Regulation of NKT cell development by SAP, the protein defective in XLP

Abstract

The adaptor molecule SAP is expressed in T lymphocytes and natural killer (NK) cells, where it regulates cytokine production and cytotoxicity. Here, we show that SAP, encoded by the SH2D1A gene locus, also has a crucial role during the development of NKT cells, a lymphocyte subset with immunoregulatory functions in response to infection, cancer and autoimmune disease. Following stimulation with the NKT cell-specific agonist alpha-galactosyl ceramide (alphaGC), Sh2d1a-/- splenocytes did not produce cytokines or activate other lymphoid lineages in an NKT cell-dependent manner. While evaluating the abnormalities in alphaGC-induced immune responses, we observed that Sh2d1a-/- animals lacked NKT cells in the thymus and peripheral organs. The defect in NKT cell ontogeny was hematopoietic cell autonomous and could be rescued by reconstitution of SAP expression within Sh2d1a-/- bone marrow cells. Seventeen individuals with X-linked lymphoproliferative disease (XLP), who harbored germline mutations in SH2D1A, also lacked NKT cells. Furthermore, a female XLP carrier showed completely skewed X chromosome inactivation within NKT cells, but not T or B cells. Thus, SAP is a crucial regulator of NKT cell ontogeny in humans and in mice. The absence of NKT cells may contribute to the phenotypes of SAP deficiency, including abnormal antiviral and antitumor immunity and hypogammaglobulinemia.

Figure 1

Absence of αGC induced immune cell activation in Sh2d1a^−/− mice. (a) Sh2d1a mRNA expression in wild-type (WT) and Sh2d1a^−/− thymocytes, and purified T, B and NKT cells. (b) In vitro αGC-induced cytokine secretion by wild-type and Sh2d1a^−/− splenocytes. (c) In vivo αGC-induced cytokine response in wild-type and Sh2d1a^−/− mice, as measured by RNase protection. (d,e) Activation of immune cells in vehicle- (filled histograms) or αGC-injected (open histograms) wild-type and Sh2d1a^−/− mice. (d) Expression of CD69 on splenocytes. The percentage of CD69⁺ cells following αGC injection is shown. (e) Intracellular IFN-γ production by TCRβ^– NK1.1⁺ NK cells. IFN-γ-positive cells are shown as a percentage of total splenocytes (dot plots) or NK cells (histograms).

Figure 2

Sh2d1a^−/− mice lack NKT cells. (a) Flow cytometric plots showing the percentages of HSA^low Tet⁺ TCRβ⁺ NKT cells in wild-type, Sh2d1a^−/−, Fyn^−/− and Cd1d1^−/− mice. (b) Absolute numbers of NKT cells in the thymi and spleens of wild-type (n = 10) and Sh2d1a^−/− mice (n = 10). Differences in NKT cell number were statistically significant in both organs; Mann-Whitney test, P < 0.001. (c) CD1d expression on thymocytes and splenocytes from wild-type (open histogram), Sh2d1a^−/− (dotted open histogram) and Cd1d1^−/− (filled histogram) mice. (d) IL-2 secretion by DN3A4–1.2 Vα14⁺ NKT cell hybridoma cells after culture with vehicle or αGC-pulsed wild-type, Sh2d1a^−/− or Cd1d1^−/− splenocytes.

Figure 3

The defect in NKT cell ontogeny is hematopoietic cell-autonomous and rescued by expression of wild-type SAP. (a,b) Bone marrow chimeric mice (n = 12) were generated using CD45.1⁺ wild-type and CD45.1^– Sh2d1a^−/− cells. (a) Flow cytometric plots showing percent donor chimerism (top) and percent HSA^low Tet⁺ TCRβ⁺ NKT cells in CD45.1⁺ (wild-type) and CD45.1^– (Sh2d1a^−/−) lymphocyte populations (bottom). (b) Percentages of B cells and T cell subsets in bone marrow chimeras. (c) Sap-deficient bone marrow cells were infected in vitro with retroviruses encoding GFP (MIGR; n = 8) or GFP plus wild-type human SAP (SAP; n = 8) and were transferred into irradiated wild-type recipients. Percentages of GFP⁺ HSA^low Tet⁺ TCRβ⁺ NKT cells in the organs of reconstituted animals are indicated.

Figure 4

XLP patients have reduced NKT cells and a female XLP carrier shows skewed X chromosome inactivation in NKT cells. (a) Genotypes of the 17 XLP patients examined. (b) Representative flow cytometric plots showing percentages of CD3⁺ Vα24⁺ Vβ11⁺ or CD3⁺ 6B11⁺ NKT cells in one control and one XLP patient. (c) The absolute number of NKT cells in 10 controls and 17 XLP patients; Mann-Whitney test, P = 0.002. (d) X chromosome inactivation in T, B and NKT cells from a female control and an XLP carrier, as determined using the human androgen receptor assay. After PCR amplification of non-HpaII-digested DNA (–), products of two lengths were obtained from control and XLP carrier T, B and NKT cells, indicating polymorphic maternal and paternal androgen receptor alleles. After amplification of HpaII-digested DNA (+), two alleles were obtained from control lymphocytes and XLP carrier T and B cells, indicating random X inactivation within these lineages. Only one allele was amplified from HpaII-digested NKT cell DNA from the XLP carrier, indicating skewed X chromosome inactivation.