Although the immune adaptor adhesion and degranulation-promoting adaptor protein (ADAP) acts as a key mediator of integrin inside-out signaling leading to T cell adhesion, the regulation of this adaptor during integrin activation and clustering remains unclear. We now identify Ubc9, the sole small ubiquitin-related modifier E2 conjugase, as an essential regulator of ADAP where it is required for TCR-induced membrane recruitment of the small GTPase Rap1 and its effector protein RapL and for activation of the small GTPase Rac1 in T cell adhesion. We show that Ubc9 interacted directly with ADAP in vitro and in vivo, and the association was increased in response to anti-CD3 stimulation. The Ubc9-binding domain on ADAP was mapped to a nuclear localization sequence (aa 674–700) within ADAP. Knockdown of Ubc9 by short hairpin RNA or expression of the Ubc9-binding–deficient ADAP mutant significantly decreased TCR-induced integrin adhesion to ICAM-1 and fibronectin, as well as LFA-1 clustering, although it had little effect on the TCR proximal signaling responses and TCR-induced IL-2 transcription. Furthermore, downregulation of Ubc9 impaired TCR-mediated Rac1 activation and attenuated the membrane targeting of Rap1 and RapL, but not Rap1-interacting adaptor molecule. Taken together, our data demonstrate for the first time, to our knowledge, that Ubc9 acts as a functional binding partner of ADAP and plays a selective role in integrin-mediated T cell adhesion via modulation of Rap1-RapL membrane recruitment and Rac1 activation.

T cell adhesion is mediated by integrins on the surface of T cells binding to integrin ligands on APCs or the endothelium. T cell adhesion is essential for T cell functions; it permits T cells to transmigrate across endothelium and enter into inflamed tissue to form immunological synapses with APCs and carry out immune functions including cytokine production and T cell cytotoxicity. T cell adhesion also enables the T cell to remain bound to the APC long enough for the T cell to become activated, increase TCR sensitivity to the MHC, and lower the threshold for T cell activation (13). Regulation of T cell adhesion is achieved by controlling the activity of integrins on the cell surface.

The major integrin receptors on T cells that mediate T cell adhesion are β2 integrins of LFA-1. LFA-1 is a heterodimeric transmembrane receptor consisting of a unique α subunit (αL; CD11a) and a β2 subunit (CD18), which are receptors for ICAMs on APCs (1). In addition, VLA-4 of the β1 integrin family is another group of T cell integrin receptors for T cell adhesion. VLA-4 is composed of an ⍺4 subunit (CD49d) and a β1 subunit (CD29) and its ligands are fibronectin or VCAM-1 on APCs (4). Integrins need to be converted to an active state to bind their ligands. A conformational change is induced that augments affinity for integrin ligand followed by clustering of these receptors on the surface of T cells, which enhances avidity for ligand binding (5). The molecular events leading to integrin activation have collectively been termed “inside-out” signaling (6), which involves an array of signaling molecules including the adaptor proteins SH2 domain-containing leukocyte protein of 76 kDa (SLP-76), adhesion and degranulation-promoting adaptor protein (ADAP), and Src kinase–associated phosphoprotein 1 (SKAP1) (711). These adaptor proteins lack enzymatic activity, but instead, carry binding modules or sites for the assembly of supra-molecular complexes that are critical for the inside-out adhesion pathway (2).

ADAP, a key mediator in the inside-out signaling of T cells, contains domains for scaffolding molecules of various classes, including a proline-rich region for binding to SKAP1, tyrosines (Y) in the two YDDV sites that can be phosphorylated and mediate SLP-76 binding, an E/K-rich region for binding to caspase recruitment domain–containing membrane-associated guanylate kinase protein 1 (CARMA1), and an FPPPP sequence for binding the Ena/vasodilator-stimulated phosphoprotein (VASP) homology 1 (EVH1) domain of VASP (2, 1214). The scaffolding function of ADAP for effectors SKAP1 and SLP-76 is necessary to support TCR-dependent integrin-mediated adhesion (15, 16). Abrogating the interaction of ADAP with SKAP1 or SLP-76 reduces TCR-mediated integrin activation and adhesion, as well as impairs immunological synapse formation and T cell–APC conjugation (1518). Besides, ADAP consists of two lysine/glutamic acid/arginine-rich clusters, which span residues 469–505, and residues 674–700, respectively. These two clusters were first proposed as the putative nuclear localization sequences (NLSs) by two independent groups in 1997 (13, 14) due to the sequence similarities to the bipartite nuclear localization motifs KR/RR-X11–12-KK/RK found in nuclear proteins such as p53, c-Rel, c-jun, and RNA polymerase I (19, 20). However, thus far, it has remained unknown about the functions of these two putative NLSs in ADAP. In particular, it has not been clear about the role of these two putative NLS motifs in the ADAP-mediated T cell adhesion and signaling.

ADAP constitutively associates with and stabilizes SKAP1 (21). The downstream effector molecules of the ADAP-SKAP1 module include the small GTPase ras-related protein 1 (Rap1) and its two effector proteins, regulator of adhesion and cell polarization enriched in lymphoid tissues (RapL) and Rap1-interacting adaptor molecule (RIAM), in the TCR-mediated regulation of integrin function and T cell adhesion (5). T cell activation induces the formation of a complex of activated Rap1 and RIAM, which is brought to the plasma membrane in close proximity to the β-chain of LFA-1, and is required to mediate binding of LFA-1 to talin, thereafter facilitating the activation of LFA-1 (22). The ADAP-SKAP1 module is recruited to the plasma membrane following TCR activation (16), which is required for the plasma membrane recruitment of Rap1 and RIAM in TCR-mediated T cell adhesion. T cells from ADAP−/− mice exhibited deficient TCR-mediated upregulation of adhesion (8, 18). In addition to its central role in TCR-mediated adhesion, ADAP also participates in the signaling pathways downstream of the TCR that result in T cell activation (8, 18). ADAP-deficient T cells displayed reduced IL-2 production and cell proliferation (8, 18).

The transfer of activated small ubiquitin-related modifier (SUMO) to the target proteins is mediated by Ubc9, the sole E2-conjugating enzyme of the SUMOylation cycle (23). Ubc9-mediated SUMOylation has been implicated in many processes, such as DNA replication/repair, cell division, movement, nuclear transport, and transcription (24, 25). In addition, Ubc9 functions as a cellular chaperone and transcriptional coregulator, which is SUMOylation-independent (2628). To date, a link between SUMOylation and T cell functions has remained unclear. In the current study, we determined that Ubc9 has a role in TCR-mediated inside-out T cell adhesion. We show that Ubc9 downregulation impaired the TCR-mediated integrin adhesion and LFA-1 clustering, whereas the CD3-induced IL-2 transcription was not affected. Mechanistically, Ubc9 interacts directly with ADAP, a major mediator of inside-out signaling in T cell adhesion, and is essential for plasma membrane recruitment of Rap1 and RapL, but not RIAM, and for the Rac1 activation upon TCR stimulation. Our data have identified for the first time, to our knowledge, that the SUMO E2 conjugase Ubc9 regulates integrin-mediated T cell adhesion through an interaction with the immune adaptor ADAP.

Jurkat T cells and murine T cell hybridoma DC27.10 and 2B4 cells were maintained in RPMI 1640 medium supplemented with 5% (v/v) FBS (Sigma) and 100 U/ml penicillin/streptomycin (29). Total CD4+ T cells were enriched from primary T cells from 6 to 8-wk-old female C57BL/6J mice (Shanghai Laboratory Animal Center, Shanghai, China) with mouse CD4+ enrichment selection column (R&D System) according to the manufacturer’s manual. The purified primary T cells were cultured in RPMI 1640 medium with 10% (v/v) FBS, 100 U/ml penicillin/streptomycin, and 50 μM 2-ME. Primary cells were activated for 2 d by plate-bound 2 μg/ml anti-CD3 (2C11; eBioscience) and 1 μg/ml anti-CD28 (eBioscience) before experimental use. HEK 293T cells and COS-7 cells were maintained in DMEM supplemented with 5% (v/v) FBS and 100 U/ml penicillin/streptomycin. Reagents and Abs were obtained from the following sources: anti-human CD3 OKT3 (American Type Culture Collection); anti-human CD28 (Sigma); anti-ADAP and anti-phosphotyrosine, clone 4G10 (Millipore); anti-Ubc9 (Santa Cruz Biotechnology); anti-hemagglutinin (anti-HA) (Sigma); anti-SKAP1 (Transduction Laboratories); anti–SLP-76 (Cell Signaling Technology); anti-Rap1 (BD Biosciences); anti–α-tubulin, anti-RIAM (Abcam); anti-CD18 (Proteintech); recombinant human ICAM-1 Fc Chimera (R&D Systems and Sino Biological); fibronectin (Sigma). The isolated NLS2 domain with the amino acid sequence ((KAKTEEKDLKKLKKQEKEEKDFRKKFK) was synthesized by Botech Shanghai.

Full-length human ADAP and SKAP1 cDNAs were cloned into pSRα expression vector containing an influenza HA epitope tag at the N terminus. The HA-tagged ADAP dNLS1 mutant was generated by Quikchange-mediated deletion of a sequence coding for aa 469–505 using pSRα-HA-ADAP wild-type (WT) plasmid as a template. The HA-tagged ADAP dNLS2 mutant was generated by a Quikchange-mediated deletion of a sequence coding for aa 674–700. Ubc9 was expressed in a pcDNA3.1 expression vector. For the short hairpin RNA (shRNA) rescue experiment, a shRNA-resistance Ubc9 (Ubc9*) was generated by introducing nine synonymous point mutations into the cDNA sequence that is targeted by Ubc9 shRNA without changing the amino acid sequence using Quikchange mutagenesis kit. For transient transfection, Jurkat T cells were transfected by electroporation (Biorad Gene Pulser Xcell), using 250 V, 800 microfarad. Sequences targeted against either Ubc9 or EGFP as a negative control were cloned into pLVX-shRNA1 expression vector using BamHI and EcoRI sites. Target sequences for shRNAs against Ubc9 are as follows: shUbc9#1 (5′-GAAGUUUGCGCCCUCAUAA-3′) and shUbc9#2 (5′-GGAACUUCUAAAUGAACCA-3′) (28, 30). Lentiviral plasmids containing shRNA against Ubc9 or EGFP were freshly prepared in 293FT cells (Invitrogen). Lentivirus was concentrated from filtered culture media (0.45 μm filters) by ultracentrifugation at 25,000 rpm for 120 min, and stored at −80°C before use. Jurkat T cells and primary T cells were infected by spinoculation using concentrated lentiviral supernatants. Both the stable Ubc9-knockdown Jurkat T cells and the control nontargeting shRNA against EGFP were in parallel selected by puromycin for more than 3 wk.

Prior to the adhesion assay, an equal number of stable cells or freshly transfected cells were stimulated with anti-CD3 or left unstimulated for 30 min. The wells of a flat-bottom 96-well plate were coated with either 10 μg/ml ICAM-1 human/mouse Fc chimera or 30 μg/ml fibronectin and the cells were incubated for 1 h at 37°C. Nonadherent cells were gently washed off, and the bound cells were counted as previously described (15).

Cells were incubated with fluorescence-labeled Abs (BD Biosciences) for 1 h at room temperature in FACS buffer (0.5% BSA in PBS), washed three times with PBS, and resuspended into FACS buffer. Analyses were performed using a FACSCalibur flow cytometer (BD Biosciences) as described before (29). For the intracellular Ca2+ measurement, cells were loaded with 5 μM of Indo-1-AM (Thermo Fisher) in RPMI 1640 and incubated for 1 h at 37°C. After incubation, the cells were washed and resuspended in RPMI 1640. Cells were maintained at room temperature for 30 min and prewarmed at 37°C for at least 5 min before measurement. Baseline measurements were collected for 30 s, 4 μg/ml OKT3 or 1 μg/ml ionomycin was added, and Ca2+ measurement was continued for an additional 3 min using LSRFortessa X-30 cytometer (BD Biosciences). The intracellular Ca2+ concentration was determined over time by monitoring the fluorescence emission ratio of the Ca2+-bound versus -free form of indo-1 at 395 and 525 nm, respectively. All FACS data were processed and analyzed using FlowJo software (Tree Star).

