Joan Sayós

Our understanding of the X-linked lymphoproliferative syndrome (XLP) has advanced significantly in the last two years. The gene that is altered in the condition (SAP/SH2D1A) has been cloned and its protein crystal structure solved. At least two sets of target molecules for this small SH2 domain-containing protein have been identified: A family of hematopoietic cell surface receptors, i.e. the SLAM family, and a second molecule, which is a phosphorylated adapter. A SAP-like protein, EAT-2, has also been found to interact with this family of surface receptors. Several lines of evidence, including structural studies and analyses of missense mutations in XLP patients, support the notion that SAP/SH2D1A is a natural inhibitor of SH2-domain-dependent interactions with members of the SLAM family. However, details of its role in signaling mechanisms are yet to be unravelled. Further analyses of the SAP/SH2D1A gene in XLP patients have made it clear that the development of dys-gammaglobulinemia and B cell lymphoma can occur without evidence of prior EBV infection. Moreover, preliminary results of virus infections of a mouse in which the SAP/SH2D1A gene has been disrupted suggest that EBV infection is not per se critical for the development of XLP phenotypes. It appears therefore that the SAP/SH2D1A gene controls signaling via the SLAM family of surface receptors and thus may play a fundamental role in T cell and APC interactions during viral infections.

X-linked lymphoproliferative disease (XLP) is a rare immune disorder commonly triggered by infection with Epstein-Barr virus. Major disease manifestations include fatal acute infectious mononucleosis, B-cell lymphoma, and progressive dys-gammaglobulinemia. SAP/SH2D1A, the product of the gene mutated in XLP, is a small protein that comprises a single SH2 domain and a short tail of 26 amino acids. SAP binds to a specific motif in the cytoplasmic tails of the cell surface receptors SLAM and 2B4, where it blocks recruitment of the phosphatase SHP-2. Here it is reported that Ly-9 and CD84, 2 related glycoproteins differentially expressed on hematopoietic cells, also recruit SAP. Interactions between SAP and Ly-9 or CD84 were analyzed using a novel yeast 2-hybrid system, by COS cell transfections and in lymphoid cells. Recruitment of SAP is most efficient when the specific tyrosine residues in the cytoplasmic tails of Ly-9 or CD84 are phosphorylated. It is concluded that in activated T cells, the SAP protein binds to and regulates signal transduction events initiated through the engagement of SLAM, 2B4, CD84, and Ly-9. This suggests that combinations of dysfunctional signaling pathways initiated by these 4 cell surface receptors may cause the complex phenotypes of XLP. (Blood. 2001;97:3867-3874)

SAP, the gene that is altered or absent in the X-linked lymphoproliferative syndrome (XLP), encodes a small protein that comprises a single SH2 domain and binds to the cell-surface protein SLAM which is present on activated or memory T and B cells. Because defective NK cell activity also has been reported in XLP patients, we studied the SAP gene in NK cells. SAP was induced upon viral infection of SCID mice and shown to be expressed in NK cells by in vitro culturing in the presence of IL-2. Moreover, SAP was expressed in the NK cell lines YT and RNK 16. Because SLAM, the cell-surface protein with which SAP interacts, and 2B4, a membrane protein having sequence homologies with SLAM, also were found to be expressed on the surfaces of activated NK and T cell populations, they may access SAP functions in these populations. Whereas we found that 2B4 also binds SAP, 2B4-SAP interactions occurred only upon tyrosine phosphorylation of 2B4. By contrast, SLAM-SAP interactions were independent of phosphorylation of Y281 and Y327 on SLAM. As CD48, the ligand for 2B4, is expressed on the surface of Epstein-Barr virus (EBV)-infected B cells, it is likely that SAP regulates signal transduction through this pair of cell-surface molecules. These data support the hypothesis that XLP is a result of both defective NK and T lymphocyte responses to EBV. The altered responses may be due to aberrant control of the signaling cascades which are initiated by the SLAM-SLAM and 2B4-CD48 interactions.