation.

To further explore how PALS1 impacts on lymphocyte proliferation, early signaling pathways emanating from the TCR were examined in Jurkat cells transfected with PALS1 siRNA. We did not detect major alteration in the general pattern of tyrosine phosphorylation, or mitogen-activated protein kinase (MAPK) extracellular signal-regulated kinases (ERK) 1/2 phosphorylation upon TCR stimulation ( Figure 3A ). Only a slight but consistent increase in TCR-mediated phosphorylation of p38 was noted ( Figure 3A ). Moreover, CD3-induced calcium mobilization was largely normal in PALS1-knockdown Jurkat cells ( Figure 3B ).

We next analyzed TCR-mediated activation of NF-AT and NF-kB transcription factors. siRNA-treated Jurkat T cells were cotransfected with firefly luciferase constructs driven by NF-AT or NF-kB binding sequences and with a renilla luciferase control. PALS1 knockdown had only a marginal effect on NF-AT activity following stimulation with PMA and ionomycin, or with antibodies to CD3 and CD28 ( Figure 3C and Figure S2 ). In sharp contrast, NF-kB activity was significantly reduced without PALS1 ( Figure 3D ). Interestingly, tumor necrosis factor-a (TNFa)induced NF-kB activation remained essentially unaffected, underscoring the selective involvement of PALS1 in the TCR-NF-kB pathway ( Figure S3 ). Altogether, our data unveiled an unexpected role for PALS1 in TCR-mediated NF-kB activation.

To gain insights on how PALS1 modulate NF-kB, we first investigated the transcription factor binding ability by electrophoretic mobility shift assay (EMSA). Less NF-kB bound to its specific probe in nuclei extracts from PALS1-siRNA transfected cells following TCR stimulation ( Figure 4A ). As expected, Oct-1 binding remained unchanged without PALS1. Consistent with a diminished NF-kB activity, both the phosphorylation and subsequent proteasomal degradation of NF-kB inhibitor, IkBa, were severely decreased in the absence of PALS1 ( Figure 4B ). Because TCR-induced NF-kB activation relies on the assembly of the CBM complex [15] , BCL10 was immunoprecipitated from nonspecific (NS-) and PALS1-siRNA transfected Jurkat cells. MALT1, which forms an heterodimer with BCL10, coprecipitated with BCL10 regardless of stimulation. Although PALS1 was not found bound to BCL10, its absence diminished CARMA1 recruitment ( Figure 4C , and data not shown). Hence, our data suggest that PALS1 participates to the optimal translocation and activation of NF-kB upon TCR stimulation, possibly by favoring the CBM assembly.

Since PALS1 nucleates a ternary complex containing CRB3 and PATJ, and further binds PAR6 to maintain cell polarity [3, 20, 21] , their contribution to TCR-mediated NF-kB was evaluated. Similarly to PALS1, mRNA for PATJ, CRB3, PAR6, were efficiently detected by RT-PCR ( Figure 5A ). The same hold true for the unrelated cell polarity protein SCRIB ( Figure 5A ). siRNA-based knockdown of PALS1 and CRB3 significantly decreased NF-kB activation in cells treated by antibodies against CD3 and CD28, or with a mixture of PMA and ionomycin. Although less dramatic, similar results were observed with PAR6 or PATJ knockdown. By contrast, NF-kB was normally activated in the absence of SCRIB ( Figure 5B ). In agreement, IkBa phosphorylation was diminished in lysates from CRB3-depleted cells, and to a lesser extent from PATJ-or PAR6-siRNA transfected cells, and not from SCRIB-depleted cells. Again, ERK phosphorylation occurred normally ( Figure 5C , D, E, and F). Altogether, our data suggest that PALS1 implication in the TCR-NF-kB pathway is inextricably linked to its cell polarity partners.

In summary, our data show that the cell polarity protein PALS1 is expressed in lymphocytes and contributes to their optimal activation. Although Dlgh1 and SCRIB were proposed to modulate NF-AT or p38 [9, 10, 17, 18] and NF-AT [7] respectively, a distinct scenario likely occurs for PALS1. Our results support a model in which PALS1 participates to NF-kB activation, upstream of IkBa phosphorylation and degradation. However, how precisely PALS1 modulates NF-kB remains unclear. Because MAGUK function as scaffold units to organize and integrate multi-molecular signaling complexes [13] , it is tempting to speculate that PALS1 nucleates its own signalosome. For example, CARMA1 anchors a .900 kDa complex including the heterodimer BCL10/MALT1 [22] , and Dlgh1 was reported to bind to Lck, Zap70, Wasp [17] , and p38 [10] . In our hands, PALS1 did not integrate the CBM, but its absence reduced CARMA1 binding to BCL10. It will therefore be interesting to identify PALS1 partners in the context of lymphocyte activation. In line with this, CRB3, PATJ and PAR6, which all bound PALS1 to maintain cell polarity [2] , also participate to NF-kB signaling upon TCR ligation in lymphocytes, and might therefore complex with PALS1 in lymphocytes. Altogether, our results strengthen the unexpected function of cell polarity proteins in lymphocyte proliferation [7, 8, 9, 10] , and unveil an original role for PALS1 during TCR-mediated NF-kB activation.

Jurkat T cells E6.1 were purchased from ATCC. CD3 + human T lymphocytes from healthy donors (Etablissement Francais du Sang) were isolated with the MidiMacs system (Miltenyi Biotec). Cells were activated with a mixture of soluble anti-CD3e (HIT3a, BD Biosciences) and anti-CD28 (BD Biosciences), or with 20-40 ng.ml 21 phorbol 12-myristate 13-acetate (PMA, Sigma) and 300 ng.ml 21 ionomycin (Calbiochem). Carboxyfluorescein Succinimydyl Ester (CFSE) and Brefeldin A were purchased from Sigma, and the calcium-sensitive dye Fluo-4 was from Invitrogen.