Immunofluorescence microscopy was conducted as described previously (29). Cells were seeded on poly-l-lysine (Sigma) -coated slides, fixed in 4% paraformaldehyde for 15 min, permeabilized with 90% methanol for 30 min, and blocked with 5% BSA for 1 h. Cells were subsequently stained with the indicated primary Abs in blocking buffer overnight and visualized with Alexa Fluor 488-conjugated goat anti-mouse or Alexa Fluor 555-conjugated donkey anti-rabbit Abs (Cell Signaling Technology) correspondingly. For LFA-1 clustering, cells were incubated with anti-CD3 (2 μg/106 cells) at 37°C for 30 min and surface-bound Ab was removed by incubation on ice for 3 min in a PBS solution acidified to pH 2 with HCl and supplemented with 0.03 M sucrose and 10% FBS (31, 32). Cells were then fixed and blocked in 5% BSA followed by incubation with anti-CD11a (1 μg/ml; BD Biosciences) and Alexa Fluor 488-conjugated goat anti-mouse Ab (Cell Signaling Technology). Cells were counterstained with DAPI for 10 min and mounted on microscopy slides. Samples were visualized on a Nikon Eclipse Ni-U microscope. For each experiment, a minimum of 200 cells from each condition were imaged and analyzed for LFA-1 clustering. Anti–LFA-1 clustering was defined by the presence of a discrete polarized cap at one end of the cell (15). For the imaging of Rap1 and LFA-1 in cells, sequential immunofluorescence staining was performed using anti-Rap1 (BD Biosciences) and anti–LFA-1 (Proteintech) followed by incubation with fluorescence-labeled secondary Abs as described above. Cells were imaged with a Nikon Eclipse Ti laser scanning confocal system. The mean intensity of green fluorescence of Rap1 at the plasma membrane or the total amount per cell was determined using ImageJ software. For the imaging of HA-tagged proteins, cells were imaged with a ZEISS LSM 880 Confocal Laser Scanning Microscope.

Cell lysis, immunoprecipitation, and detection were performed as described previously with minor modifications (33). Briefly, 5 × 106 cells were lysed with 200 μl of lysis buffer (1% Triton X-100 [v/v] in 20 mM Tris-HCl [pH 8.3], 150 mM NaCl, 1 mM Na4VO3, and 0.1% protease inhibitor mixture solution [Roche]). For immunoprecipitation, cell lysate was incubated with the indicated Abs at 4°C overnight before conjugating with 20 μl protein G Sepharose beads (Amersham) at 4°C for 1 h. Bound beads were washed three times with lysis buffer and precipitates were dissociated from beads by boiling in sample buffer for 10 min. Proteins were separated by SDS-PAGE and transferred to nitrocellulose membranes. Membranes were blocked with 5% BSA and incubated with indicated Abs at 4°C overnight. Bound Abs were visualized using IRDye secondary Abs (LI-COR) followed by detection with Odyssey Imaging Systems (LI-COR).

Isolation of cytosolic and plasma membrane fractions has been previously described (16, 34). Briefly, freshly harvested lymphocytes from C57BL/6 mice were enriched with CD4+ enrichment column and infected with lentiviruses expressing shRNAs against Ubc9 or EGFP, and cultured ex vivo for 2 d. Cells were either left unstimulated or stimulated with 4 μg/ml anti-CD3 (2C11) for 10 min, and were washed in ice-cold PBS and resuspended on ice in a hypotonic buffer. Cells were sheared and lysates were centrifuged at low speed to precipitate nuclei. The remaining supernatant was recentrifuged, and the supernatant (cytosolic fraction) was collected. The pellet (membrane fraction) was washed twice with hypotonic buffer and resuspended on ice in lysis buffer containing 1% NP-40. The protein concentrations of the cytosolic and membrane fractions were determined by the Bradford assay (Sigma), and an equal amount of protein from each fraction was analyzed by Western blotting.

The Rac1 pull-down activation assay kit (Cytoskeleton) was used to measure endogenous Rac1 GTPase activity. The assay uses the Cdc42/Rac interactive binding region (also called the p21 binding domain, PBD) of the Cdc42/Rac effector protein, p21-activated kinase 1 (PAK). The PAK-PBD is in the form of a GST fusion protein, and this allows Rac-GTP to be precipitated from the cell lysate by binding to the PAK-PBD domain fused to glutathione affinity beads. Either resting or stimulated (4 μg/ml OKT3 for 60 min) control shEGFP cells or stable Ubc9-knockdown T cells (3 × 107) were washed with ice-cold PBS and lysed, and equal amounts of cell lysates were processed according to the manufacturer’s protocol. The GTP-bound Rac1 protein and total Rac1 protein (10 μg) were detected by Western blotting using an anti-Rac1 Ab (Cytoskeleton).

Cell lysates and precipitates were prepared as described in immunoprecipitation and immunoblotting and resolved in a 12% SDS polyacrylamide gel. The gel was subjected to GelCode Blue staining (Thermo Scientific) according to the manufacturer’s instructions. Proteins of interest were excised and subjected to reduction, alkylation, in-gel digestion, and extraction according to standard protocols (35). Peptides were desalted with ZipTip C18 pipette tips (Millipore) before analysis by mass spectrometry. Samples were loaded and eluted into an EASY-nLC 1000 system coupled online to a linear trap quadrupole Orbitrap Elite (Thermo Fisher Scientific) through an Acclaim PepMap 100 Column (75 μm × 15 cm, C18, 3 μm, 100 Å; Thermo Fisher Scientific) with a reversed-phase binary gradient (Solvent A: 0.1% formic acid; Solvent B: 0.01% formic acid, 100% acetonitrile in 2 h). Results were evaluated with Mascot software (Matrix Science). Mass spectrometry spectra were searched against the human UniProt nonredundant protein database using Mascot (Matrix Science). The Mascot database search was performed using the following parameters: trypsin enzyme specificity, one possible missed cleavage, 10 ppm mass tolerance for peptide ions, and 0.5 Da mass tolerance for fragment ions. Search parameters specified a differential modification of methionine oxidation, (+15.9949 Da) as well as a static modification of carbamidomethylation (+57.0215 Da) on cysteine. To provide high confidence sequence assignments, Mascot results were filtered by a cut-off value <0.05.

Jurkat T cells (5 × 106) were transfected with 5–10 μg of plasmid constructs and in combination with 4 μg of luciferase-driven IL-2 promoter as described previously (36). Cells were stimulated at 37°C with 2 μg/ml anti-CD3 mAb (OKT3) or left unstimulated for 6 h after transfection. After stimulation, cells were lysed in 50 μl of lysis buffer (Promega kit). Luciferase activity was subsequently determined using a microplate reader (PHERAstar FS; BMG Labtech). Luciferase units of the experimental vector were normalized to the level of control vectors in each sample.

All data were analyzed with Prism 7 (GraphPad Software). Differences between two group means were analyzed using an unpaired Student t test or the Mann–Whitney U test for nonparametric data. A one-way or two-way ANOVA followed by correction for multiple comparisons was used to compare more than two groups. Wherever indicated, p values are as follows: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Data are shown as mean ± SE.

Although the posttranscriptional modification with SUMO, namely SUMOylation, is one of the important mechanisms that regulates a series of key cellular processes, a role for SUMOylation in T cell functions has remained unclear. The binding of integrin molecules on the surface of T cells to ICAM-1 (e.g., β2 integrin LFA-1) and fibronectin (e.g., β1 integrin VLA-4) facilitates T cell adhesion and enables T cells to form firm conjugates with APCs (6). Ubc9 is the sole SUMO E2 conjugase, and thus depletion of Ubc9 leads to the switching off of SUMO pathway activity (3739). To examine the functional role of the SUMO pathway in TCR-mediated T cell adhesion, we initially assessed the effect of Ubc9 knockdown on β1 and β2 integrin–mediated T cell adhesion and LFA-1 clustering. Stable Ubc9-knockdown Jurkat T cells and control shEGFP Jurkat T cells were generated using lentiviral delivery of shRNA expression vectors for the expressions of two individual sequence-specific shRNAs against Ubc9 and shRNA against EGFP, respectively. Western blotting analysis confirmed both shRNAs targeted against Ubc9 showed efficient knockdown (80%) of Ubc9 in stable cells (right panels in Fig. 1A, 1B). The in vitro ICAM-1–binding assays were performed by plating Ubc9-knockdown cells or control shEGFP cells on ICAM-1–coated or fibronectin-coated plates. Adhesion was analyzed by counting the bound T cells on the plate as previously described (33). Stimulation with anti-CD3 increased the number of control shEGFP Jurkat T cells bound to ICAM-1–coated plates and fibronectin-coated plates by 5.7- and 1.8-fold, respectively (Fig. 1A, 1B, left panels, shEGFP, anti-CD3 versus unstimulated). By contrast, Ubc9-knockdown cells displayed a substantially attenuated binding to ICAM-1–coated plates upon anti-CD3 stimulation (Fig. 1A, left panel, shUbc9#1 and shUbc9#2 versus shEGFP). Similar effects were observed for stable expressions of shRNAs targeted to different sites of Ubc9. Meanwhile, a reduction in the number of cells binding to fibronectin-coated plate was shown in the Ubc9-knockdown cells as well (Fig. 1B, left panel). Further, similar results were also found using mouse primary T cells. Ubc9 was knocked down in primary T cells by lentiviral-shRNA infections. Although the primary T cells infected with shEGFP showed a 2–3-fold increase in adhesion to ICAM-1–coated or fibronectin-coated plates after anti-CD3 stimulation (Fig. 1C, 1D, shEGFP, anti-CD3 versus unstimulated), a marked decrease in adhesion to both ICAM-1–coated and fibronectin-coated plates was apparent in shUbc9#2-infected cells upon anti-CD3 stimulation (Fig. 1C, 1D, shUbc9#2 versus shEGFP). The specific effect of Ubc9 knockdown in T cell adhesion was further supported by the observation of a dose-dependent effect of Ubc9 knockdown on T cell adhesion to ICAM-1, and by the observation that re-expression of Ubc9 in stable Ubc9-knockdown cells restored up to 75% of the impairment of anti-CD3–induced adhesion to ICAM-1 (Supplemental Fig. 1A, 1B).

FIGURE 1.

Knockdown of Ubc9 impairs integrin-mediated T cell adhesion and LFA-1 clustering. (A and B) Stable Ubc9-knockdown Jurkat T cells (shUbc9#1 and shUbc9#2) or control shEGFP Jurkat T cells (shEGFP) were either left untreated or stimulated with anti-CD3, followed by the measurement of cell adhesion to ICAM-1–coated plates (A) and fibronectin-coated plates (B) as described in 2Materials and Methods. Error bars indicate the SE from three individual experiments for the ICAM-1–binding assay and four individual experiments for the fibronectin-binding assay. Immunoblots on the right panels show anti-Ubc9 Western blot of whole cell lysates of cells used in the ICAM-1–/fibronectin-binding assays. (C and D) Murine primary T cells infected with lentiviral-encoded shEGFP or shUbc9#2 were stimulated with anti-CD3 and assessed for adhesion to ICAM-1–coated plates (C) and fibronectin-coated plates (D). Error bars indicate the SE from three individual experiments. (E) Upper panel: Stable Ubc9-knockdown and control shEGFP Jurkat T cells were stimulated with anti-CD3 for 30 min and imaged for LFA-1 clustering as described in 2Materials and Methods. The arrow designates LFA-1 cluster; scale bars, 10 μm. Lower panel: Histogram showing the percentage of T cells with LFA-1 clustering. Error bars indicate the SE from four individual experiments. (F) Representative FACS profile showing the integrin LFA-1 α and β subunits and TCR surface expression levels in the stable Ubc9-knockdown Jurkat T cells (shUbc9#2) and in mock control T cells (shEGFP). Data are representative of at least three independent experiments. *p < 0.05, **p < 0.01.

FIGURE 1.

