| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Centenary Institute for Cancer Medicine and Cell Biology and
University of Sydney, Sydney, New South Wales, Australia; and
Children’s Hospital of Philadelphia, Philadelphia, PA 19104
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
X-linked lymphoproliferative syndrome (XLP) is an often fatal inherited immunodeficiency characterized by extreme sensitivity to infection with the herpes group virus, EBV (26, 27). The genetic defect in XLP has been identified as SAP (SLAM-associated protein; also known as SH2D1A and DSHP), a small Src homology 2 (SH2) domain-containing protein expressed predominantly by T and NK cells (28, 29, 30, 31). SAP interacts with the CD2 IgSF members SLAM, 2B4, CD84, Ly9, and NTB-A via a unique tyrosine-based motif (TxYxxV/I) present in the cytoplasmic domains of these receptors (3, 8, 21, 30, 32, 33, 34, 35, 36, 37). Thus, signaling through these receptors may be regulated by the recruitment of SAP. Consequently, altered signaling via SAP-associating cell surface molecules may contribute to some of the immunological defects observed in XLP patients. This is consistent with the finding that NK cells from SAP-deficient XLP patients are refractory to activating signals delivered via 2B4 (38, 39, 40).
We have recently characterized the expression and function of CD84 on human B cells and found that it becomes rapidly phosphorylated on tyrosine residues following ligation (35). CD84 is also expressed on human T cells, but not NK cells (19, 35). While a significant amount of information is available regarding the association of SAP with SLAM and, to a lesser extent, 2B4 as well as the contribution that these receptors make to lymphocyte activation, little is known about the function(s) of other CD2 IgSF receptors on human leukocytes or the molecular requirements for the recruitment of SAP or other SH2-domain containing proteins. For these reasons we have now examined the mechanism underlying the interaction between SAP and CD84 and compared the function of CD84 on normal human T cells with that on cells from XLP patients.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
FITC-labeled anti-CD4 and anti-CD8, and PE-labeled anti-CD3 mAb were purchased from BD Biosciences (San Jose, CA). PE-labeled anti-CD84 mAb was obtained from BD PharMingen (San Diego, CA). Purified anti-CD84 mAb (clone 152-1D5) was purchased from NeoMarkers (Fremont, CA). Anti-phosphotyrosine mAb (4G10) was purchased from Upstate Biotechnology (Lake Placid, NY); anti-Myc mAb (9E10) and anti-Lck mAb (clone 3A5) were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). HRP-conjugated, donkey anti-rabbit and donkey anti-mouse IgG antiserum were purchased from Amersham Pharmacia Biotech (Castle Hill, Australia), F(ab')2 of goat anti-mouse IgG (heavy and light chain specific) were purchased from Jackson ImmunoResearch Laboratories (West Grove, PA); anti-
-tubulin mAb was purchased from Sigma-Aldrich (St. Louis, MO). Anti-CD3 mAb (clone SpVT3) (41) and rabbit anti-human SAP polyclonal antiserum have been previously described (21, 39).
Cells and cell lines
Normal human spleens were obtained from organ donors (Australian Red Cross Blood Service). Splenic mononuclear cells (MNC) were prepared as previously described (35). T cells were purified from human spleens using a combination of CD4 and CD8 Dynabeads and CD4/CD8 DetachaBead, according to the manufacturer’s instructions (Dynal Biotech, Oslo, Sweden). The purity of the resulting T cell population was typically >95%, as determined by flow cytometric analysis using anti-CD3 mAb. MNC were isolated from peripheral blood of normal healthy donors and XLP patients by centrifuging diluted whole blood over Ficoll-Paque (Amersham Pharmacia Biotech, Uppsala, Sweden). The three XLP patients used in the study had the following mutations in SAP: patient 1, Tyr54
Cys; patient 2, Phe87Ser; and patient 3, complete deletion of the SAP locus. PHA blasts were generated by stimulating MNC (106/ml) with PHA (5 µg/ml; Sigma-Aldrich) and IL-2 (10 U/ml; Endogen, Woburn, MA). The cell lines used in this study were the human T cell lines F2F7, Jurkat, Lck-deficient Jurkat mutant JCam.1, JCam.1 reconstituted with Lck (42) (provided by Prof. A. Weiss, University of California, San Francisco, CA), and the ZAP-70-deficient Jurkat mutant P116 (43) (provided by Prof. Robert Abraham, Duke University, Raleigh, NC). All cell lines were cultured at 37°C in 5% CO2 in RPMI 1640 tissue culture medium (JRH Biosciences, Lenexa, KS) supplemented with 10% FBS, penicillin, streptomycin, and L-glutamine.
Immunofluorescent staining
Cell lines were incubated with PE-conjugated control IgG1 or anti-CD84 mAb for 20–30 min on ice before washing with cold PBS containing 0.1% BSA and 0.1% sodium azide. To determine the expression of CD84 on cell lines as well as resting and stimulated T cells isolated from peripheral blood or spleens of healthy donors or XLP patients, cells were incubated with PE-conjugated control IgG1 or anti-CD84 mAb and either FITC-anti-CD4 or FITC-anti-CD8 mAb. CD4+ or CD8+ T cells were gated, and the fluorescence of the population of cells incubated with anti-CD84 mAb was compared with the fluorescence of cells incubated with control IgG1. The cells were analyzed on a FACScan flow cytometer using CellQuest software (BD Biosciences).