Cells were washed twice with PBS 1X and lysed with 50 mM Tris pH 7.4, 150 mM NaCl, 1% Triton X-100, 1% Igepal, 2 mM EDTA, supplemented with complete protease inhibitors (Roche). Lysates were cleared by a centrifugation at 10,000g at 4 o C, and protein concentration determined (micro BCA kit, Pierce). Samples were resolved on 5-20% SDS-PAGE gels and transferred to nitrocellulose membranes (Amersham). For Immunoprecipitations, samples were precleared with protein G-sepharose beads (Roche) for 30 min prior to overnight incubation with antibodies and additional protein G-sepharose beads at 4uC, as previously described [23] . Antibodies to BCL10 (A-6), IkBa (C-21), MALT1 (B-12), Tubulin (TU-02), PALS1 (H-250), SCRIB (C-6), PAR6 (G-9) and p65 (C-20) were purchased from Santa Cruz. Phosphospecific antibodies against IkBa, ERK, p38, and antibodies to CARMA1 and to ERK were from Cell Signaling Technologies. Anti-phosphorylated Tyrosine (4G10, Millipore), anti-GAPDH (Sigma), and Immobilon (Millipore) chemiluminescent substrates were also used.

Firefly luciferase constructs downstream of promoters for NF-kB or NF-AT were co-transfected with renilla luciferase pRL-TK (Int -) plasmid (Promega). Luciferase activities were analyzed using the Dual-Luciferase Kit (Promega), with firefly fluorescence units normalized to renilla luciferase fluorescence units (BMG microplate reader). were purified from blood on Ficoll-isopaque gradients. PBL were nucleofected with the Nucleofactor system and T cell solution (Amaxa, program U14), and left for three days in culture medium prior treatment.

Nuclear protein extraction and electrophoretic mobility shift assay (EMSA) 4 mg of nuclear extracts from Jurkat cells were examined for NF-kB-and Oct1-binding activity by electromobility shift assay (Panomics kit). Samples were resolved on a 6% native polyacrylamide DNA retardation gel in 0.5X TBE buffer and analyzed using a FUJI LA4000 system. 

Cells were left for 10 min on poly-lysine coated slides (Thermo Scientific) prior fixation with PBS1X containing 4% paraformaldehyde. For TCR crosslinking experiments, cells were incubated with 5 mg.ml 21 anti-CD3 at 4uC for 15 min. After two washes, cells were incubated with 5 mg.ml 21 of goat antimouse (Jackson) for 20 min either at 4uC or 37uC. To disassemble Golgi apparatus, cells were treated with 10 mg.ml 21 Brefeldin A for 60 min. Samples were permeabilized with 0.05% Triton-X100 in PBS1X for 5 min, and non-specific sites blocked with 10% FCS in PBS1X. Antibodies used were: PALS1 (Millipore), 58K Golgi (Abcam), Alexa-488 conjugated goat anti-rabbit IgG or Alexa-594 conjugated goat anti-mouse IgG (Invitrogen). Samples were analyzed using a Leica confocal microscope SP6.

Cells were incubated for 30 min at 4uC with FITC-and PEconjugated antibodies against CD25 and CD69 (ImmunoTools) and the respective isotype controls in PBS containing 0.5% BSA. After one wash with ice-cold PBS-BSA, cells were analyzed by flow cytometry with a FACSCalibur (BD Biosciences). 

Crosstalk between small GTPases and polarity proteins in cell polarization

Polarity complex proteins

Direct interaction of two polarity complexes implicated in epithelial tight junction assembly

A network of PDZ-containing proteins regulates T cell polarity and morphology during migration and immunological synapse formation

Cutting edge: Atypical PKCs regulate T lymphocyte polarity and scanning behavior

Upsides and downsides to polarity and asymmetric cell division in leukemia

The PDZ domain-binding motif of the human T cell leukemia virus type 1 tax protein induces mislocalization of the tumor suppressor hScrib in T cells

Maintenance and modulation of T cell polarity

DLGH1 is a negative regulator of T-lymphocyte proliferation

Scaffold protein Dlgh1 coordinates alternative p38 kinase activation, directing T cell receptor signals toward NFAT but not NF-kappaB transcription factors

Molecular cloning and characterization of Pals, proteins associated with mLin-7

The Maguk protein, Pals1, functions as an adapter, linking mammalian homologues of Crumbs and Discs Lost

Membrane-associated guanylate kinases regulate adhesion and plasticity at cell junctions

CARMA1-mediated NF-kappaB and JNK activation in lymphocytes

Antigen receptor signaling to NF-kappaB via CARMA1, BCL10, and MALT1

Interplay between BCL10, MALT1 and IkappaBalpha during T-cell-receptormediated NFkappaB activation

Dlgh1 coordinates actin polymerization, synaptic T cell receptor and lipid raft aggregation, and effector function in T cells

Discs large (Dlg1) complexes in lymphocyte activation

The SARS coronavirus E protein interacts with PALS1 and alters tight junction formation and epithelial morphogenesis

PATJ regulates tight junction formation and polarity in mammalian epithelial cells

The Crumbs3-Pals1 complex participates in the establishment of polarity in mammalian epithelial cells

Malt1 ubiquitination triggers NF-kappaB signaling upon T-cell activation

Casein kinase 1alpha governs antigen-receptor-induced NF-kappaB activation and human lymphoma cell survival

We thank D. Arnoult and J. Gavard for helpful discussions and critical reading of the manuscript.

Conceived and designed the experiments: GC NB. Performed the experiments: GC CD KP EH NB. Analyzed the data: GC CD KP EH AV NB. Wrote the paper: GC NB.