Knockdown of Ubc9 impairs integrin-mediated T cell adhesion and LFA-1 clustering. (A and B) Stable Ubc9-knockdown Jurkat T cells (shUbc9#1 and shUbc9#2) or control shEGFP Jurkat T cells (shEGFP) were either left untreated or stimulated with anti-CD3, followed by the measurement of cell adhesion to ICAM-1–coated plates (A) and fibronectin-coated plates (B) as described in 2Materials and Methods. Error bars indicate the SE from three individual experiments for the ICAM-1–binding assay and four individual experiments for the fibronectin-binding assay. Immunoblots on the right panels show anti-Ubc9 Western blot of whole cell lysates of cells used in the ICAM-1–/fibronectin-binding assays. (C and D) Murine primary T cells infected with lentiviral-encoded shEGFP or shUbc9#2 were stimulated with anti-CD3 and assessed for adhesion to ICAM-1–coated plates (C) and fibronectin-coated plates (D). Error bars indicate the SE from three individual experiments. (E) Upper panel: Stable Ubc9-knockdown and control shEGFP Jurkat T cells were stimulated with anti-CD3 for 30 min and imaged for LFA-1 clustering as described in 2Materials and Methods. The arrow designates LFA-1 cluster; scale bars, 10 μm. Lower panel: Histogram showing the percentage of T cells with LFA-1 clustering. Error bars indicate the SE from four individual experiments. (F) Representative FACS profile showing the integrin LFA-1 α and β subunits and TCR surface expression levels in the stable Ubc9-knockdown Jurkat T cells (shUbc9#2) and in mock control T cells (shEGFP). Data are representative of at least three independent experiments. *p < 0.05, **p < 0.01.

Close modal

Integrin, especially LFA-1, forms clusters or accumulates at one side of the cell to increase its avidity for ICAM-1 binding (40). Thus, we examined whether the impaired anti-CD3–induced adhesion of Ubc9-knockdown T cells to ICAM-1 and fibronectin was also accompanied by a reduction in LFA-1 clustering on the surface of T cells. In the control shEGFP Jurkat T cells, anti-CD3 stimulation resulted in a 4-fold increase in the number of cells with LFA-1 clusters (Fig. 1E, shEGFP, anti-CD3 versus unstimulated, from 5 to 20%), whereas fewer Ubc9-knockdown cells displayed clustering after parallel stimulation with anti-CD3 (Fig. 1E, shUbc9#2, anti-CD3 versus unstimulated, from 4 to 10%). Examples of cells with LFA-1 clustering were shown in the upper panel of Fig. 1E. The defect in TCR-mediated T cell adhesion to ICAM-1 and fibronectin, and LFA-1 clustering in Ubc9-knockdown T cells was not due to impaired expressions of the LFA-1 (α and β subunit) or the TCR as determined by flow cytometry analysis (Fig. 1F). In addition, the impaired T cell adhesion and integrin activation were not due to the altered expression levels of SKAP1, Rap1, RIAM, or RapL in stable Ubc9-knockdown Jurkat T cells (Supplemental Fig. 1C). Taken together, these data indicated that Ubc9 is needed for TCR-induced integrin-mediated T cell adhesion and LFA-1 clustering.

Cytosolic adaptor proteins ADAP and SKAP1 have been implicated in integrin-mediated T cell adhesion and identified as crucial mediators for TCR-mediated inside-out signaling (7, 15, 33). Because we showed here that Ubc9 is required for integrin-mediated T cell adhesion, we hypothesized that there may be a functional interaction between Ubc9 and the adhesion-related adaptors ADAP and SKAP1. To test this, we performed coimmunoprecipitation experiments in mouse T cell hybridoma DC27.10 cells with or without anti-CD3 stimulation. Western blot analysis of immunoprecipitates with anti-Ubc9 Abs revealed that Ubc9, at ∼18-kDa, was coimmunoprecipitated with ADAP (Fig. 2A, lanes 4 and 5) and with SKAP1 (Fig. 2A, lanes 6 and 7). However, in comparison, the amount of Ubc9 was more abundant in the anti-ADAP precipitates than in the anti-SKAP1 precipitates (lanes 4 and 5 versus lanes 6 and 7 in Fig. 2A). As a positive control (41), Ubc9 was readily detected in the anti-RanGAP1 immunoprecipitates (Fig. 2A, lane 8). Importantly, anti-CD3 stimulation significantly increased the level of Ubc9 in the precipitations (lane 5 versus lane 4 and lane 7 versus lane 6).

FIGURE 2.

Ubc9 binds directly to the immune adaptor ADAP but not SKAP1 in vivo and in vitro. (A) Left panel: Mouse hybridoma DC27.10 T cells were preactivated with anti-CD3 (2C11) for 5 min and prepared for immunoprecipitation with either IgG control, anti-ADAP, anti-SKAP1, or anti-RanGAP1 followed by blotting with anti-ADAP (top) and anti-Ubc9 Ab (bottom). Right panel: Relative bound Ubc9 to each Ab was calculated as the band intensity by ImageJ and was plotted as fold difference relative to the anti-CD3–stimulated IgG control. (B) Upper panel: The interaction of Ubc9 with ADAP in primary murine T cells. Primary murine CD4+ T cells were stimulated with anti-CD3 for the indicated time periods. Endogenous ADAP was immunoprecipitated with anti-ADAP Ab or IgG control, followed by immunoblotting with Abs against ADAP or Ubc9. Lower panel: Time course of Ubc9 binding to ADAP following anti-CD3 stimulation. 2B4 cells were stimulated with anti-CD3 for the indicated time periods. Endogenous ADAP was immunoprecipitated with anti-ADAP Ab and assayed for interaction with Ubc9 by immunoblotting with Abs against ADAP or Ubc9. (C) HEK 293T cells were transfected with Ubc9 alone, or together with either HA-ADAP, HA-SKAP1, HA-ADAP dNLS2, or HA-RanGAP1 constructs. Immunoprecipitation was performed using anti-HA Ab, followed by blotting with anti-HA (top) or anti-Ubc9 Ab (bottom). The black lines indicate where parts of the image were joined. (D) Detection of Ubc9 in anti-ADAP immunoprecipitates using mass spectrometry analysis. Left panel: Coomassie brilliant blue staining analysis of anti-CD3–stimulated Jurkat T cell lysate prepared for immunoprecipitation with IgG control and anti-ADAP. Bands indicated with rectangle were excised and subjected to liquid chromatography tandem mass spectrometry analysis as described in 2Materials and Methods. Right panel: List of detected peptides assigned to Ubc9 and their abundance. (E) The putative NLS domains in ADAP. Schematic drawing of domain structure of ADAP and the ADAP NLS1 deletion mutant (ADAP dNLS1), and NLS2 deletion mutant (ADAP dNLS2) constructs (upper panel). The corresponding sequences of both NLSs are listed and compared (lower panel). NLS1 and NLS2 share an identical lysine-containing sequence of KKFK (in bold), and both have several positively charged residues as KR/KK that resemble the bipartite nuclear localization motif (in shaded). A unique 5-aa motif of KKLKK (in bold and underlined) is present in NLS2 but not NLS1, which is also found in the Ubc9-interacting NLSs of AR and Vsx-1 proteins (26, 50). (F) HEK 293T cells were cotransfected with Ubc9 and vector control, ADAP WT, ADAP dNLS1, or ADAP dNLS2, followed by precipitation with anti-HA Ab and immunoblotting with anti-Ubc9 (top) and anti-ADAP Ab (bottom).

FIGURE 2.

Ubc9 binds directly to the immune adaptor ADAP but not SKAP1 in vivo and in vitro. (A) Left panel: Mouse hybridoma DC27.10 T cells were preactivated with anti-CD3 (2C11) for 5 min and prepared for immunoprecipitation with either IgG control, anti-ADAP, anti-SKAP1, or anti-RanGAP1 followed by blotting with anti-ADAP (top) and anti-Ubc9 Ab (bottom). Right panel: Relative bound Ubc9 to each Ab was calculated as the band intensity by ImageJ and was plotted as fold difference relative to the anti-CD3–stimulated IgG control. (B) Upper panel: The interaction of Ubc9 with ADAP in primary murine T cells. Primary murine CD4+ T cells were stimulated with anti-CD3 for the indicated time periods. Endogenous ADAP was immunoprecipitated with anti-ADAP Ab or IgG control, followed by immunoblotting with Abs against ADAP or Ubc9. Lower panel: Time course of Ubc9 binding to ADAP following anti-CD3 stimulation. 2B4 cells were stimulated with anti-CD3 for the indicated time periods. Endogenous ADAP was immunoprecipitated with anti-ADAP Ab and assayed for interaction with Ubc9 by immunoblotting with Abs against ADAP or Ubc9. (C) HEK 293T cells were transfected with Ubc9 alone, or together with either HA-ADAP, HA-SKAP1, HA-ADAP dNLS2, or HA-RanGAP1 constructs. Immunoprecipitation was performed using anti-HA Ab, followed by blotting with anti-HA (top) or anti-Ubc9 Ab (bottom). The black lines indicate where parts of the image were joined. (D) Detection of Ubc9 in anti-ADAP immunoprecipitates using mass spectrometry analysis. Left panel: Coomassie brilliant blue staining analysis of anti-CD3–stimulated Jurkat T cell lysate prepared for immunoprecipitation with IgG control and anti-ADAP. Bands indicated with rectangle were excised and subjected to liquid chromatography tandem mass spectrometry analysis as described in 2Materials and Methods. Right panel: List of detected peptides assigned to Ubc9 and their abundance. (E) The putative NLS domains in ADAP. Schematic drawing of domain structure of ADAP and the ADAP NLS1 deletion mutant (ADAP dNLS1), and NLS2 deletion mutant (ADAP dNLS2) constructs (upper panel). The corresponding sequences of both NLSs are listed and compared (lower panel). NLS1 and NLS2 share an identical lysine-containing sequence of KKFK (in bold), and both have several positively charged residues as KR/KK that resemble the bipartite nuclear localization motif (in shaded). A unique 5-aa motif of KKLKK (in bold and underlined) is present in NLS2 but not NLS1, which is also found in the Ubc9-interacting NLSs of AR and Vsx-1 proteins (26, 50). (F) HEK 293T cells were cotransfected with Ubc9 and vector control, ADAP WT, ADAP dNLS1, or ADAP dNLS2, followed by precipitation with anti-HA Ab and immunoblotting with anti-Ubc9 (top) and anti-ADAP Ab (bottom).

Close modal

The effect of anti-CD3 stimulation on the binding of Ubc9 to ADAP was also assessed using 2B4 cells stimulated with anti-CD3 over a time course of 30 min. Endogenous ADAP was immunoprecipitated with anti-ADAP Ab and assayed for interaction with Ubc9 by immunoblotting with Abs against ADAP and Ubc9. Anti-CD3 stimulation induced the coprecipitation of Ubc9 with ADAP, which peaked at 10 min after anti-CD3 stimulation (Fig. 2B, lower panel). Interestingly, the binding of Ubc9 to ADAP declined at 15 min after anti-CD3 stimulation (Fig. 2B, lower panel, lane 4 versus lane 3). The interaction of ADAP and Ubc9 was also detected in primary murine CD4+ T cells and anti-CD3 ligation enhanced the Ubc9-ADAP binding (Fig. 2B, upper panel). These data indicate that Ubc9 binds to the immune adaptors ADAP and SKAP1, and anti-CD3 stimulation promotes their interaction in T cells.

To determine whether the binding of Ubc9 to ADAP and/or SKAP1 is direct or indirect, we carried out immunoprecipitation assays in HEK 293T cells in which endogenous ADAP and SKAP1 are both absent. Ubc9 was coexpressed with either HA-tagged ADAP or SKAP1 proteins in HEK 293T cells, followed by anti-HA precipitation and anti-Ubc9 blotting. Ubc9 was readily coprecipitated with HA-tagged ADAP (Fig. 2C, lane 3). The presence of Ubc9 in anti-ADAP immunoprecipitates was further confirmed by liquid chromatography tandem mass spectrometry analysis. Coomassie brilliant blue staining identified proteins that were coprecipitated by anti-ADAP. Six peptides corresponding to Ubc9 were detected (Fig. 2D). However, Ubc9 failed to coimmunoprecipitate with HA-tagged SKAP1 in HEK 293T cells (Fig. 2C, lane 4), suggesting that the coimmunoprecipitation of Ubc9 with SKAP1 in Jurkat T cells was indirect (Fig. 2A, lanes 6 and 7) and likely mediated via ADAP. Thus, we concluded that Ubc9 interacts directly with ADAP but not SKAP1 in the ADAP-SKAP1 module.