PCR, construction of retroviral expression plasmids, and cell transfection
The pMX plasmid containing N-terminal FLAG-tagged full-length wild-type (WT) human CD84 has been described previously (35). Tyrosine residues at positions 262 (Y262) and 298 (Y298) in the cytoplasmic domain of CD84 were mutated to phenylalanine (F) by site-directed mutagenesis using the following primers: Y262F, 5'-GCC TCA AAG AAA ACC ATA TTC AC; Y262F, 3'-GTG AAT ATG GTT TTC TTT GAG GC; Y298F, 5'-CCA GTG AAC ACA GTT TTT TCC G; Y298F, 3'-CGG AAA AAA CTG TGT TCA CTG G, in combination with 5'-GCA TGC ATC GAT GAC TCA GAA ATC TTC ACA GTG AA (ClaI restriction site underlined) and 3'-GCA TGC GCG GCC GCC CAG CAG CCT AGA TCA CAA TTT (NotI restriction site underlined) (35) primers located at the boundary of the open reading frame of CD84, to generate Y262F and Y298F constructs. The conditions for PCR were 30 cycles of 1-min denaturation at 94°C, 1-min annealing at 55°C, and 1-min extension at 72°C. The amplified products were digested with ClaI and NotI and cloned into the pMX-neo plasmid downstream of the FLAG epitope (DYKDDDK) and the CD8 leader sequence, generating FLAG-CD84 constructs (32). The Y262F mutant was used as template in PCR using Y298F 5' and 3' primers to generate the Y262/298F construct.
The pMX-puro construct containing a C-terminal c-myc epitope-tagged version of human SAP and human EWS/FLI1 activated transcript-2 (EAT-2) upstream of enhanced green fluorescence protein (eGFP) has been previously described (35). The arginine at position 32 (R32) was mutated to a lysine (K) by site-directed mutagenesis using the following primers: R32K, 5'-CTA TTT GCT GAA GGA CAG CGA G; and R32K, 3'-CTC GCT GTC CTT CAG CAA ATA G, in combination with 5'-GAA GAA GGA TCC GCC ATG GAC GCA GTG GCT G (BamHI restriction site underlined) and 3'-GCA TTA GAA TTC TGG GGC TTT CAG GCA GAC ATC (EcoRI restriction site underlined) (32) primers located at the boundary of the open reading frame of SAP, to generate a SAP R32K construct. The PCR-amplified product was digested with BamHI and EcoRI, then cloned into the pMX-puro retroviral expression vector upstream of an in-frame sequence encoding the c-Myc epitope (32), generating a human SAP R32Kmyc construct. This vector also contained a cDNA encoding eGFP (35). The nucleotide sequences of all mutated constructs were confirmed by automated DNA sequencing (performed by SUPAMAC, University of Sydney, Sydney, Australia).
Constructs were packaged using the Phoenix cell line (44), and virus was used to infect the mouse pre-B cell line Ba/F3 (21, 32). Infected Ba/F3 cells were selected using either puromycin or neomycin. The expression of CD84 and SAP or EAT-2 by the transfected cells was assessed by flow cytometry using PE-conjugated anti-CD84 mAb and eGFP fluorescence, respectively (35). The expression of SAP and EAT-2 cDNA in activated human T cells was assessed by PCR using primers and amplification conditions described previously (32, 35, 39).
Immunoprecipitation, SDS-PAGE, and Western blotting
To determine the effect of cross-linking CD84, human T cell lines, primary T cells, or PHA blasts were resuspended in PBS containing 0.5% BSA and incubated with a control IgG1 mAb or anti-CD84 mAb (5 µg/ml) for 30 min at 4° C (35). The cells were then washed three times, and bound mAb was cross-linked with F(ab')2 goat anti-mouse Ig (10 µg/ml) for 5 min at 37°C. Cells were then solubilized in ice-cold lysis buffer containing 10 mM Tris-HCl (pH 7.8), 1% Nonidet P-40, 150 mM NaCl, and enzyme inhibitors (35). Transfected Ba/F3 cells or F2F7 cells were resuspended in PBS and either untreated or treated with 100 µM sodium pervanadate prepared in 0.1% H2O2 (Sigma-Aldrich) for 5 min at room temperature (32). Cells were then solubilized in the above-described ice-cold lysis buffer. CD84 or SAP were immunoprecipitated from lysates using anti-CD84 mAb or anti-SAP polyclonal antiserum absorbed onto protein A-Sepharose beads (Sigma-Aldrich). To determine the expression of Lck in Jurkat cell lines and of SAP in PHA blasts, whole cell lysates were prepared by solubilizing the cells in ice-cold lysis buffer. Precipitated proteins or cell lysates were electrophoresed through 12% acrylamide gels containing 0.1% SDS and transferred to polyvinylidene difluoride membranes (Bio-Rad, Hercules, CA). Membranes were probed with Abs against phosphotyrosine, CD84, SAP, Myc, or Lck, followed by HRP-conjugated donkey anti-rabbit or anti-mouse-IgG antiserum. The membranes were developed using enhanced chemiluminescence (Pierce, Rockford, IL) and autoradiography.
