Non-cleavable talin rescues defect in the T-cell conjugation of T-cells deficient in the immune adaptor SKAP1

Highlights • Skap1−/− T-cells show impaired talin and RIAM localization at the anti-CD3 beads.• Talin cleavage is altered in Skap1−/− T-cells.• Cleavage resistant talin (L432G) restored normal conjugation of Skap1−/− T-cells.• Immune cell adaptor SKAP1 interfaces with regulation of talin and RIAM in T-cells.

Talin is a ubiquitous high-molecular-weight protein of the cytoskeleton that is essential for the activation of integrins [32,33]. It consists of a C-terminal rod domain, a N-terminal FERM domain and multiple alpha helices [34,35]. The F3 subdomain of the FERM domain binds to the integrin ␤ integrin and is needed to activate integrins [36]. Talin also links the cytoplasmic domain of integrin ␤-chains to actin filaments and is localised in contact regions between T-cell and antigen-presenting cells such as cytolytic Tcells and their targets [37][38][39][40]. Talin also couples integrins to the actin cytoskeleton by interacting with vinculin and alpha-actinin [41,42]. Further, it is a substrate for the calcium activated protease, calpain II [43]. Quantification of adhesion assembly and disassembly rates has demonstrated that this proteolysis is a rate-limiting step for adhesion turnover [44]. In this context, disassembly of adhesion components such as paxillin, vinculin and zyxin, is dependent on this cleavage event [44]. Calpain cleavage also promotes talin binding to the ␤3 integrin cytoplasmic domain and clustering [45]. Talin is targeted to the plasma membrane via the shuttle pro- tein Rap1-GTP-interacting adaptor molecule (RIAM), a member of the MRL (Mig-10/RIAM/Lamellipodin) protein family [46].
Given the binding of talin to the ␤-chain and the SKAP1-RapL-Rap1 complex to the ␣-chain in LFA-1, it has been unclear whether these mediators can affect each other. In this study, we report that T-cells from Skap1−/− mice show altered processing and localization of talin in T-cells, concurrent with reduced dwell times with DCs, and further that a cleavage resistant L432G talin rescued impaired Skap1−/− T-cell conjugation. This observation finding demonstrates cross-regulation between SKAP1 and talin in T-cells despite binding to distinct chains of LFA-1.

T-cell isolation
Spleens isolated from C57Bl6 or SKAP1-deficient mice were meshed through cell strainers, followed by removal of red blood cells (RBC) with hypotonic buffer (0.15 M NH 4 Cl, 1 mM NaHCO 3 , 0.1 mM EDTA, pH 7.25). CD3 + T-cells were purified from the splenocytes using a Mouse T cell Enrichment column (R&D Systems). Cells were then used immediately for experiments. Primary naïve mouse cells were transfected with various vectors using the Amaxa Nucleofector Kit (Lonza, Germany). Jurkat T-cells were transfected by microporation (Digital Bio Technology) using a single pulse of 30 ms at 1410 V. In certain experiments, mouse and Jurkat T-cells were stimulated with 2-5 g/ml of 145-2C11 or OKT3, respectively [47].

Immunofluorescence staining
Immunofluorescence staining was conducted as described. Anti-CD3 coated beads were prepared by incubating 4 g of anti-CD3 (2C11) with 10 6 Dynabeads M-450 Epoxy beads in phosphate buffer for 30 min at 4 • C prior to supplementing with FBS to a final concentration and a further incubation of 0.3% overnight. Alternately, T-cells were plated on polylysine-coated coverslips incubated with anti-CD3 (2 g/ml) for the stipulated time points. The cells were then washed with PBS to remove any non-adherent cells before fixing in Cytofix (BD Biosciences, Oxford, UK). Cells were then permeabilised using 0.5% Saponin before staining with the relevant antibodies. Anti-mouse Alexa568, anti-rabbit Alexa488, anti-rabbit Alexa647 and anti-mouse Alexa568 were used as appropriate secondary antibodies.