Next, we asked which site/region in ADAP contributes to its interaction with Ubc9. ADAP consists of a proline-rich domain, multiple tyrosine phosphorylation sites, two SH3 domains, an EVH1 domain binding site, and two putative NLSs (2, 13, 14). We first constructed different HA-tagged truncated deletions of ADAP (Fig. 2E). Ubc9 was then coexpressed with either HA-tagged ADAP WT or its deletion mutants in HEK 293T cells, followed by anti-HA immunoprecipitation and immunoblotting with anti-Ubc9. WT ADAP coprecipitated with Ubc9 (Fig. 2F, lane 2). Deletion of NLS1 (aa 469–505) resulted in an approximate 50% decrease in binding of Ubc9 compared with WT ADAP (lane 3 versus lane 2). By contrast, deletion of NLS2 (aa 674–700) abolished the binding of Ubc9 with ADAP (lane 4 versus lanes 2 and 3). Similar amounts of HA-tagged ADAP WT and mutant expressions were confirmed by anti-ADAP blotting (Fig. 2F, lower panel). Further, in vitro pull-down assay showed that Ubc9 was pulled down by ADAP, and addition of the synthesized isolated NLS2 domain significantly decreased the amount of Ubc9 pulled down by ADAP (Supplemental Fig. 2A, lane 2 versus lane 3). These data indicate that the NLS2 of ADAP is responsible for the binding of ADAP with Ubc9.

TCR engagement with APCs is rapidly followed by tyrosine phosphorylation of downstream substrates, including receptors, kinases, and adaptor proteins, which is critical for signal initiation and integrin inside-out signaling (42, 43). To assess whether the attenuation of TCR-induced T cell adhesion in Ubc9-knockdown cells was a result of impaired TCR signaling, we compared the level of CD3-mediated tyrosine phosphorylation between control shEGFP T cells and Ubc9-knockdown Jurkat T cells (Fig. 3A). The total tyrosine phosphorylation level of Ubc9-knockdown cells and control shEGFP cells was examined by blotting with the anti-phosphotyrosine Ab 4G10. Both control shEGFP and Ubc9-knockdown T cells responded effectively and retained the similar responsiveness to the anti-CD3 stimulation as indicated by a significant increase in the overall level of tyrosine phosphorylation upon anti-CD3 stimulation (Fig. 3A, lane 2 versus lane 1 for control shEGFP T cells and lane 4 versus lane 3 for Ubc9-knockdown cells). In addition, stimulation of Ubc9-knockdown T cells with anti-CD3 revealed no significant difference in total tyrosine phosphorylation as compared with control shEGFP T cells (Fig. 3A, lanes 3 and 4 versus lanes 1 and 2). As a control, the expression level of Ubc9 in Ubc9-knockdown cells was confirmed to be significantly reduced as shown by the anti-Ubc9 blot. We next examined the effect of Ubc9 knockdown on IL-2 transcription, a long-term parameter of T cell activation. Jurkat T cells were cotransfected with either shEGFP or shUbc9#2 constructs in combination with IL-2 promoter-driven luciferase reporter. The luciferase assay demonstrated that stimulation with anti-CD3 induced significant enhancement in IL-2 promoter activity. However, shEGFP-transfected cells and shUBC9#2-transfected cells showed a similar level of increase in IL-2 promoter activity (Fig. 3B, left panel, shEGFP versus shUBC9#2, 9-fold versus 7-fold, respectively). A similar level of increment in the transactivation of NF-κB promoter after anti-CD3 stimulation was observed (Fig. 3B, right panel, shEGFP versus shUBC9#2, 3.2-fold versus 3-fold, respectively). Further, we have also measured the effects of Ubc9 knockdown on other TCR proximal and distal signaling events, including tyrosine phosphorylation of individual molecules of ZAP-70 and PLC-γ1 (Fig. 3C), calcium mobilization (Fig. 3D), and the upregulation of CD69 (Fig. 3E), which were not significantly affected upon suppression of Ubc9.

FIGURE 3.

ADAP-Ubc9 binding fails to affect TCR proximal signaling events and TCR-induced IL-2 transcription. (A) Stable Ubc9-knockdown and control shEGFP Jurkat T cells were either left untreated or stimulated with anti-CD3 (OKT3) for 3 min. The cell lysates were extracted and subjected to Western blotting with Abs against phosphotyrosine, α-tubulin, and Ubc9. (B) Jurkat T cells were cotransfected with IL-2 (upper left panel) or NF-κB (upper right panel) promoter-driven luciferase reporter with either shEGFP or shUbc9#2 constructs and were stimulated with anti-CD3 for 6 h, followed by a measurement of luciferase activity as indicated in 2Materials and Methods. Error bars indicate the SE from three individual experiments. The knockdown of Ubc9 was assessed by anti-Ubc9 Western blotting in the lower panel. (C) Stable Ubc9-knockdown Jurkat T cells (shUbc9#2) or the mock control T cells (shEGFP) were either left unstimulated or were stimulated with anti-CD3 for 5 min. Whole cell lysate was subjected to immunoprecipitation with anti–ZAP-70 (left panel) and anti–PLC-γ1 (right panel), followed by Western blotting with indicated Abs. (D) Stable Ubc9-knockdown Jurkat T cells (shUbc9#2) or the mock control cells (shEGFP) were loaded with indo-1. Intracellular Ca2+ flux after anti-CD3 or ionomycin stimulation was measured by flow cytometry. Arrows indicate the time of addition of stimulus. (E) Stable Ubc9-knockdown Jurkat T cells (shUbc9#2) or the mock control cells (shEGFP) were cultured with anti-CD3 for 24 h, stained with FITC-CD69, and analyzed by FACS. (F) Left panel: Jurkat T cells were cotransfected with SKAP1 and either ADAP WT, ADAP dNLS1, or ADAP dNLS2 constructs, followed by immunoprecipitation with anti-SKAP1 Ab and blotting with anti-ADAP and anti-SKAP1. Right panel: COS-7 cells were cotransfected with SLP-76 and either ADAP WT, ADAP dNLS1, or ADAP dNLS2 constructs, followed by immunoprecipitation with anti–SLP-76 Ab and immunoblotting with anti-ADAP and anti–SLP-76. (G) Jurkat T cells were cotransfected with IL-2 promoter-driven luciferase reporter and either vector, ADAP WT, ADAP dNLS1, or ADAP dNLS2 constructs, stimulated with anti-CD3 for 6 h followed by luciferase activity analysis as described in 2Materials and Methods. Error bars indicate the SE from three individual experiments. (H) HA-tagged ADAP WT and deletion mutants were expressed in Jurkat T cells and stimulated with ionomycin plus PMA for 30 min. Confocal images of the cells showed subcellular localizations of ADAP WT and deletion mutants (green) by immunofluorescence microscopy using anti-HA Ab. Scale bar, 5 μm. ns, not significant.

FIGURE 3.

ADAP-Ubc9 binding fails to affect TCR proximal signaling events and TCR-induced IL-2 transcription. (A) Stable Ubc9-knockdown and control shEGFP Jurkat T cells were either left untreated or stimulated with anti-CD3 (OKT3) for 3 min. The cell lysates were extracted and subjected to Western blotting with Abs against phosphotyrosine, α-tubulin, and Ubc9. (B) Jurkat T cells were cotransfected with IL-2 (upper left panel) or NF-κB (upper right panel) promoter-driven luciferase reporter with either shEGFP or shUbc9#2 constructs and were stimulated with anti-CD3 for 6 h, followed by a measurement of luciferase activity as indicated in 2Materials and Methods. Error bars indicate the SE from three individual experiments. The knockdown of Ubc9 was assessed by anti-Ubc9 Western blotting in the lower panel. (C) Stable Ubc9-knockdown Jurkat T cells (shUbc9#2) or the mock control T cells (shEGFP) were either left unstimulated or were stimulated with anti-CD3 for 5 min. Whole cell lysate was subjected to immunoprecipitation with anti–ZAP-70 (left panel) and anti–PLC-γ1 (right panel), followed by Western blotting with indicated Abs. (D) Stable Ubc9-knockdown Jurkat T cells (shUbc9#2) or the mock control cells (shEGFP) were loaded with indo-1. Intracellular Ca2+ flux after anti-CD3 or ionomycin stimulation was measured by flow cytometry. Arrows indicate the time of addition of stimulus. (E) Stable Ubc9-knockdown Jurkat T cells (shUbc9#2) or the mock control cells (shEGFP) were cultured with anti-CD3 for 24 h, stained with FITC-CD69, and analyzed by FACS. (F) Left panel: Jurkat T cells were cotransfected with SKAP1 and either ADAP WT, ADAP dNLS1, or ADAP dNLS2 constructs, followed by immunoprecipitation with anti-SKAP1 Ab and blotting with anti-ADAP and anti-SKAP1. Right panel: COS-7 cells were cotransfected with SLP-76 and either ADAP WT, ADAP dNLS1, or ADAP dNLS2 constructs, followed by immunoprecipitation with anti–SLP-76 Ab and immunoblotting with anti-ADAP and anti–SLP-76. (G) Jurkat T cells were cotransfected with IL-2 promoter-driven luciferase reporter and either vector, ADAP WT, ADAP dNLS1, or ADAP dNLS2 constructs, stimulated with anti-CD3 for 6 h followed by luciferase activity analysis as described in 2Materials and Methods. Error bars indicate the SE from three individual experiments. (H) HA-tagged ADAP WT and deletion mutants were expressed in Jurkat T cells and stimulated with ionomycin plus PMA for 30 min. Confocal images of the cells showed subcellular localizations of ADAP WT and deletion mutants (green) by immunofluorescence microscopy using anti-HA Ab. Scale bar, 5 μm. ns, not significant.

Close modal

In addition to its central role in TCR-mediated adhesion, ADAP participates in the signaling pathways downstream of the TCR that result in T cell activation (8, 18). Next, it was important to assess whether Ubc9-ADAP binding is required for the proper functions of ADAP in TCR proximal responses, including the formation of signaling complex of ADAP with its binding partners such as SKAP1 and SLP-76, as well as TCR-induced IL-2 transcription in T cells. SKAP1 or SLP-76 were coexpressed with either ADAP, ADAP mutant dNLS1, or the Ubc9-binding–deficient ADAP mutant dNLS2 in Jurkat T cells or COS-7 cells respectively, followed by coimmunoprecipitation with anti-SKAP1 or anti–SLP-76 and blotting with anti-ADAP. ADAP dNLS2 bound similar amounts of SKAP1 (Fig. 3F, left panel, lane 3 versus lanes 1 and 2) and SLP-76 (Fig. 3F, right panel, lane 3 versus lanes 1 and 2, and Supplemental Fig. 2B, lane 3 versus lanes 1 and 2) compared with WT ADAP and the ADAP dNLS1 mutant.

ADAP positively regulates the signaling downstream of the TCR, and its phosphorylation at both the Tyr595 and Tyr651 sites is essential for IL-2 production related to T cell activation (15, 44). To determine whether Ubc9-ADAP binding is required for ADAP-mediated IL-2 production, we next examined if the Ubc9-ADAP interaction is required for TCR-induced IL-2 transcription. Jurkat T cells were cotransfected with an IL-2 promoter-driven luciferase reporter together with either WT ADAP or ADAP mutants, including the Ubc9-binding–deficient ADAP mutant dNLS2. The luciferase assay revealed that in the absence of anti-CD3 stimulation, the luciferase activity was minimal. In the presence of anti-CD3 stimulation, a significant 7-fold increase in luciferase activity was observed (Fig. 3G, vector, anti-CD3 versus unstimulated). Transfection with the ADAP WT expression construct caused a further 2-fold increase in the luciferase activity when anti-CD3 stimulation was present (Fig. 3G, ADAP WT versus vector). Surprisingly, luciferase activity in cells transfected with the Ubc9-binding–deficient mutant dNLS2, or the other mutant dNLS1, showed levels of TCR-mediated IL-2 transcription comparable to those of cells transfected with WT ADAP (Fig. 3G, ADAP dNLS2 and dNLS1 versus ADAP WT). To assess whether deletion of Ubc9-ADAP binding site affects the subcellular distribution of ADAP following T cell activation, we next transfected Jurkat T cells with HA-tagged ADAP WT or truncation mutants, and stimulated the cells with ionomycin plus PMA followed by visualization with immunofluorescence microscopy using anti-HA Abs. The staining pattern revealed a distinct localization of ADAP proteins throughout the cytoplasm, whereas cells expressing ADAP dNLS1 and ADAP dNLS2 displayed no difference in the subcellular distribution (Fig. 3H).