Cell cultures
PHA blasts were generated by stimulating splenic or peripheral blood MNC with PHA (5 µg/ml) and IL-2 (10 U/ml; Endogen, Woburn, MA). Ninety-six-well, round-bottom culture plates (Costar, Cambridge MA) were coated with goat anti-mouse Ig (5 µg/ml, in 0.05 M carbonate buffer, pH 9.6) at 37°C for 6–8 h. Anti-CD84 mAb or a control IgG1 mAb (5 µg/ml) was added to the wells and incubated overnight at 37°C (45). T cell blasts were collected after 4 days of stimulation with PHA and IL-2, added to all wells (5 x 104/well), and cultured in a total volume of 200 µl either with no additional stimulus or with IL-2 (5 U/ml), anti-CD3 mAb (500 ng/ml), or anti-CD3 mAb plus IL-2. Proliferation was determined by assessing incorporation of [3H]thymidine (1 µCi/well; Amersham Pharmacia Biotech, Arlington Heights, IL) by quadruplicate cultures of T cells during the final 8 h of a 72-h culture period. Incorporation of [3H]thymidine was measured as counts per minute by liquid scintillation counting with a beta counter (Pharmacia-LKB-Wallac, Turku, Finland).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have recently reported that CD84 expressed by human B cell lines becomes tyrosine phosphorylated when ligated with anti-CD84 mAb (35). To determine whether this also occurred for other lymphoid cells, the human T cell lines F2F7 and Jurkat were examined. The expression of CD84 by these immortalized T cell lines was similar to primary CD4+ and CD8+ T cells present in human peripheral blood and spleen (Fig. 1, a–d) (35). Jurkat and F2F7 cells were labeled with a control mAb or anti-CD84 mAb, then treated with anti-mouse Ig Ab to induce cross-linking. The cells were lysed, CD84 was immunoprecipitated, and the phosphorylation status of CD84 was assessed. In cells treated with a control mAb, immunoprecipitated CD84 was not phosphorylated (Fig. 2a, upper panel). However, following receptor ligation, an 
85 kDa protein phosphorylated on tyrosine was detected in CD84 immunoprecipitates from both T cell lines (Fig. 2a, upper panel). Probing duplicate blots with anti-CD84 mAb revealed this phosphoprotein to be CD84 (Fig. 2a, lower panel). To determine whether CD84 became tyrosine phosphorylated in primary cells, normal splenic MNC were expanded with PHA and IL-2 for 10–14 days to generate PHA blasts. After this time, both CD4+ and CD8+ T cell blasts continued to express CD84, with the level being increased on activated CD8+ T cells compared with unstimulated cells (Fig. 1, c–f). Immunoprecipitation of CD84 from lysates of PHA T cell blasts indicated that CD84 was also specifically tyrosine phosphorylated in these cells following treatment with anti-CD84 mAb, but not a control mAb (Fig. 2b). Thus, an early event following ligation of CD84 is induction of phosphorylation on tyrosine residues.
|
|
By using transfectants and GST fusion proteins, we and others have demonstrated that CD84, like 2B4 and SLAM, can interact with SAP (33, 34, 35). A limitation of the transfectants used in our previous study was that following overexpression, CD84 became constitutively phosphorylated on tyrosines (35), presumably as a result of homophilic interactions between CD84 expressed on nearby cells (19), a phenomenon also observed for SLAM transfectants (46, 47). It was therefore not possible to determine whether recruitment of SAP to CD84 was dependent on phosphorylation of the cytoplasmic domain of CD84. It was thus necessary to establish whether this interaction 1) required CD84 to be phosphorylated and 2) occurred in human immune cells expressing endogenous SAP and CD84. For these experiments CD84 was phosphorylated by treating F2F7 cells with anti-CD84 mAb or sodium pervanadate, an inhibitor of protein tyrosine phosphatases that induces hyperphosphorylation of proteins on tyrosine residues (48, 49). When CD84 was immunoprecipitated from F2F7 cells treated with anti-CD84, but not control mAb, a phosphoprotein corresponding to CD84 was detected (Fig. 2c, upper and middle panels, lanes 1 and 2), but it did not associate with endogenous SAP (Fig. 2c, lower panel, lane 2). However, when F2F7 cells were treated with sodium pervanadate, SAP did associate with phosphorylated CD84 (Fig. 2c, lower panel, lane 4). The failure to observe an association between CD84 and SAP in cells stimulated with anti-CD84 mAb in this experiment may reflect the lower level of phosphorylation of CD84 compared with that induced by sodium pervanadate (Fig. 2c, upper panel, compare lanes 2 and 4). To investigate this further, F2F7 cells were stimulated with control Ig or anti-CD84 mAb; the resulting lysates were immunoprecipitated with anti-CD84, anti-SAP, or the appropriate control Ab; and the presence of phosphorylated CD84 in the different precipitates was determined. Phosphorylated CD84 was detected in anti-CD84 mAb (Fig. 2d, left panel, lane 4) as well as anti-SAP Ab (Fig. 2d, right panel, lane 8) immunoprecipitates of F2F7 cells following stimulation with anti-CD84 mAb, but not following treatment with a control Ig (Fig. 2d, lanes 3 and 7). The amount of phosphorylated CD84 precipitated with anti-SAP Ab appeared to be less than with anti-CD84 mAb (Fig. 2d, compare lanes 4 and 8). Indeed, when duplicate blots were probed for CD84, the amount detected in the anti-SAP Ab precipitates was much less than in the anti-CD84 mAb precipitates (data not shown). The small amount of phosphorylated CD84 detected in the control precipitates of anti-CD84 mAb-treated cells (Fig. 2d, lanes 2 and 6) is due to immunoprecipitation by the stimulating mAb, some of which becomes immobilized to protein A during the immunoprecipitation procedure, rather than nonspecific immunoprecipitation by the control Ig, as well as increased exposure times of the membranes during development. Thus, SAP is recruited to CD84 in human immune cells, and this interaction is dependent on phosphorylation of tyrosines present in the cytoplasmic domain of CD84.