Immunoprecipitation and western blotting
Membranes of cells were isolated from detergent solubilisation for immunoprecipitation. Cells were centrifuged at 1850 rpm for 5 min and wash with PBS before resuspending in cold hypotonic buffer (10 mM HEPES, 1.5 mM MgCl 2 , 10 mM KCl, 0.5 mM PMSF, 5 mM DTT, 0.1 mM NaV) supplemented with protease inhibitors (Roche) for 10 min at 4 • C. Cells were then homogenised before centrifugation at 3300 rpm for 15 min at 4 • C. The pellet is discarded and supernatant is centrifuged at 15000 rpm for 1 h to separate cytosolic fraction from membranes. The cytosolic fraction is collected from the supernatant and the membrane fraction is solubilised with RIPA buffer (50 mM Tris-HCl pH 8.0, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate and 0.1% SDS). Immunoprecipitation and Western blotting was conducted as described [23,50].

Statistical analysis
Results are given as the mean ± standard deviation (SD). Statistical significance was tested using unpaired student's T-test using GraphPad Prism version 3.02 (GraphPad Software, San Diego, California, U.S.A.), with p < 0.05 was considered as significant.

SKAP1 is needed for optimal contact times with DCs
We previously reported that SKAP1 was needed for LFA-1 adhesion and T-cell conjugation in response to super-antigen [18,20,23]. To assess the role of SKAP1 in an antigen-specific model, Skap1−/− OT-1 (SKOT1) mice were used. Skap1+/+ OT-1 (OT1) vs. Skap1−/− OT-1 (SKOT1) T-cells were activated for 3 days with 10 g/ml OVA peptide, washed and rested for 24 h followed by a measure of dwell times with DCs and motility (Fig. 1A). Mature DCs were prepared as described previously by labeling with SNARF-1 and pre-incubating with OVA 257-264 peptide (DC-OVA) prior to incubation, as described [51]. The presence of OVA peptide increased contact times from a mean of 237-788 s for OT1 T-cells (p < 0.0001). SKOT1 T-cells also showed an increase in contact time from 287 s to 645 s (p < 0.0001) in the presence of OVA peptide. However, the mean contact time between SKOT1 T-cells and DCs was significantly lower than between OT1 T-cells and DCs (p = 0.0125). The decreased contact time between SKOT1 T-cells and DCs was also accompanied by a decrease in displacement in SKOT1 T-cells (Fig. 1B). To assess an effect of SKAP1 on LFA-1 affinity, we transfected Jurkat T-cells with scrambled vs. SKAP1 siRNA, ligated with anti-CD3 for 30 min and stained cells with the MAb KIM-127 (Fig. 1C). KIM-127 recognizes the intermediate affinity form of human LFA-1 (CD11a/CD18) [52]. Activated human Jurkat T-cells expressing SKAP1 siRNA showed a lower staining than cells expressing scrambled siRNA (Fig. 1C). Overall, these data demonstrated that SKAP1 is needed for LFA-1 activation and contact between T-cells and DCs.

SKAP1 is required for the optimal translocation of talin and RIAM and elongation of T-cell in response to TCR activation
While SKAP1 regulates the Rap1-RapL binding to the ␣L chain of the LFA-1 [23], talin binds to the ␤-subunit and is needed for LFA-1 activation [53,54]. We initially asked whether SKAP1 might affect talin and RIAM localization with LFA-1 at an activation interface (Fig. 2). To assess this, beads coated with anti-CD3 were initially used. They were incubated with primary naïve T-cells from WT and Skap1−/− mice followed by staining of talin and its associated protein RIAM by confocal immunofluorescence. Anti-CD3 induced the translocation of SKAP1 to the contact site with beads as seen in wild-type cells ( Fig. 2A, upper panels). Concurrent with this, the clustering of LFA-1 was observed at the contact with beads in Skap1+/+ T-cells and this was reduced in Skap1−/− T-cells ( Fig. 2A, middle and lower panels), as reported [23,24].
We next looked at the translocation of talin and its associated adaptor RIAM to the contact site (Fig. 2B). With Skap1+/+ T-cells, 81% of the cells showed translocation of talin to the beads that was significantly reduced to 42% with Skap1−/− T-cells (Fig. 2Bi upper  panel and ii). A similar difference was noted with the localization of RIAM where 61% of Skap1+/+ T-cells showed RIAM localization compared to 38% for Skap1−/− T-cells (p < 0.005) (Fig. 2Bii). Further, most of the RIAM was also found outer contact region with beads ( Fig. 2Bi upper panel) and was co-localized with talin (Fig. 2Biii). By contrast, no difference in the number of Skap1+/+ vs. Skap1−/− T-cells was found associated with the anti-CD3 beads (Fig. 2Biv). A similar situation was observed between OT-1 T-cells (CTLs) with peptide presenting EL4 cells with localized cap of talin at the contact site of Skap1+/+ T-cells (Fig. 2Bv, left panel, see arrow), with was diffused in Skap1−/− (SKOT1) cells (right panel). These findings showed that anti-CD3 on beads or OVA peptide presentation b induction of the translocation of talin to the contact point requires the expression of SKAP1. This differences of IS localization correlated with the reduced polarisation of skap1−/− T-cells at the interface of anti-CD3 coated beads (Fig. 2Bvi). Polarization of Tcells was defined as 1.5 times the mean diameter (Fig. 2Bvi). 58% of the WT T-cells were polarized relative to 9% of Skap1−/− T-cells. Skap1−/− T-cells also showed less length extension when in contact with an anti-CD3 bead (Fig. 2Bvii).