Together, these data suggest that the Ubc9-ADAP interaction is not required for TCR proximal signaling responses as well as for the positive regulatory effects of ADAP on TCR-induced IL-2 transcription.

ADAP plays an important role in the adhesion of integrin to ICAM-1 and fibronectin in T cells (8, 15). Given that TCR stimulation increased the association between ADAP and Ubc9 (Fig. 2A, lane 5 versus lane 4), we next assessed whether the Ubc9-ADAP module is essential for TCR-mediated T cell adhesion by an in vitro ICAM-1–binding assay in Jurkat T cells. Consistent with previous reports, overexpression of ADAP enhanced T cell adhesion to ICAM-1–coated plates, which was further increased ∼5-fold upon TCR activation (Fig. 4A, ADAP versus vector). Interestingly, this profound potentiation of T cell adhesion induced by anti-CD3 stimulation was diminished sharply by 70% in cells expressing the Ubc9-binding–deficient mutant ADAP dNLS2 (Fig. 4A, ADAP dNLS2 versus ADAP). In addition, similarly impaired anti-CD3–induced adhesion to fibronectin, another ligand of β1 integrins, was observed in ADAP dNLS2–transfected Jurkat T cells compared with ADAP WT–transfected Jurkat T cells (Fig. 4B, ADAP dNLS2 versus ADAP).

FIGURE 4.

Ubc9-ADAP binding is required for integrin adhesion and clustering. (A and B) Jurkat T cells transfected with either the empty vector, ADAP WT, ADAP dNLS1, or ADAP dNLS2 constructs were either left unstimulated or stimulated with anti-CD3 for 30 min, followed by an in vitro assay for adhesion to either ICAM-1–coated plates (A) or fibronectin-coated plates (B). Error bars indicate the SE from four individual experiments for the ICAM-1–binding assay, and SE from three individual experiments for fibronectin-binding assay. (C) Upper panel: Representative microscopy images of Jurkat T cells transfected with either empty vector, ADAP WT, ADAP dNLS1, or ADAP dNLS2 constructs. Cells were stimulated with anti-CD3 for 30 min, then fixed and stained with anti-CD11a and Alexa Fluor 488-conjugated anti-mouse Abs. Arrowhead designates LFA-1 clustering. Scale bar, 10 μm. Lower left panel: Histogram showing the percentage of T cells with LFA-1 clustering. Error bars indicate the SE from three individual experiments. Lower right panel: Anti-HA Western blot analysis of cells transfected with either ADAP WT, ADAP dNLS1, or ADAP dNLS2 construct. *p < 0.05, ***p < 0.001, ****p < 0.0001. ns, not significant.

FIGURE 4.

Ubc9-ADAP binding is required for integrin adhesion and clustering. (A and B) Jurkat T cells transfected with either the empty vector, ADAP WT, ADAP dNLS1, or ADAP dNLS2 constructs were either left unstimulated or stimulated with anti-CD3 for 30 min, followed by an in vitro assay for adhesion to either ICAM-1–coated plates (A) or fibronectin-coated plates (B). Error bars indicate the SE from four individual experiments for the ICAM-1–binding assay, and SE from three individual experiments for fibronectin-binding assay. (C) Upper panel: Representative microscopy images of Jurkat T cells transfected with either empty vector, ADAP WT, ADAP dNLS1, or ADAP dNLS2 constructs. Cells were stimulated with anti-CD3 for 30 min, then fixed and stained with anti-CD11a and Alexa Fluor 488-conjugated anti-mouse Abs. Arrowhead designates LFA-1 clustering. Scale bar, 10 μm. Lower left panel: Histogram showing the percentage of T cells with LFA-1 clustering. Error bars indicate the SE from three individual experiments. Lower right panel: Anti-HA Western blot analysis of cells transfected with either ADAP WT, ADAP dNLS1, or ADAP dNLS2 construct. *p < 0.05, ***p < 0.001, ****p < 0.0001. ns, not significant.

Close modal

Because knockdown of Ubc9 caused impaired LFA-1 clustering in response to anti-CD3 stimulation, we next assessed if expression of ADAP dNLS2 has a similar effect on this event. Cells were considered to have clustered LFA-1 if the staining pattern showed LFA-1 polarized to one side of the cell. In vector-transfected Jurkat T cells, clustered LFA-1 (indicated by the white arrows) was observed and anti-CD3 stimulation increased the percentage of cells with clustering from a basal 1–4% (Fig. 4C). Transfection of ADAP or ADAP dNLS1 further increased the percentages of cells with clustering by 2-fold to ∼9%. In contrast, expression of the Ubc9-binding–deficient mutant ADAP dNLS2 failed to further augment LFA-1 clustering, with percentages comparable to that in the vector control cells (Fig. 4C). Expressions of the transfected ADAP WT and mutant proteins were confirmed by immunoblotting with anti-ADAP (Fig. 4C, lower right panel). These data indicated that dissociation of Ubc9 from ADAP impairs TCR-induced integrin adhesion and clustering of T cells.

Thus, either knockdown of Ubc9 or expression of the Ubc9-binding–deficient ADAP mutant led to impaired integrin-mediated T cell adhesion and LFA-1 clustering, indicating that ADAP-Ubc9 interaction is required for T cell adhesion.

Rap1 and its binding proteins RIAM and RapL are critical components of β2 integrin activation downstream of the TCR-ADAP-SKAP1 axis, and ADAP-mediated plasma membrane recruitment of Rap1 and RIAM following TCR activation is required for LFA-1 activation (5, 16). Thus, we investigated whether Ubc9 is required for the TCR-induced plasma membrane translocation or recruitment of Rap1, RapL, and RIAM. The membrane fractions extracted from murine primary T cells infected with lentiviruses expressing shRNAs against Ubc9 or EGFP were subjected to Western blot analysis for the presence of Rap1, RapL, and RIAM. As shown in Fig. 5A, TCR stimulation substantially increased the plasma membrane recruitments of Rap1 and RapL in mock control shEGFP-infected cells (Fig. 5A, shEGFP, lane 6 versus lane 5). In contrast, the anti-CD3–induced membrane targeting of Rap1 and RapL was significantly reduced in the shUbc9#2-infected cells. However, the membrane recruitment of RIAM is unaffected (Fig. 5A, lane 8 versus lane 6).

FIGURE 5.

Effects of Ubc9 knockdown on LFA-1–mediated membrane localization of Rap1-RIAM-RapL and fibronectin-mediated Rac1 activation in response to anti-CD3 stimulation. (A) Purified mouse primary CD4+ cells were infected with lentiviruses expressing shRNAs against Ubc9 or EGFP, and cultured ex vivo for 2 d. Cells were either left unstimulated or stimulated with 4 μg/ml anti-CD3 (2C11) for 10 min, and subjected to cytosolic and plasma membrane fractionation, followed by Western blotting with Abs against Rap1, RapL, RIAM, tubulin, and Ubc9 as indicated. The anti-tubulin blotting was used as an internal control. (B) Double immunofluorescence staining of Rap1 (green) and LFA-1 (red) in stable Ubc9-knockdown or control shEGFP Jurkat T cells with or without anti-CD3 stimulation. Cells were fixed, permeabilized, stained, and visualized for LFA-1 (red) and Rap1 (green) as described in the 2Materials and Methods. Images (Ba–Bd) are representative of three independent experiments. Scale bar, 5 μm. The graph (Be) indicates the percentage of Rap1 localized at the plasma membrane from a total of 30–50 cells in each condition. (C) Stable Ubc9-knockdown Jurkat T cells (shUbc9#2) or the mock control T cells (shEGFP) were either left unstimulated or stimulated with 5 μg/ml anti-CD3 (OKT3) for 60 min. Rac1 activity was assessed by pull-down assays using GST-PBD-PAK. GTP-bound Rac1 and total Rac1 were determined by immunoblot with anti-Rac1 Ab. Right panel: the bar graph represented the relative intensity of the bands (Rac-GTP/Total Rac1) measured by ImageJ software. The fold induction of each sample was normalized to that of unstimulated shEGFP cells. Error bars indicate the SE from three individual experiments. ****p < 0.0001.

FIGURE 5.

Effects of Ubc9 knockdown on LFA-1–mediated membrane localization of Rap1-RIAM-RapL and fibronectin-mediated Rac1 activation in response to anti-CD3 stimulation. (A) Purified mouse primary CD4+ cells were infected with lentiviruses expressing shRNAs against Ubc9 or EGFP, and cultured ex vivo for 2 d. Cells were either left unstimulated or stimulated with 4 μg/ml anti-CD3 (2C11) for 10 min, and subjected to cytosolic and plasma membrane fractionation, followed by Western blotting with Abs against Rap1, RapL, RIAM, tubulin, and Ubc9 as indicated. The anti-tubulin blotting was used as an internal control. (B) Double immunofluorescence staining of Rap1 (green) and LFA-1 (red) in stable Ubc9-knockdown or control shEGFP Jurkat T cells with or without anti-CD3 stimulation. Cells were fixed, permeabilized, stained, and visualized for LFA-1 (red) and Rap1 (green) as described in the 2Materials and Methods. Images (Ba–Bd) are representative of three independent experiments. Scale bar, 5 μm. The graph (Be) indicates the percentage of Rap1 localized at the plasma membrane from a total of 30–50 cells in each condition. (C) Stable Ubc9-knockdown Jurkat T cells (shUbc9#2) or the mock control T cells (shEGFP) were either left unstimulated or stimulated with 5 μg/ml anti-CD3 (OKT3) for 60 min. Rac1 activity was assessed by pull-down assays using GST-PBD-PAK. GTP-bound Rac1 and total Rac1 were determined by immunoblot with anti-Rac1 Ab. Right panel: the bar graph represented the relative intensity of the bands (Rac-GTP/Total Rac1) measured by ImageJ software. The fold induction of each sample was normalized to that of unstimulated shEGFP cells. Error bars indicate the SE from three individual experiments. ****p < 0.0001.

Close modal

Additionally, double immunofluorescence with Rap1 and LFA-1 was also performed to assess the effect of downregulation of Ubc9 on the plasma membrane translocation of Rap1 and LFA-1 clustering in response to anti-CD3 stimulation. The level of Rap1 localized on the plasma membrane was analyzed by quantifying the Rap1 signal intensity of the plasma membrane area compared with the whole cell. As shown in Fig. 5B, in the absence of anti-CD3 stimulation, Rap1 distributed diffusely over the cells in both control shEGFP (Fig. 5Ba) and Ubc9-knockdown T cells (Fig. 5Bc). In the control shEGFP cells, anti-CD3 stimulation led to an increase in the level of Rap1 at the plasma membrane, and the Rap1 at the plasma membrane was colocalized with clustered LFA-1 (indicated by the white arrows in Fig. 5Bb). In contrast, the level of anti-CD3–induced increases in the translocation of Rap1 on plasma membrane was significantly decreased in the Ubc9-knockdown cells, accompanied by a reduction in LFA-1 cluster formation (Fig. 5Bd; shUbc9#2, anti-CD3 versus unstimulated in Fig. 5Be).