The Src tyrosine kinase Lck is involved in tyrosine phosphorylation of CD84
The molecular requirements for tyrosine phosphorylation of CD84 were investigated by comparing the parental Jurkat cell line with cell lines lacking expression of protein tyrosine kinases Lck (JCam.1) (42) and ZAP-70 (P116) (43). While the expression of CD84 on these mutant cell lines was comparable to that of the parental Jurkat line (data not shown), Lck was clearly absent from JCam.1 cells (Fig. 3a). Similarly, P116 cells are devoid of ZAP-70 (43). CD84 was tyrosine phosphorylated to a comparable extent following ligation on parental Jurkat T cells and on P116 cells (Fig. 3b). However, CD84 did not become tyrosine phosphorylated following ligation on JCam.1 cells (Fig. 3b). When JCam.1 cells were reconstituted with Lck, the level of tyrosine phosphorylation of CD84 was restored to that observed for P116 (Fig. 3c) and parental Jurkat cells (Fig. 3b). Thus, the Src family protein tyrosine kinase Lck is involved in the phosphorylation of CD84 in Jurkat cell lines.
|
Four tyrosine residues are present in the cytoplasmic domain of CD84 at amino acid positions 262, 279, 298, and 324 (5) (Fig. 4a). Two of these tyrosine residues (Y262 and Y298) are embedded within the putative SAP-binding motif, TxYxxV/I (i.e., TIY262TYI, TVY298SEV; Fig. 4a). To identify which of these tyrosines is involved in the recruitment of SAP, they were mutated to phenylalanine (F), either singly or in combination (Fig. 4a). The mouse pre-B cell line Ba/F3 was then transduced with the resulting CD84 Y262F, CD84 Y298F, and CD84 Y262/298F retroviral constructs as well as with a retroviral construct encoding SAPmyc (35). Ba/F3 cells were used for these studies because they are readily transduced with retroviruses for ectopic protein expression (44), thus facilitating the analysis of protein-protein interactions (50). They also lack the expression of endogenous SAP (data not shown), thus avoiding any competition with transduced SAP for binding to CD84. Following selection and cell sorting, the levels of expression of CD84 and SAP by these cells were similar, as revealed by flow cytometry (Fig. 4b) and Western blotting (Fig. 4c). When WT CD84 was overexpressed in Ba/F3 cells, it became constitutively phosphorylated on tyrosine and recruited SAP (Fig. 4d, lane 2) (35). Mutating Y262 or Y298 to phenylalanine reduced the level of basal tyrosine phosphorylation of CD84 compared with WT CD84, while mutating both tyrosine residues to phenylalanine abrogated phosphorylation in untreated cells (Fig. 4d, upper panel, compare lanes 2, 5, 9, and 13). Despite reduced tyrosine phosphorylation, CD84 Y298F was still capable or recruiting SAP (Fig. 4d, lower panel, lane 9). In contrast, the ability of CD84 to do so was abolished by the Y262F mutation (Fig. 4d, lower panel, lane 5). Consistent with this finding, SAP failed to bind to the double tyrosine mutant (Fig. 4d, lower panel, lane 13).
|
The interaction between SH2 domain-containing proteins and phosphotyrosine residues usually involves an arginine (R) residue located within the SH2 domain (51). For human SAP, this is located at position 32, within the consensus sequence F/Y-L-L/I/V-R*D-S (28, 29, 30). To determine the role of R32 in the CD84-SAP interaction, it was mutated to lysine (K) by site-directed mutagenesis (SAP R32K). Stable Ba/F3 cell transfectants expressing CD84 only, both CD84 and WT SAP (CD84/WT SAP), or CD84 and SAP R32K (CD84/SAP R32K) were generated, and associations between CD84 and SAP were determined. WT SAP was immunoprecipitated from lysates of untreated and sodium pervanadate-treated CD84/SAP transfectants with anti-CD84 mAb, but not from transfectants expressing CD84 only or CD84 and SAP R32K (Fig. 5a, anti-human SAP blot), despite precipitation of comparable amounts of (p)CD84 (Fig. 5a, anti-phosphotyrosine and anti-CD84 blots). In a reciprocal experiment CD84 was detected in anti-SAP Ab immunoprecipitates of lysates prepared from CD84/WT SAP transfectants (Fig. 5b, anti-phosphotyrosine blot), but not from transfectants expressing CD84 only or CD84 and SAP R32K (Fig. 5b, anti-phosphotyrosine blot), although comparable amounts of SAP were precipitated from the WT and R32K SAP-expressing cell lines (Fig. 5b, anti-SAP blot). Thus, R32 of SAP is required for its interaction with CD84.