SKAP1 deficiency alters anti-CD3 induced talin cleavage
Given the effect of SKAP1 deficiency on talin recruitment to the IS, we next assessed whether SKAP1 might affect the activation status of talin. Talin is cleaved by calpain in a process needed for the disassembly of the focal adhesion complex [44]. Anti-talin antibody [8D 4 ] recognises intact talin molecule (225 kDa) as well as the proteolytic calpain-cleaved fragment at 190 kDa [55]. Anti-CD3 ligation resulted in the degradation of talin between 10 and 30 min in membranes from Skap1+/+ T-cells (Fig. 3, lanes 2,3) [55]. None of the lower Mr proteolytic fragment was found associated with membranes. By contrast, talin and the calpain fragment was observed in membranes of resting skap1−/− T-cells (lane 4). Further, full-length talin was more quickly cleaved from 0 to 10 min with the appearance of the cleaved fragment (lane 4-6) (lower histogram). This observation showed that talin was more readily cleaved in the absence of SKAP1 with the accompanying presence of the cleaved fragment of talin with the membranes of resting and anti-CD3 activated cells. Naïve T-cells were incubated with anti-CD3 coated beads for 1 h were fixed and stained with anti-RIAM plus Alexa-Fluor 568 followed by anti-talin plus Alexa Fluor 468, and were imaged using confocal microscopy. (ii) Histogram shows the percentage of naïve T-cells with RIAM or talin localised at the bead contact site (n = 3, 120-160/experiment). The means ± SD are displayed. (iii) Histogram shows the percentage of naïve T-cells with RIAM and talin co-localised at the bead contact site (n = 3, 120-160/experiment). (iv) Histogram shows the percentage of naïve T-cells in conjugation with an anti-CD3 bead with RIAM and talin co-localised at the bead contact site (n = 3, 120-160/experiment). (v) CTLs from SKOT1 or OT1 show similar distribution of talin and RIAM when in contact with an APC previously incubated with OVA peptide. CTLs from SKOT1 or OT1 were generated from splenocytes with OVA peptide for 3 days before rested for 24 h before incubating with DC-OVA for 10 min. Cells were fixed and stained with anti-talin plus Alexa Fluor 468 (green) and were imaged using confocal microscopy. (vi) Histogram shows the percentage of polarised cells. A polarised cell was defined as at least 1.5 times the mean diameter of an unstimulated cell (n = 3, 120-160/experiment). (vii) Histogram shows the length of cells, measured lengthwise from the point of anti-CD3 bead contact site (n = 3, 120-160/experiment). The means ± SD are displayed. Differences between means are compared using two-tailed unpaired Student's T-test (*, p < 0.05).