Rac1 belongs to the Rho subgroup of a family of small GTPases, and the active GTP-bound Rac1 is critical for α4β1 (VLA-4) -mediated T cell adhesion to fibronectin (4548). ADAP facilitates Rac1 activation and α4β1-mediated adhesion (45). To address whether the contribution of Ubc9 to T cell adhesion is involved in Rac1 activation signaling, we performed a Rac1 activity assay by detecting the level of active Rac1 (GTP-bound Rac1) in the lysate of control shEGFP Jurkat T cells or Ubc9-knockdown T cells with or without anti-CD3 stimulation. Consistent with a previous observation (49), anti-CD3 treatment led to an increase in the amount of GTP-Rac1 detected in control shEGFP Jurkat T cells (Fig. 5C, shEGFP, anti-CD3 versus unstimulated), indicating that Rac1 was activated upon TCR activation. In contrast, the anti-CD3 stimulation-induced Rac1 activation was significantly decreased in Ubc9-knockdown T cells in both unstimulated and anti-CD3–stimulated cells (Fig. 5C, anti-CD3, shUbc9#2 versus shEGFP). These results suggested that the effect of Ubc9 on T cell adhesion relies on its modulation of TCR-mediated Rac1 activation and plasma membrane recruitment of Rap1 and its effector protein RapL, but not RIAM (Fig. 6).

FIGURE 6.

Proposed model for the selective role of Ubc9-ADAP in the regulation of TCR-mediated T cell adhesion and signaling. ADAP possesses regulatory roles in the integrin inside-out signaling for T cell adhesion and in TCR signaling events. The ADAP-SKAP1 module regulates the activation of integrin LFA-1 via its downstream two parallel signaling axis: TCR-Rap1-RapL-LFA-1 αL and Rap1-RIAM-Talin-LFA-1 β2 (10, 43, 60). The distinct pool of ADAP-SKAP independently associates with RapL and RIAM, but collectively determines the integrin LFA-1 activation (22). In addition, the ADAP-SKAP1 module facilitates VLA-4 activation via signaling axis TCR-Rac1-VLA-4/fibronectin (4547). Our data suggest a model for the selective role of Ubc9-ADAP in the TCR-induced integrin activation for T cell adhesion. In this model, TCR stimulation induces Ubc9 association with ADAP. Although it has no effect on the TCR proximal signaling responses and TCR-induced IL-2 transcription, the interaction of Ubc9 with ADAP is selectively required for TCR-induced Rac1 activation that leads to the enhancements of VLA-4–mediated T cell adhesion to fibronectin (signaling axis 2 in circle), and the membrane targeting of Rap1 and RapL, but not RIAM for the activation of LFA-1 integrin (signaling axis 1 in circle). Thus, our findings reconfigure the existing model of inside-out signaling and introduce the Ubc9-ADAP module as a new regulatory layer on the controls of integrin-mediated T cell adhesion.

FIGURE 6.

Proposed model for the selective role of Ubc9-ADAP in the regulation of TCR-mediated T cell adhesion and signaling. ADAP possesses regulatory roles in the integrin inside-out signaling for T cell adhesion and in TCR signaling events. The ADAP-SKAP1 module regulates the activation of integrin LFA-1 via its downstream two parallel signaling axis: TCR-Rap1-RapL-LFA-1 αL and Rap1-RIAM-Talin-LFA-1 β2 (10, 43, 60). The distinct pool of ADAP-SKAP independently associates with RapL and RIAM, but collectively determines the integrin LFA-1 activation (22). In addition, the ADAP-SKAP1 module facilitates VLA-4 activation via signaling axis TCR-Rac1-VLA-4/fibronectin (4547). Our data suggest a model for the selective role of Ubc9-ADAP in the TCR-induced integrin activation for T cell adhesion. In this model, TCR stimulation induces Ubc9 association with ADAP. Although it has no effect on the TCR proximal signaling responses and TCR-induced IL-2 transcription, the interaction of Ubc9 with ADAP is selectively required for TCR-induced Rac1 activation that leads to the enhancements of VLA-4–mediated T cell adhesion to fibronectin (signaling axis 2 in circle), and the membrane targeting of Rap1 and RapL, but not RIAM for the activation of LFA-1 integrin (signaling axis 1 in circle). Thus, our findings reconfigure the existing model of inside-out signaling and introduce the Ubc9-ADAP module as a new regulatory layer on the controls of integrin-mediated T cell adhesion.

Close modal

Although the SUMOylation pathway has been established to be necessary for various key cellular processes, its role in T cell function remains unknown. We investigated the role of Ubc9, the sole SUMO E2 conjugase of the SUMO pathway, in the regulation of T cell signaling and adhesion. The present study demonstrates that Ubc9 regulates T cell adhesion via a mechanism involving a direct interaction with the immune adaptor ADAP: 1) although proximal T cell signaling responses are not affected, T cell adhesion is impaired in Ubc9-knockdown T cells; 2) Ubc9 directly binds to ADAP within a segment of NLS spanning residues 674–700, and the binding is enhanced by anti-CD3 stimulation; 3) Ubc9-ADAP interaction is required for integrin-mediated T cell adhesion and LFA-1 clustering but not for TCR-mediated IL-2 transcription; 4) the mechanism of Ubc9 regulation of T cell adhesion involves the plasma membrane translocation of Rap1 and RapL, but not RIAM, and Rac1 activation upon TCR stimulation. Our study revealed the importance of Ubc9 in T cell adhesion, and further presented the underlying mechanisms of the regulation via a direct interaction between Ubc9 with immune adaptor ADAP that is required for the TCR-induced plasma membrane translocation of Rap1 and RapL, but not RIAM, as well as Rac1 activation.

There are two NLS sites within ADAP with unknown functions. A comparison of the exact sequence of the NLS1 and NLS2 of ADAP reveals that both NLS motifs bear several positively charged residues as KR/KK that resemble the bipartite nuclear localization motif, but the major difference between them in sequence is that there is a unique 5-aa motif of KKLKK present in NLS2 but not in NLS1. Of note, interestingly, this KKLKK motif is also found in the known Ubc9-interacting NLSs of proteins visual system homeobox 1 (Vsx-1) (26) and androgen receptor (AR) (50) (Supplemental Fig. 3B). The Ubc9 binding site in ADAP was mapped to NLS2, which was in line with previous reports that Ubc9 binds to the NLSs of other known NLS-containing molecules including Vsx-1, AR, and poly(A) polymerase (51). Despite this, deletion of NLS1 also decreased the binding of ADAP to Ubc9 (Fig. 2F), and led to a significant defect in T cell adhesion to ICAM-1 (Fig. 4A). A possible mechanism could be that the NLS1 may indirectly contribute to the ADAP binding affinity and specificity to Ubc9, for example, stabilizing a favorable confirmation for the interaction mediated by the KKLKK motif, which is present in NLS2 and the other known Ubc9-interacting NLSs of Vsx-1 and AR but not in NLS1.

ADAP consists of two SH3 domains, which are each located at the C-terminal side of NLS1 and NLS2, respectively (Supplemental Fig. 3A). A previous report has defined a pitch of 12 aa that forms an amphipathic helix at the N terminus of each SH3 domain (52). Although the SH3 domain mediates the interaction with the phospholipids, the amphipathic helix is not required for the phospholipid binding but it stabilizes the hSH3 domain. Interestingly, deletion of the hSH3 domain led to a modest increase in the TCR-mediated T cell adhesion (53), suggesting a nonoverlapping function between NLS2 and hSH3 in the regulation of T cell adhesion. This excludes the possibility that the deletion of NLS2 could interfere with the adjacent hSH3 domain in its phospholipid binding, which accounts for the defective adhesion.

The in vivo coimmunoprecipitation in T cells showed that Ubc9 was pulled down not only by anti-ADAP but also by anti-SKAP1 (Fig. 2A), suggesting that there is a formation of a signaling complex containing Ubc9, ADAP, and SKAP1 at least at some stage of the event. However, in an overexpression system in HEK 293T cells where both ADAP and SKAP1 are absent, Ubc9 only associated with overexpressed ADAP but not with SKAP1 (Fig. 2C), suggesting that Ubc9 interacts directly with ADAP rather than SKAP1 within the ADAP-SKAP1 module. Given the fact that ADAP constitutively associates with SKAP1 and expression of ADAP is required for stable expression of SKAP1 in T cells (10), the precipitation of Ubc9 by SKAP1 is indirect and bridged via ADAP. Apart from its enzymatic activity as to conjugate SUMO moieties to protein substrates, Ubc9 also functions independently of SUMOylation in regulating cellular physiology (26, 54, 55). Proteins that interact with Ubc9 without being SUMOylated include high mobility group A1, Vsx-1, and chicken OVA upstream promoter transcriptional factors (26, 56, 57). Our coimmunoprecipitation suggests that ADAP was not SUMOylated by Ubc9 with or without anti-CD3 stimulation, indicating that Ubc9 modulation of integrin-mediated T cell adhesion is independent of Ubc9-mediated SUMO modification of ADAP. However, it has not excluded the possibility that the recruited Ubc9 in the complex with ADAP can mediate SUMOylation of a binding partner or a downstream effector molecule of ADAP to exert additional functions rather than T cell adhesion.

Removal of the Ubc9-binding domain NLS2 at the C terminus of ADAP decreased significantly the TCR-induced T cell adhesion to a level comparable to that caused by the deletion of the SKAP1-binding region at the N terminus of ADAP, which is in agreement with the previous reports showing a full function of ADAP in the TCR-mediated integrin activation requires both N terminus and C terminus of ADAP (10, 58). Interestingly, additional deletion of the NLS2 on the base of truncation of the SKAP1-binding region did not result in a further deduction in the TCR-mediated T cell adhesion (Supplemental Fig. 2C). This suggests that the modules of ADAP-Ubc9 and ADAP-SKAP1 do not operate separately, but converge to target the same downstream effector molecules Rap1 and RapL for integrin activation. Previous studies identified two layers of modulation on the inside-out T cell adhesion pathway, the adaptor proteins SLP-76-ADAP-SKAP1 module (2, 43) and the Rap1-RapL-RIAM module (5). Rap1 is an important coordinator between integrins and TCR signals, and T cell activation results in the enrichment of active Rap1 at the plasma membrane destined for LFA-1 activation (59). Active Rap1 binds with RapL in response to TCR ligation, which is required for the spatial distribution of LFA-1 in T cells. Rap1 binding with a second effector RIAM is also necessary for the association of talin with the β-chain of LFA-1 (22, 60). Distinct pools of the ADAP-SKAP1 module in T cells selectively associate with RapL and RIAM at the plasma membrane but collectively facilitate TCR-mediated adhesion (22). Interestingly, in cells where Ubc9 was downregulated, although the TCR-induced RIAM membrane translocation was not affected, the TCR-induced Rap1 and RapL membrane targeting was abolished (Fig. 5A), concomitant with the impaired formation of LFA-1 clusters (Fig. 5B). In addition to LFA-1, the other integrin molecules on the surface of T cells mediating T cell adhesion are VLA-4 (⍺4β1 integrin), whose ligands are fibronectin and VCAM-1 on APCs (4). ADAP facilitates the Rac1 activation and regulates the Rac1-mediated ⍺4β1 integrin activation and T cell adhesion (4548). Our data also showed that Ubc9 downregulation resulted in a significant reduction of T cell adhesion to fibronectin (Fig. 1B, 1D) as well as the abolishment of Rac1 activation (Fig. 5C). Thus, the involvement of Ubc9-ADAP in the regulation of T cell adhesion is a selective and combinative effect via a mechanism of facilitating the activation of Rac1 and the membrane targeting of Rap1-RapL but not RIAM. A possible explanation is that the TCR-induced Ubc9 interaction with ADAP could preferentially direct the recruitment of the ADAP-SKAP1 module to the signaling pool of ADAP-SKAP1-Rap1-RapL and to the pool of ADAP-SKAP1-Rac1-VLA-4 for integrin activation (Figs. 3F, 5A).