|
EAT-2 is a single SH2-domain containing protein (52) exhibiting 45% homology with SAP that is also capable of being recruited to the cytoplasmic domains of CD84, SLAM, 2B4, and Ly9. This interaction is believed to occur via a similar tyrosine-based motif, i.e., TxYxxI/V (31, 35, 36, 53, 54). To determine whether EAT-2 is recruited to the same motif in the cytoplasmic domain of CD84 as SAP, Ba/F3 cells expressing WT or tyrosine mutants of CD84 were transduced to express a C-terminal Myc-tagged version of human EAT-2 (35). Following selection and cell sorting, the resulting transfectants expressed comparable levels of CD84 and EAT-2 (data not shown). The transfectants were untreated or treated with sodium pervanadate as described above for the CD84/SAP transfectants, and the ability of EAT-2 to be recruited to CD84 was determined by immunoprecipitating with anti-CD84 mAb and assessing for the presence of coassociating EAT-2myc. In contrast to SAP, EAT-2 remained capable of binding to CD84 when Y262 or Y298 was mutated to phenylalanine (Fig. 4e, lanes 5 and 6, and 9 and 10), yet the level of binding appeared to be lower than to WT CD84 (Fig. 4e, compare lanes 2, 5, and 9). However, binding was completely abrogated when both tyrosine residues were mutated (Fig. 4e, lanes 13 and 14). Thus, there is functional redundancy in the tyrosine-based motifs used by EAT-2 for binding to CD84, with EAT-2 being able to interact with Y262 and Y298.
Induction of tyrosine phosphorylation of CD84 does not require SAP expression
Recent studies have revealed that SLAM becomes tyrosine phosphorylated when ligated, and this event requires coexpression of SAP (21, 46, 47, 55). In contrast, tyrosine phosphorylation of CD84 was shown to occur in transfectants expressing CD84 only (Figs. 4d and 5a) or CD84 and SAP R32K, where binding to SAP was abrogated (Fig. 5a), suggesting that the expression and/or function of SAP was not required for the induction of CD84 phosphorylation. Furthermore, CD84 was tyrosine phosphorylated in SAP-deficient human B cell lines (35), again indicating that SAP expression may not be necessary for CD84 to become phosphorylated following receptor ligation. To investigate the requirement for SAP in the induction of phosphorylation of CD84 in primary T cells, PHA blasts were generated from MNC isolated from two patients with XLP with missense mutations present in their SAP genes. Western blotting of activated cell lysates revealed that these mutations greatly reduced or abolished the expression of SAP (Fig. 6a). However, the level of phosphorylation of CD84 in PHA blasts generated from the XLP patients following ligation of CD84 was comparable to that observed in F2F7 cells (Fig. 6b, upper panel) as well as normal PHA blasts (see Fig. 2b), suggesting that SAP is not required for this initial signaling event to occur.
|
Signals delivered through 2B4, SLAM, NTB-A, and CRACC enhance the responses of NK cells and T cells (4, 8, 12, 20, 21, 22, 23, 24, 25). To test whether CD84 contributed to lymphocyte activation, freshly isolated T cells or PHA blasts were cultured with immobilized anti-CD84 mAb alone or in combination with IL-2 and/or anti-CD3 mAb, and cell proliferation was determined by the uptake of [3H]thymidine. Ligation of CD84 alone on primary T cells or PHA blasts did not induce proliferation, nor did it enhance the proliferative response of primary T cells to anti-CD3 mAb alone or in the presence of IL-2 (data not shown). However, the proliferation of PHA blasts induced by anti-CD3 mAb was enhanced 3- to 10-fold following coligation of CD84 (Fig. 7a). The level of proliferation induced by anti-CD3 and anti-CD84 mAb approximated or exceeded that induced by anti-CD3 mAb plus IL-2 (Fig. 7a). Consequently, proliferation of PHA blasts induced by anti-CD3 mAb plus IL-2 was only slightly increased (1.5- to 2-fold) by culture in the presence of immobilized anti-CD84 mAb.
|
2-fold in the presence of immobilized anti-CD84 mAb. Thus, although CD84 is capable of augmenting proliferative responses of activated human T cells, SAP does not appear to be required for this process. Activated human T cells express EAT-2
The finding that activated T cells from XLP patients were capable of responding to stimulation with anti-CD84 mAb suggested that a SAP homologue, such as EAT-2 (52), may substitute for its function in SAP-deficient lymphocytes. EAT-2 is predominantly expressed by APC, such as B cells and myeloid cells (31, 35, 53); however, its expression in T cells remains to be examined. For this reason, the expression of EAT-2 cDNA in purified human CD4+ and CD8+ T cells that had been stimulated in vitro with PHA and IL-2 was assessed by PCR. Although SAP was clearly evident in both populations of activated T cells as well as activated MNC (Fig. 8a), EAT-2 transcripts were also detectable in these cells (Fig. 8b). The lack of available reagents (i.e., anti-EAT-2 Ab) precluded the assessment of expression of EAT-2 protein in primary human cells and cell lines, as well as analysis of the association between endogenous CD84 and EAT-2. This notwithstanding, these data suggest that the expression of EAT-2 may compensate for SAP deficiency in signaling pathways elicited via SAP-associating receptors in activated T cells.