Non-cleavable talin L432G rescued the conjugation defect of SKOT1CTLs
Given this result, we next postulated that a calpain resistant form of talin (L432G-EGFP) might rescue the defect in APC conjugation (Fig. 4). For this, Skap1−/− OT 1 T-cells were transfected with either empty EGFP (EGFP), talin-EGFP or talin-L432G-EGFP followed by an assessment of conjugation. Green cells expressing EGFP and their behaviour were visualized and scored under the confocal microscope. Skap1+/+ OT-1 (OT1) vs. Skap1−/− OT-1 (SKOT1) were activated for 3 days with 10 g/ml OVA peptide, washed and rested for 24 h followed by a measure of dwell times with DCs and motility (Fig. 4). No difference in the expression of LFA-1 was observed on the cells. Mature DCs were prepared as described previously by labeling with SNARF-1 [51]. In the presence of OVA peptide, as expected, OT1 cells showed longer contact times with APCs as compared to SKOT1 cells (a mean of 765 s [OT1 EGFP] relative to 610 s [SKOT1 EGFP] (p value = 0.001) (Fig. 4A). The expression of talin-EGFP in SKOT1 T-cells was unable to increase the contact time with DC-OVA. However, the expression of the L432G mutant significantly increased contact times of SKOT1 cells (a mean of 770 s [SKOT1 talin-L432G EGFP] vs. 594 s [SKOT1 talin EGFP]) (Fig. 4A). The mean contact of SKOT1 expressing talin-L432G EGFP was comparable to wild-type OT1 T-cells expressing vector-EGFP. L432G-EGFP also increased displacement of SKOT1 cells (Fig. 4B). These data demonstrated that the reduced conjugation of Skap1−/− T-cells could be rescued by the expression of a form of talin that is resistant to cleavage.
In summary, our findings show that the interface localization and processing of talin is regulated by SKAP1 and that the expression of a cleavage resistant talin (L432G) reversed the reduced conjugation of skap1−/− OT-1 T-cells with DCs. Skap1−/− OT-1 Tcells also showed reduced talin localization to the T-cell interface with antigenic surfaces and DCs and enhanced talin cleavage in resting and anti-CD3 activated T-cells. Talin is necessary for F-actin polarization, the stability of the IS and sustained T cell-APC interactions [56]. Conversely, talin cleavage is needed for the disassembly of the focal adhesion complex [44]. Talin processing may therefore be needed for the disassembly of the contact interface between Tcells and antigen-presenting cells. The enhanced talin cleavage of  Skap1−/− cells might involve a more rapid disassembly of the contact between T-cells and APCs. Expression of non-cleavable talin might therefore stabilize the contact interface and increase dwell times. The SKAP1 pathway also modulates the transport of RapL to Rap1 in binding to the ␣L chain of the LFA-1 receptor [23,24].
T-cell interaction with APCs involves the formation of the central supramolecular activation complex (SMAC [cSMAC]) and peripheral SMAC (pSMAC) in a variety of different cell types [57]. The c-SMAC includes molecules such as CD2, CD28, PKC-, Lck, Fyn, CD4, and CD8, while talin and LFA-1 reside in the p-SMAC [58]. Con-sistent with this, we observed talin to be localized in outer contact regions with anti-CD3 on beads. The loss of SKAP1 in skap1−/− primary T-cells resulted in a significant reduction on the localization of talin and RIAM in this outer region and in the case of interactions between T-cells and antigen-presenting cells showed differences in the degree of polarization and length of cell extension. Whether this altered localization of talin was responsible for the reduce localization and increased polarization and size of contact or vice versa is unclear. This reduction in levels of localized talin at the activation interface could predispose Skap1−/− T-cells to have less stable conjugation due to deficiency of the scaffold for cytoskeletal reorganization [12]. Talin interacts with paxillin, vinculin and zyxin of the cytoskeleton [44]. The importance of full-length talin in conjugation was demonstrated by the ability of protease resistant talin-L432G to restore the dwell times of SKOT1 cells to that of OT1 cells. Consistent with this, a previous report showed that full-length talin rescued adhesion defects of Talin1−/− T-cells [40]. Overall, our findings indicate that SKAP1 regulates both the affinity and avidity of integrins [23,24] leading to altered conjugation with the activating interface of anti-CD3 beads or cells.
The SKAP1-talin connection may interface with other adhesion pathways. Talin has also been reported to act downstream of Rap1A, as constitutively active Rap1A (G12 V) failed to activate ␣IIb␤3integrin in cells expressing low levels of talin [59]. Rap1-interacting adaptor molecule RIAM has also been implicated in talin activation in the absence of Rap1A (G12 V) (22). Calpain-specific inhibitors block T-cell proliferation and cell shape changes upon TCR activation [55]. Calpain I is sensitive to micromolar levels of Ca2 + , in comparison to calpain II which requires millimolar levels of Ca 2+ [60,61]. It is possible that calpain I, the more sensitive enzyme to Ca 2+ present in the cytoplasm after T-cell activation in Skap1−/− T-cells is freely accessible to the talin present in the cytoplasm, resulting in the pronounced cleavage of talin. Overall, our findings suggest that an interplay between two sets of molecules that bind to different chains of LFA-1 in regulating adhesion, and that the altered cleavage and translocation of talin to the IS may be play a role in the reduced ability of Skap1−/− T-cells to form stable conjugates with APCs.