In addition to its central role in TCR-mediated adhesion, ADAP participates in the signaling pathways downstream of the TCR by an interaction with SLP-76 at two phosphorylation sites at Tyr595 and Tyr651, thereby causing IL-2 production and T cell activation (8, 18). The Ubc9 binding site deletion mutant retained the ability to form a signaling complex with SKAP1 and SLP-76 (Fig. 3F). Overexpression of the Ubc9-binding–deficient mutant dNLS2 or downregulation of Ubc9 hardly impacts the TCR signaling readouts, including the phosphorylation of the kinases ZAP-70 and PLC-γ1, calcium flux, and transactivation of NF-κB, and subsequently IL-2 transcription and CD69 expression (Fig. 3), suggesting that the Ubc9-ADAP module selectively regulates T cell adhesion but not TCR signaling that leads to T cell activation. One possible mechanism is that the pool of the ADAP-Ubc9 module for adhesion regulation could be segregated from the pool of ADAP-SLP-76 for T cell activation as indicated by IL-2 production. Similar findings were observed in the Grb2-related adaptor downstream of Shc (Gads)-deficient T cells, where Gads is required for TCR-mediated IL-2 release but not TCR-induced adhesion (61). As the region of ADAP involved in Ubc9-dependent regulation of T cell adhesion is distinct from the two major sites involved in the regulation of TCR-mediated IL-2 transcription for its binding partner SLP-76, it is not surprising that the Ubc9-ADAP module regulates TCR-mediated T cell adhesion and IL-2 transcription differently. Thus, these results provide evidence for a novel interaction involving ADAP and Ubc9 that has profound and specific consequences for TCR-dependent adhesion but not for T cell signaling.

The TCR activates integrin-mediated T cell adhesion, which is mediated by either the LFA-1 or VLA-4 binding to the ICAMs or fibronectin, respectively. The ADAP-SKAP1 module regulates the activation of integrin LFA-1 via its downstream two parallel signaling axis: TCR-Rap1-RapL-LFA-1 ⍺L and Rap1-RIAM-Talin-LFA-1 β2 (10, 43, 60). The distinct pool of ADAP-SKAP independently associates with RapL and RIAM, but collectively determines the integrin LFA-1 activation (22). In addition, the ADAP-SKAP1 module facilitates VLA-4 activation via the signaling axis TCR-Rac1-VLA-4/fibronectin (4547). Our data suggest a model for the selective role of Ubc9-ADAP in the TCR-induced integrin activation for T cell adhesion but not in the TCR-induced signaling events. In this model, TCR stimulation induces Ubc9 association with ADAP. Although it has no effect on the TCR proximal signaling responses and TCR-induced IL-2 transcription, the interaction of Ubc9 with ADAP is selectively required for TCR-induced Rac1 activation that leads to the enhancements of VLA-4–mediated T cell adhesion to fibronectin, and the membrane targeting of Rap1 and RapL, but not RIAM for the activation of LFA-1 integrin.

Together, our data identify Ubc9 as a new ADAP interaction partner, and Ubc9-ADAP interaction is selectively required for TCR-mediated integrin activation and T cell adhesion but not for TCR signaling. Our findings reconfigure the existing model of inside-out signaling and introduce the Ubc9-ADAP module as a new regulatory layer on the controls of integrin-mediated T cell adhesion.

We thank Hanshuo Lu and Chang Yang from the Department of Biological Sciences, Xi’an Jiaotong-Liverpool University for technical assistance in biochemistry analysis, and Luna Dong from BD Center of Excellence (Shanghai) for technical help in FACS analysis of calcium flux in T cells.

This work was supported by National Natural Science Foundation of China Grant 31470840 (to H.L.) and Jiangsu Science and Technology Grant BK20131181 (to H.L.).

The online version of this article contains supplemental material.

Abbreviations used in this article:

ADAP

adhesion and degranulation-promoting adaptor protein

AR

androgen receptor

HA

hemagglutinin

NLS

nuclear localization sequence

PAK

p21-activated kinase 1

PBD

p21 binding domain

Rap1

ras-related protein 1

RapL

regulator of adhesion and cell polarization enriched in lymphoid tissues

RIAM

Rap1 interacting adaptor molecule

shRNA

short hairpin RNA

SKAP1

Src kinase–associated phosphoprotein 1

SLP-76

SH2 domain-containing leukocyte protein of 76 kDa

SUMO

small ubiquitin-related modifier

Vsx-1

visual system homeobox 1

WT

wild-type.