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In this study we found that CD84 expressed by human T cells rapidly became tyrosine phosphorylated following receptor ligation (Fig. 2) by a SAP-independent mechanism (Fig. 6). This was observed repeatedly for different human T cell lines as well as for activated primary T cells isolated from multiple different normal donors and XLP patients. The kinetics of CD84 phosphorylation in T cells resembled that previously observed in human B cell lines (35), since it was found to be initiated within 1 min of receptor ligation, to peak after 5 min, and then return to basal levels after 60 min (data not shown). CD84 phosphorylation did not occur in Lck-deficient Jurkat cells, suggesting that Lck is involved in the phosphorylation of CD84 in this cell line (Fig. 3). On the other hand, the observed reduction in CD84 tyrosine phosphorylation in the absence of Lck could have represented an indirect effect of the requirement for Lck in the phosphorylation and subsequent activation of another kinase such as ZAP70 (58). However, this appears unlikely because the level of phosphorylation of CD84 in the ZAP70-deficient Jurkat cell line was similar to that in WT Jurkat cells (Fig. 3). Interestingly, CD84 did not become tyrosine phosphorylated when ligated on resting primary T cells (data not shown), which points to a dependence of this event on prior activation. This possibility is consistent with the earlier observation of increases in the levels of activity of the Src kinases Lck and Fyn following stimulation of human T cell lines (49). In support of our finding of a role for Lck in CD84 tyrosine phosphorylation in T cells is the recent demonstration of Lck-mediated tyrosine phosphorylation of the related molecule 2B4 following treatment of transfected Jurkat cells with sodium pervanadate (38) and that coexpression of Lck and CD84 in 293 cells resulted in tyrosine phosphorylation of CD84 (34). The kinase required for phosphorylating SLAM has not been formally identified; however, SLAM becomes tyrosine phosphorylated when expressed in COS-7 cells cotransfected with constitutively active Fyn (30), while SLAM is weakly phosphorylated or fails to become tyrosine phosphorylated in response to stimulation with anti-SLAM Ab in thymocytes isolated from Fyn-deficient mice (46, 55). Based on our data, as well as those presented by Nakajima et al. (38), Lck may also mediate tyrosine phosphorylation of SLAM under physiological conditions. Indeed, it was recently proposed that the residual level of SLAM tyrosine phosphorylation observed in Fyn-/- thymocytes may be attributable to Lck, which is also expressed in these cells (55).
Preliminary studies have assessed additional downstream events that occur following ligation of CD84 on human T cells. In addition to CD84 itself undergoing tyrosine phosphorylation, a phosphoprotein of 60 kDa has been detected in lysates of F2F7 cells stimulated with anti-CD84 mAb, but not with control Ig (data not shown). The molecular mass of this protein together with recent data demonstrating that ligation of SLAM induced activation of FynT (46, 47) suggest that it could be a Src kinase. However, we have been unable to demonstrate phosphorylation of either Lck or Fyn in activated F2F7 cells (data not shown). Alternatively, it may be another adaptor molecule such as Dok1 (62 kDa), Dok2 (56 kDa), or Shc (52 kDa), all of which have been reported to associate with or become activated following lymphocyte activation through SLAM (46). The identification of this phosphoprotein as well as other downstream signaling events elicited in response to activation through CD84 are currently being pursued.
Consistent with previous studies (33, 34, 35), CD84 was capable of recruiting SAP (Fig. 2). However, our results extended these findings by demonstrating an association between CD84 and SAP when expressed endogenously in a human T cell line (Fig. 2, c and d), thus obviating the need to use transfected cell lines overexpressing these proteins (33, 35) or GST fusion proteins (34). According to our data, the CD84-SAP interaction was 1) dependent on tyrosine phosphorylation of CD84 (Fig. 2, c and d) and 2) mediated by a specific interaction between the SAP-binding motif located at Y262 in the cytoplasmic domain of CD84 (Fig. 4d) and R32 in the SH2 domain of SAP (Fig. 5). Thus, although there are two consensus SAP binding sites within the cytoplasmic domain of CD84 (Y262, Y298), only one (Y262) appears to be critical for SAP recruitment. The requirement for a single tyrosine-based motif for recruitment of SAP to the cytoplasmic domain of CD84 resembles the mechanism of binding of SAP to SLAM, where it was found that Y281 in human SLAM (Y288 in mouse SLAM) mediated the SLAM-SAP association (30, 46). Interestingly, the SAP homologue EAT-2 was capable of binding to either Y262 and Y298 (Fig. 4e). Although the amino acids comprising these tyrosine-based motifs are similar, an important difference is the amino acid at the -1 position relative to the tyrosine. The sequence surrounding Y262 is TIYTYI, while at Y298 it is TVYSEV (see Fig. 4a). Thus, the isoleucine (I) upstream of Y262 appears to confer specificity for SAP, while binding of EAT-2 can be achieved in the presence of either an isoleucine or a valine (V) residue. This is consistent with data derived from screening of peptide libraries that demonstrated SAP preferentially bound to the TIpYxxV/I motif, while EAT-2 could bind to either a TIpYxxV/I or a TVpYxxV/I motif (36). The TVpYxxV/I motif is also present in the cytoplasmic domains of SLAM, 2B4, NTB-A, Ly9, and CRACC (3); however, its function in these receptors is unknown. SLAM, 2B4, CD84, NTB-A, and Ly9 have been found to recruit other SH2 domain-containing proteins, such as EAT-2, and the phosphatases SH2-containing inositol polyphosphate 5-phosphatase, Src homology protein tyrosine phosphatase-1, and Src homology protein tyrosine phosphatase-2 (33, 34, 37, 46, 53, 54, 59) and data not shown). Consequently, the TVpYxxV/I motif may be important for the recruitment of these additional mediators of intracellular signaling pathways to the cytoplasmic domains of SLAM, NTB-A, Ly9, and CD84. Furthermore, the absence of a TIpYxxV/I motif from CRACC may explain the inability of SAP to bind this receptor (12).
SAP is capable of binding to both phosphorylated as well as nonphosphorylated isoforms of SLAM (30). In contrast, recruitment of SAP to the cytoplasmic domains of CD84 (Fig. 2) as well as to 2B4, Ly9, and NTB-A (8, 32, 33), is strictly dependent on tyrosine phosphorylation. Thus, SLAM is unique in its requirements for binding to SAP, whereas tyrosine phosphorylation of CD84, 2B4, and Ly9 is a prerequisite for such a binding event to occur. The finding that SAP R32 mediated the interaction with phosphorylated CD84 is consistent with studies investigating binding of members of the CD2 IgSF to SAP molecules that contain specific mutations at this residue. XLP patients have been identified with mutations affecting R32, resulting in SAP R32T (34) and SAP R32Q proteins (60). Both these SAP mutants failed to associate with CD84, SLAM, or Ly9 (34, 60), confirming the critical role played by this residue in mediating the interaction between SAP and members of the CD2 IgSF.
An important consequence of ligating CD84 was the ability to enhance the proliferation of anti-CD3 mAb-stimulated T cell blasts (Fig. 7). These results are consistent with those reported recently demonstrating that ligation of CD84 on anti-CD3 mAb-stimulated T cells led to increased IFN-
production (19). The level of proliferation induced by anti-CD3 in conjunction with anti-CD84 mAb approximated or exceeded that induced by anti-CD3 mAb plus IL-2 (Fig. 7a), suggesting the enhancing effect of anti-CD84 mAb may be mediated by increased production of endogenous cytokines, such as IL-2, as has been observed for 2B4 (20, 23), CD84 (19), and SLAM (4, 21). The effect of ligating CD84 on cytokine production by anti-CD3 mAb-stimulated T cells from both normal healthy donors and XLP patients as well as the effect of coligating CD3 and CD84 on signaling pathways elicited via these molecules are currently under investigation. The inability of anti-CD84 mAb to induce the proliferation of resting primary T cells is further evidence that T cells need to be preactivated for the stimulatory effects of ligating CD84 to be realized. The facts that anti-SLAM mAb increases proliferation and IFN- secretion of human T cell blasts and T cells clones, but not of resting T cells (24), and anti-2B4 mAb elicited cytotoxicity by activated, but not resting, NK cells (6) indicate that a common feature of members of the CD2 IgSF is their ability to promote responses of activated, but not resting, lymphocytes. Although CD84 was capable of recruiting SAP, a deficiency of SAP did not appear to compromise the ability of anti-CD84 mAb to enhance the proliferation of stimulated XLP T cells (Fig. 7b). Consistent with this finding are recent studies demonstrating that receptors belonging to the CD2 IgSF are also capable of exerting their effector function independently of SAP. For instance, anti-SLAM mAb has been shown to increase proliferation and IFN- production by both WT and SAP-deficient murine T cells stimulated with anti-CD3 mAb (61), and ligation of SLAM on transfectants induced activation of extracellular signal-regulated kinases 1/2 (ERK1/2) in the absence of SAP (2, 62). Similarly, CRACC can activate human NK cells from both normal individuals and XLP patients (12). Lastly, although SAP was capable of associating with Dok1 and activating NF-
B, mutated versions of SAP from XLP patients remained capable of activating NF-B in this system, suggesting not only that NF-B activation may be independent of WT SAP, but also that signaling pathways using NF-B may not be involved in the pathogenesis of XLP (63). Thus, CD84 may be capable of eliciting multiple signaling pathways, not all of which involve SAP, even though each can result in distinct effector functions. Alternatively, the residual amount of SAP expressed in lymphocytes from XLP patients used in this study may have been sufficient to maintain CD84 effector function, or homologues of SAP, such as EAT-2 (31, 35, 52, 53), may be able to substitute functionally for SAP in situations of SAP deficiency. Indeed, EAT-2 transcripts could be detected in activated CD4+ and CD8+ T cells (Fig. 8). The expression of EAT-2 in activated T cells may also explain the ability of anti-mouse SLAM mAb to enhance the proliferation and IFN- production by SAP-deficient T cells (61).
The molecular mechanism(s) underlying lymphocyte activation via SAP-associating receptors in the absence of SAP is unknown; however, it may result from divergent signaling pathways and effector functions and the ability of these receptors to interact with multiple intracellular molecules (2, 35, 37, 46, 47, 53, 54, 64). There are several precedents for such a suggestion. First, agonistic anti-2B4 mAb has been shown to induce cytotoxicity via transcription-dependent and mitogen-activate protein kinase (p38, ERK1/2)-dependent pathways, whereas IFN- secretion in response to the same stimulus was only dependent on p38 (64). Secondly, 2B4 has been found to associate constitutively with linker for activation of T cells, and ligation of 2B4 induced phosphorylation of 2B4, followed by recruitment of PLC and Grb2 (64, 65). Thirdly, although SLAM was found to activate ERK1/2 in transfectants in the absence of SAP, introduction of SAP into these cell lines resulted in activation of the Akt signaling pathway (2, 62). Lastly, elegant studies by Latour et al. (46, 47) demonstrated that signaling through SLAM induced phosphorylation and subsequent activation of a number of different signaling intermediates, including SHIP, Dok1, Dok2 and FynT, which depended on the expression of SAP. Consequently, the exact downstream mediators involved in signaling via CD84 and other CD2 IgSF, and their contributions to activation events that occur in the presence and the absence of SAP remain to be elucidated. Further investigation of the mechanisms of lymphocyte activation mediated via SAP-associating receptors will facilitate a greater understanding of immune regulation not only in patients with XLP, but in normal individuals as well.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Stuart Tangye, Centenary Institute of Cancer Medicine and Cell Biology, Locked Bag #6, Newtown 2042, New South Wales, Australia. E-mail address: s.tangye{at}centenary.usyd.edu.au
3 Abbreviations used in this paper: IgSF, Ig superfamily; CRACC, CD2-like receptor-activating cytotoxic cells; eGFP, enhanced green fluorescent protein; ERK, extracellular signal-regulated kinase; MNC, mononuclear cells; p, phosphorylated; SAP, SLAM-associated protein; SH2, Src homology 2; SLAM, signaling lymphocytic activation molecule; XLP, X-linked lymphoproliferative syndrome; WT, wild type; EAT-2, EWS/FLI1 activated transcript-2.