1
Billadeau
,
D. D.
,
J. C.
Nolz
,
T. S.
Gomez
.
2007
.
Regulation of T-cell activation by the cytoskeleton.
Nat. Rev. Immunol.
7
:
131
143
.
2
Wang
,
H.
,
C. E.
Rudd
.
2008
.
SKAP-55, SKAP-55-related and ADAP adaptors modulate integrin-mediated immune-cell adhesion.
Trends Cell Biol.
18
:
486
493
.
3
Hogg
,
N.
,
I.
Patzak
,
F.
Willenbrock
.
2011
.
The insider’s guide to leukocyte integrin signalling and function.
Nat. Rev. Immunol.
11
:
416
426
.
4
Chen
,
W.
,
C.
Zhu
.
2013
.
Mechanical regulation of T-cell functions.
Immunol. Rev.
256
:
160
176
.
5
Kinashi
,
T.
2005
.
Intracellular signalling controlling integrin activation in lymphocytes.
Nat. Rev. Immunol.
5
:
546
559
.
6
Hogg
,
N.
,
A.
Smith
,
A.
McDowall
,
K.
Giles
,
P.
Stanley
,
M.
Laschinger
,
R.
Henderson
.
2004
.
How T cells use LFA-1 to attach and migrate.
Immunol. Lett.
92
:
51
54
.
7
Wang
,
H.
,
E. Y.
Moon
,
A.
Azouz
,
X.
Wu
,
A.
Smith
,
H.
Schneider
,
N.
Hogg
,
C. E.
Rudd
.
2003
.
SKAP-55 regulates integrin adhesion and formation of T cell-APC conjugates.
Nat. Immunol.
4
:
366
374
.
8
Griffiths
,
E. K.
,
C.
Krawczyk
,
Y. Y.
Kong
,
M.
Raab
,
S. J.
Hyduk
,
D.
Bouchard
,
V. S.
Chan
,
I.
Kozieradzki
,
A. J.
Oliveira-Dos-Santos
,
A.
Wakeham
, et al
.
2001
.
Positive regulation of T cell activation and integrin adhesion by the adapter Fyb/Slap.
Science
293
:
2260
2263
.
9
Burbach
,
B. J.
,
R. B.
Medeiros
,
K. L.
Mueller
,
Y.
Shimizu
.
2007
.
T-cell receptor signaling to integrins.
Immunol. Rev.
218
:
65
81
.
10
Kliche
,
S.
,
D.
Breitling
,
M.
Togni
,
R.
Pusch
,
K.
Heuer
,
X.
Wang
,
C.
Freund
,
A.
Kasirer-Friede
,
G.
Menasche
,
G. A.
Koretzky
,
B.
Schraven
.
2006
.
The ADAP/SKAP55 signaling module regulates T-cell receptor-mediated integrin activation through plasma membrane targeting of Rap1.
Mol. Cell. Biol.
26
:
7130
7144
.
11
Horn
,
J.
,
X.
Wang
,
P.
Reichardt
,
T. E.
Stradal
,
N.
Warnecke
,
L.
Simeoni
,
M.
Gunzer
,
D.
Yablonski
,
B.
Schraven
,
S.
Kliche
.
2009
.
Src homology 2-domain containing leukocyte-specific phosphoprotein of 76 kDa is mandatory for TCR-mediated inside-out signaling, but dispensable for CXCR4-mediated LFA-1 activation, adhesion, and migration of T cells.
J. Immunol.
183
:
5756
5767
.
12
Medeiros
,
R. B.
,
B. J.
Burbach
,
K. L.
Mueller
,
R.
Srivastava
,
J. J.
Moon
,
S.
Highfill
,
E. J.
Peterson
,
Y.
Shimizu
.
2007
.
Regulation of NF-kappaB activation in T cells via association of the adapter proteins ADAP and CARMA1.
Science
316
:
754
758
.
13
da Silva
,
A. J.
,
Z.
Li
,
C.
de Vera
,
E.
Canto
,
P.
Findell
,
C. E.
Rudd
.
1997
.
Cloning of a novel T-cell protein FYB that binds FYN and SH2-domain-containing leukocyte protein 76 and modulates interleukin 2 production.
Proc. Natl. Acad. Sci. USA
94
:
7493
7498
.
14
Musci
,
M. A.
,
L. R.
Hendricks-Taylor
,
D. G.
Motto
,
M.
Paskind
,
J.
Kamens
,
C. W.
Turck
,
G. A.
Koretzky
.
1997
.
Molecular cloning of SLAP-130, an SLP-76-associated substrate of the T cell antigen receptor-stimulated protein tyrosine kinases.
J. Biol. Chem.
272
:
11674
11677
.
15
Wang
,
H.
,
F. E.
McCann
,
J. D.
Gordan
,
X.
Wu
,
M.
Raab
,
T. H.
Malik
,
D. M.
Davis
,
C. E.
Rudd
.
2004
.
ADAP-SLP-76 binding differentially regulates supramolecular activation cluster (SMAC) formation relative to T cell-APC conjugation. [Published erratum appears in 2005 J. Exp. Med. 201: 659.]
J. Exp. Med.
200
:
1063
1074
.
16
Ménasché
,
G.
,
S.
Kliche
,
E. J.
Chen
,
T. E.
Stradal
,
B.
Schraven
,
G.
Koretzky
.
2007
.
RIAM links the ADAP/SKAP-55 signaling module to Rap1, facilitating T-cell-receptor-mediated integrin activation.
Mol. Cell. Biol.
27
:
4070
4081
.
17
Baker
,
R. G.
,
C. J.
Hsu
,
D.
Lee
,
M. S.
Jordan
,
J. S.
Maltzman
,
D. A.
Hammer
,
T.
Baumgart
,
G. A.
Koretzky
.
2009
.
The adapter protein SLP-76 mediates “outside-in” integrin signaling and function in T cells.
Mol. Cell. Biol.
29
:
5578
5589
.
18
Peterson
,
E. J.
,
M. L.
Woods
,
S. A.
Dmowski
,
G.
Derimanov
,
M. S.
Jordan
,
J. N.
Wu
,
P. S.
Myung
,
Q. H.
Liu
,
J. T.
Pribila
,
B. D.
Freedman
, et al
.
2001
.
Coupling of the TCR to integrin activation by Slap-130/Fyb.
Science
293
:
2263
2265
.
19
Dingwall
,
C.
,
R. A.
Laskey
.
1991
.
Nuclear targeting sequences--a consensus?
Trends Biochem. Sci.
16
:
478
481
.
20
Robbins
,
J.
,
S. M.
Dilworth
,
R. A.
Laskey
,
C.
Dingwall
.
1991
.
Two interdependent basic domains in nucleoplasmin nuclear targeting sequence: identification of a class of bipartite nuclear targeting sequence.
Cell
64
:
615
623
.
21
Huang
,
Y.
,
D. D.
Norton
,
P.
Precht
,
J. L.
Martindale
,
J. K.
Burkhardt
,
R. L.
Wange
.
2005
.
Deficiency of ADAP/Fyb/SLAP-130 destabilizes SKAP55 in Jurkat T cells.
J. Biol. Chem.
280
:
23576
23583
.
22
Kliche
,
S.
,
T.
Worbs
,
X.
Wang
,
J.
Degen
,
I.
Patzak
,
B.
Meineke
,
M.
Togni
,
M.
Moser
,
A.
Reinhold
,
F.
Kiefer
, et al
.
2012
.
CCR7-mediated LFA-1 functions in T cells are regulated by 2 independent ADAP/SKAP55 modules.
Blood
119
:
777
785
.
23
Johnson
,
E. S.
,
G.
Blobel
.
1997
.
Ubc9p is the conjugating enzyme for the ubiquitin-like protein Smt3p.
J. Biol. Chem.
272
:
26799
26802
.
24
Flotho
,
A.
,
F.
Melchior
.
2013
.
Sumoylation: a regulatory protein modification in health and disease.
Annu. Rev. Biochem.
82
:
357
385
.
25
Gareau
,
J. R.
,
C. D.
Lima
.
2010
.
The SUMO pathway: emerging mechanisms that shape specificity, conjugation and recognition.
Nat. Rev. Mol. Cell Biol.
11
:
861
871
.
26
Kurtzman
,
A. L.
,
N.
Schechter
.
2001
.
Ubc9 interacts with a nuclear localization signal and mediates nuclear localization of the paired-like homeobox protein Vsx-1 independent of SUMO-1 modification.
Proc. Natl. Acad. Sci. USA
98
:
5602
5607
.
27
Liu
,
L. B.
,
W.
Omata
,
I.
Kojima
,
H.
Shibata
.
2007
.
The SUMO conjugating enzyme Ubc9 is a regulator of GLUT4 turnover and targeting to the insulin-responsive storage compartment in 3T3-L1 adipocytes.
Diabetes
56
:
1977
1985
.
28
Kurihara
,
I.
,
H.
Shibata
,
S.
Kobayashi
,
N.
Suda
,
Y.
Ikeda
,
K.
Yokota
,
A.
Murai
,
I.
Saito
,
W. E.
Rainey
,
T.
Saruta
.
2005
.
Ubc9 and protein inhibitor of activated STAT 1 activate chicken ovalbumin upstream promoter-transcription factor I-mediated human CYP11B2 gene transcription.
J. Biol. Chem.
280
:
6721
6730
.
29
Liu
,
H.
,
H.
Schneider
,
A.
Recino
,
C.
Richardson
,
M. W.
Goldberg
,
C. E.
Rudd
.
2015
.
The immune adaptor SLP-76 binds to SUMO-RANGAP1 at nuclear pore complex filaments to regulate nuclear import of transcription factors in T cells.
Mol. Cell
59
:
840
849
.
30
Kim
,
J. H.
,
H. J.
Choi
,
B.
Kim
,
M. H.
Kim
,
J. M.
Lee
,
I. S.
Kim
,
M. H.
Lee
,
S. J.
Choi
,
K. I.
Kim
,
S. I.
Kim
, et al
.
2006
.
Roles of sumoylation of a reptin chromatin-remodelling complex in cancer metastasis.
Nat. Cell Biol.
8
:
631
639
.
31
Cefai
,
D.
,
M.
Ferrer
,
N.
Serpente
,
T.
Idziorek
,
A.
Dautry-Varsat
,
P.
Debre
,
G.
Bismuth
.
1992
.
Internalization of HIV glycoprotein gp120 is associated with down-modulation of membrane CD4 and p56lck together with impairment of T cell activation.
J. Immunol.
149
:
285
294
.
32
Céfaï
,
D.
,
H.
Schneider
,
O.
Matangkasombut
,
H.
Kang
,
J.
Brody
,
C. E.
Rudd
.
1998
.
CD28 receptor endocytosis is targeted by mutations that disrupt phosphatidylinositol 3-kinase binding and costimulation.
J. Immunol.
160
:
2223
2230
.
33
Wang
,
H.
,
H.
Liu
,
Y.
Lu
,
M.
Lovatt
,
B.
Wei
,
C. E.
Rudd
.
2007
.
Functional defects of SKAP-55-deficient T cells identify a regulatory role for the adaptor in LFA-1 adhesion.
Mol. Cell. Biol.
27
:
6863
6875
.
34
Meller
,
N.
,
Y. C.
Liu
,
T. L.
Collins
,
N.
Bonnefoy-Bérard
,
G.
Baier
,
N.
Isakov
,
A.
Altman
.
1996
.
Direct interaction between protein kinase C theta (PKC theta) and 14-3-3 tau in T cells: 14-3-3 overexpression results in inhibition of PKC theta translocation and function.
Mol. Cell. Biol.
16
:
5782
5791
.
35
Shevchenko
,
A.
,
H.
Tomas
,
J.
Havlis
,
J. V.
Olsen
,
M.
Mann
.
2006
.
In-gel digestion for mass spectrometric characterization of proteins and proteomes.
Nat. Protoc.
1
:
2856
2860
.
36
Raab
,
M.
,
H.
Kang
,
A.
da Silva
,
X.
Zhu
,
C. E.
Rudd
.
1999
.
FYN-T-FYB-SLP-76 interactions define a T-cell receptor zeta/CD3-mediated tyrosine phosphorylation pathway that up-regulates interleukin 2 transcription in T-cells.
J. Biol. Chem.
274
:
21170
21179
.
37
He
,
X.
,
J.
Riceberg
,
S. M.
Pulukuri
,
S.
Grossman
,
V.
Shinde
,
P.
Shah
,
J. E.
Brownell
,
L.
Dick
,
J.
Newcomb
,
N.
Bence
.
2015
.
Characterization of the loss of SUMO pathway function on cancer cells and tumor proliferation.
PLoS One
10
:
e0123882
.
38
Lin
,
X.
,
M.
Liang
,
Y. Y.
Liang
,
F. C.
Brunicardi
,
X. H.
Feng
.
2003
.
SUMO-1/Ubc9 promotes nuclear accumulation and metabolic stability of tumor suppressor Smad4.
J. Biol. Chem.
278
:
31043
31048
.
39
Nacerddine
,
K.
,
F.
Lehembre
,
M.
Bhaumik
,
J.
Artus
,
M.
Cohen-Tannoudji
,
C.
Babinet
,
P. P.
Pandolfi
,
A.
Dejean
.
2005
.
The SUMO pathway is essential for nuclear integrity and chromosome segregation in mice.
Dev. Cell
9
:
769
779
.
40
van Kooyk
,
Y.
,
C. G.
Figdor
.
2000
.
Avidity regulation of integrins: the driving force in leukocyte adhesion.
Curr. Opin. Cell Biol.
12
:
542
547
.
41
Swaminathan
,
S.
,
F.
Kiendl
,
R.
Körner
,
R.
Lupetti
,
L.
Hengst
,
F.
Melchior
.
2004
.
RanGAP1*SUMO1 is phosphorylated at the onset of mitosis and remains associated with RanBP2 upon NPC disassembly.
J. Cell Biol.
164
:
965
971
.
42
Balagopalan
,
L.
,
V. A.
Barr
,
L. E.
Samelson
.
2009
.
Endocytic events in TCR signaling: focus on adapters in microclusters.
Immunol. Rev.
232
:
84
98
.
43
Raab
,
M.
,
H.
Wang
,
Y.
Lu
,
X.
Smith
,
Z.
Wu
,
K.
Strebhardt
,
J. E.
Ladbury
,
C. E.
Rudd
.
2010
.
T cell receptor “inside-out” pathway via signaling module SKAP1-RapL regulates T cell motility and interactions in lymph nodes.
Immunity
32
:
541
556
.
44
Geng
,
L.
,
M.
Raab
,
C. E.
Rudd
.
1999
.
Cutting edge: SLP-76 cooperativity with FYB/FYN-T in the Up-regulation of TCR-driven IL-2 transcription requires SLP-76 binding to FYB at Tyr595 and Tyr651.
J. Immunol.
163
:
5753
5757
.
45
Dios-Esponera
,
A.
,
S.
Isern de Val
,
S.
Sevilla-Movilla
,
R.
García-Verdugo
,
D.
García-Bernal
,
N.
Arellano-Sánchez
,
C.
Cabañas
,
J.
Teixidó
.
2015
.
Positive and negative regulation by SLP-76/ADAP and Pyk2 of chemokine-stimulated T-lymphocyte adhesion mediated by integrin α4β1.
Mol. Biol. Cell
26
:
3215
3228
.
46
Faroudi
,
M.
,
M.
Hons
,
A.
Zachacz
,
C.
Dumont
,
R.
Lyck
,
J. V.
Stein
,
V. L.
Tybulewicz
.
2010
.
Critical roles for Rac GTPases in T-cell migration to and within lymph nodes.
Blood
116
:
5536
5547
.
47
Price
,
L. S.
,
J.
Leng
,
M. A.
Schwartz
,
G. M.
Bokoch
.
1998
.
Activation of Rac and Cdc42 by integrins mediates cell spreading.
Mol. Biol. Cell
9
:
1863
1871
.
48
D’Souza-Schorey
,
C.
,
B.
Boettner
,
L.
Van Aelst
.
1998
.
Rac regulates integrin-mediated spreading and increased adhesion of T lymphocytes.
Mol. Cell. Biol.
18
:
3936
3946
.
49
Kaminuma
,
O.
,
M.
Deckert
,
C.
Elly
,
Y. C.
Liu
,
A.
Altman
.
2001
.
Vav-Rac1-mediated activation of the c-Jun N-terminal kinase/c-Jun/AP-1 pathway plays a major role in stimulation of the distal NFAT site in the interleukin-2 gene promoter.
Mol. Cell. Biol.
21
:
3126
3136
.
50
Poukka
,
H.
,
P.
Aarnisalo
,
U.
Karvonen
,
J. J.
Palvimo
,
O. A.
Jänne
.
1999
.
Ubc9 interacts with the androgen receptor and activates receptor-dependent transcription.
J. Biol. Chem.
274
:
19441
19446
.
51
Vethantham
,
V.
,
N.
Rao
,
J. L.
Manley
.
2008
.
Sumoylation regulates multiple aspects of mammalian poly(A) polymerase function.
Genes Dev.
22
:
499
511
.
52
Heuer
,
K.
,
M.
Kofler
,
G.
Langdon
,
K.
Thiemke
,
C.
Freund
.
2004
.
Structure of a helically extended SH3 domain of the T cell adapter protein ADAP.
Structure
12
:
603
610
.
53
Heuer
,
K.
,
M.
Sylvester
,
S.
Kliche
,
R.
Pusch
,
K.
Thiemke
,
B.
Schraven
,
C.
Freund
.
2006
.
Lipid-binding hSH3 domains in immune cell adapter proteins.
J. Mol. Biol.
361
:
94
104
.
54
Kaul
,
S.
,
J. A.
Blackford
Jr.
,
S.
Cho
,
S. S.
Simons
Jr.
.
2002
.
Ubc9 is a novel modulator of the induction properties of glucocorticoid receptors.
J. Biol. Chem.
277
:
12541
12549
.
55
Chakrabarti
,
S. R.
,
R.
Sood
,
S.
Ganguly
,
S.
Bohlander
,
Z.
Shen
,
G.
Nucifora
.
1999
.
Modulation of TEL transcription activity by interaction with the ubiquitin-conjugating enzyme UBC9.
Proc. Natl. Acad. Sci. USA
96
:
7467
7472
.
56
Li
,
Y.
,
J.
Lu
,
E. V.
Prochownik
.
2007
.
Dual role for SUMO E2 conjugase Ubc9 in modulating the transforming and growth-promoting properties of the HMGA1b architectural transcription factor.
J. Biol. Chem.
282
:
13363
13371
.
57
Kobayashi
,
S.
,
H.
Shibata
,
I.
Kurihara
,
K.
Yokota
,
N.
Suda
,
I.
Saito
,
T.
Saruta
.
2004
.
Ubc9 interacts with chicken ovalbumin upstream promoter-transcription factor I and represses receptor-dependent transcription.
J. Mol. Endocrinol.
32
:
69
86
.
58
Burbach
,
B. J.
,
R.
Srivastava
,
R. B.
Medeiros
,
W. E.
O’Gorman
,
E. J.
Peterson
,
Y.
Shimizu
.
2008
.
Distinct regulation of integrin-dependent T cell conjugate formation and NF-kappa B activation by the adapter protein ADAP.
J. Immunol.
181
:
4840
4851
.
59
Katagiri
,
K.
,
M.
Imamura
,
T.
Kinashi
.
2006
.
Spatiotemporal regulation of the kinase Mst1 by binding protein RAPL is critical for lymphocyte polarity and adhesion.
Nat. Immunol.
7
:
919
928
.
60
Katagiri
,
K.
,
A.
Maeda
,
M.
Shimonaka
,
T.
Kinashi
.
2003
.
RAPL, a Rap1-binding molecule that mediates Rap1-induced adhesion through spatial regulation of LFA-1.
Nat. Immunol.
4
:
741
748
.
61
Bilal
,
M. Y.
,
E. Y.
Zhang
,
B.
Dinkel
,
D.
Hardy
,
T. M.
Yankee
,
J. C.
Houtman
.
2015
.
GADS is required for TCR-mediated calcium influx and cytokine release, but not cellular adhesion, in human T cells.
Cell. Signal.
27
:
841
850
.

The authors have no financial conflicts of interest.

Supplementary data