Received for publication February 14, 2003. Accepted for publication June 27, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production. Blood 100:2899.B. Proc. Natl. Acad. Sci. USA 97:7470. release in human NK cells involves distinct pathways. J. Immunol. 167:6210.This article has been cited by other articles:
|
|
 |
|
  J. A. Wahle, K. H. T. Paraiso, R. D. Kendig, H. R. Lawrence, L. Chen, J. Wu, and W. G. Kerr Inappropriate Recruitment and Activity by the Src Homology Region 2 Domain-Containing Phosphatase 1 (SHP1) Is Responsible for Receptor Dominance in the SHIP-Deficient NK Cell J. Immunol., December 15, 2007; 179(12): 8009 - 8015. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Eissmann and C. Watzl Molecular Analysis of NTB-A Signaling: A Role for EAT-2 in NTB-A-Mediated Activation of Human NK Cells. J. Immunol., September 1, 2006; 177(5): 3170 - 3177. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. J. Hare, C. S. Ma, F. Alvaro, K. E. Nichols, and S. G. Tangye Missense mutations in SH2D1A identified in patients with X-linked lymphoproliferative disease differentially affect the expression and function of SAP Int. Immunol., July 1, 2006; 18(7): 1055 - 1065. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M-C Zhang, N Misu, H Furukawa, Y Watanabe, M Terada, H Komori, T Miyazaki, M Nose, and M Ono An epistatic effect of the female specific loci on the development of autoimmune vasculitis and antinuclear autoantibody in murine lupus Ann Rheum Dis, April 1, 2006; 65(4): 495 - 500. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Bhat, P. Eissmann, J. Endt, S. Hoffmann, and C. Watzl Fine-tuning of immune responses by SLAM-related receptors J. Leukoc. Biol., March 1, 2006; 79(3): 417 - 424. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. H.J. Looijenga, R. Hersmus, A. J.M. Gillis, R. Pfundt, H. J. Stoop, R. J.H.L.M. van Gurp, J. Veltman, H. B. Beverloo, E. van Drunen, A. Geurts van Kessel, et al. Genomic and Expression Profiling of Human Spermatocytic Seminomas: Primary Spermatocyte as Tumorigenic Precursor and DMRT1 as Candidate Chromosome 9 Gene Cancer Res., January 1, 2006; 66(1): 290 - 302. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Tassi and M. Colonna The Cytotoxicity Receptor CRACC (CS-1) Recruits EAT-2 and Activates the PI3K and Phospholipase C{gamma} Signaling Pathways in Human NK Cells J. Immunol., December 15, 2005; 175(12): 7996 - 8002. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Nanda, P. Andre, M. Bao, K. Clauser, F. Deguzman, D. Howie, P. B. Conley, C. Terhorst, and D. R. Phillips Platelet aggregation induces platelet aggregate stability via SLAM family receptor signaling Blood, November 1, 2005; 106(9): 3028 - 3034. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Chen and A. C. Ross Inaugural Article: Vitamin A and immune function: Retinoic acid modulates population dynamics in antigen receptor and CD38-stimulated splenic B cells PNAS, October 4, 2005; 102(40): 14142 - 14149. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Selvaraj, D. Bunick, C. Finnigan-Bunick, R. W. Johnson, H. Wang, L. Liu, and P. S. Cooke Gene Expression Profiling of 17{beta}-Estradiol and Genistein Effects on Mouse Thymus Toxicol. Sci., September 1, 2005; 87(1): 97 - 112. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Dupre, G. Andolfi, S. G. Tangye, R. Clementi, F. Locatelli, M. Arico, A. Aiuti, and M.-G. Roncarolo SAP controls the cytolytic activity of CD8+ T cells against EBV-infected cells Blood, June 1, 2005; 105(11): 4383 - 4389. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Chtanova, S. G. Tangye, R. Newton, N. Frank, M. R. Hodge, M. S. Rolph, and C. R. Mackay T Follicular Helper Cells Express a Distinctive Transcriptional Profile, Reflecting Their Role as Non-Th1/Th2 Effector Cells That Provide Help for B Cells J. Immunol., July 1, 2004; 173(1): 68 - 78. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), anti-CD3 mAb plus immobilized control Ig (
), or anti-CD3 mAb plus immobilized anti-CD84 mAb (
) in the absence (-) or the presence (+) of 10 U/ml human IL-2. b and c, PHA blasts generated from MNC of normal healthy donors (Normal) and from two different XLP patients (XLP 1 and 3) were cultured with anti-CD3 mAb alone or with immobilized control Ig (