Genetic analysis of SH2D4A, a novel adapter protein related to T cell-specific adapter and adapter protein in lymphocytes of unknown function, reveals a redundant function in T cells

Abstract

T cell-specific adapter (TSAd) protein and adapter protein in lymphocytes of unknown function (ALX) are two related Src homology 2 (SH2) domain-containing signaling adapter molecules that have both been shown to regulate TCR signal transduction in T cells. TSAd is required for normal TCR-induced synthesis of IL-2 and other cytokines in T cells and acts at least in part by promoting activation of the LCK protein tyrosine kinase at the outset of the TCR signaling cascade. By contrast, ALX functions as a negative-regulator of TCR-induced IL-2 synthesis through as yet undetermined mechanisms. In this study, we report a novel T cell-expressed adapter protein named SH2D4A that contains an SH2 domain that is highly homologous to the TSAd protein and ALX SH2 domains and that shares other structural features with these adapters. To examine the function of SH2D4A in T cells we produced SH2D4A-deficient mice by homologous recombination in embryonic stem cells. T cell development, homeostasis, proliferation, and function were all found to be normal in these mice. Furthermore, knockdown of SH2D4A expression in human T cells did not impact upon their function. We conclude that in contrast to TSAd and ALX proteins, SH2D4A is dispensable for TCR signal transduction in T cells.

FIGURE 1

The SH2D4A adapter protein. A, Protein sequence alignment of human and murine SH2D4A. B, Protein sequence alignment of the SH2 domains of murine SH2D4A, TSAd, and ALX. C, Domain organization of murine SH2D4A, TSAd and ALX proteins. In A and B, residues that are identical or homologous between sequences are boxed. In A and C, the locations of SH2 domains, proline-rich regions (PRR) and conserved tyrosine residues in consensus phosphorylation motifs are indicated.

FIGURE 2

SH2D4A expression. A, Anti-SH2D4A Western blot of human PBMC stimulated with CD3 and CD28 mAb for the indicated times in hours (0 represents 16 h mock-stimulated cells). NS is a non-specific reactive band that serves as an equal loading control (recombinant HA-tagged human SH2D4A was run in the leftmost track). B, Expression of SH2D4A in murine thymocytes, purified splenic CD4+ and CD8+ T cells, purified splenic B cells and bone marrow-derived macrophages (Mac) and DC was determined by Western blotting using an anti-SH2D4A antiserum. Cells were stimulated with CD3/CD28 mAb (T cells) or LPS (all other cells) or were mock-stimulated for different times as indicated. Blots were reprobed with a GAPDH antibody (below). C, Intracellular localization of SH2D4A. 293T cells or human peripheral blood T cells (PBT) were transfected with SH2D4A-GFP or GFP control. Intracellular localization of transfected proteins was determined by fluorescent microscopy. The position of nuclei was determined by Hoechst staining.

FIGURE 3

Generation of sh2d4a gene-targeted mice. A, Shown is the organization of the endogenous sh2d4a locus, targeting vector and targeted sh2d4a allele before and after Cre and Flp-mediated recombination. Locations of HindIII restriction sites and 5’ and 3’ probes used in Southern blotting experiments and positions of PCR primers used in genotyping experiments are indicated. B, Shown is a Southern blot of HindIII digested genomic DNA from two ES cell clones electroporated with the sh2d4a targeting vector. The blot was probed with 5’ and 3’ probes simultaneously. Positions of bands from wild type and targeted sh2d4a alleles are shown. Clone B is correctly targeted. C, Mice that carried a germline targeted sh2d4a allele were crossed with actin-Flp transgenic mice to delete the NeoR cassette. Progeny were subsequently crossed with CMV-Cre transgenic mice to delete exon 8 of the sh2d4a gene in all tissues. Heterozygote NeoR-deleted sh2d4a exon 8-deleted mice were then intercrossed to generate wild type (+/+), heterozygote (+/−) and homozygote (−/−) NeoR-deleted sh2d4a exon 8-deleted progeny. Progeny were genotyped by PCR of tail DNA using the indicated primer combinations to detect wild-type and NeoR-deleted sh2d4a exon 8-deleted alleles. D, Expression of SH2D4A protein in whole splenocytes of littermate wild-type, heterozygote and homozygote NeoR-deleted sh2d4a exon 8-deleted mice was determined by Western blotting using an anti-SH2D4A antiserum. NS, non-specific. Note the absence of the 48 kDa SH2D4A band in the homozygous mutant animals. The membrane was re-probed with a GAPDH antibody to demonstrate equivalent protein loading.

FIGURE 4

Flow cytometric analysis of SH2D4A-deficient mice. Depicted are flow cytometric dot plots of thymocytes and splenocytes from littermate wild-type and SH2D4A-deficient mice showing expression of the indicated markers on live populations. DN, CD4-CD8- double-negative thymocytes. Percentages of cells in quadrants are indicated.

FIGURE 5

T cell function in SH2D4A-deficient mice. A, Splenocytes from littermate wild-type, heterozygote or homozygote SH2D4A-deficient mice were stimulated with the indicated concentrations of a CD3 mAb and 0.5µg/ml of a CD28 mAb. Concentrations of IL-2 (at 24 h), IFN-γ and IL-4 (at 48 h) in culture supernatants were determined by ELISA. Data are represented as means +/− 1 SD of triplicate determinations. B, Splenocytes from littermate mice of the indicated genotypes were stimulated with CD3 and CD28 mAb for 48 h and then grown in IL-2 for a further 48 h. Cytokine synthesis in response to restimulation with CD3 and CD28 mAb was then determined as in A. C, Purified splenic T cells from littermate wild type and SH2D4A-deficient mice were stained with CFSE and stimulated with plate-bound CD3 mAb and 0.5µg/ml of soluble CD28 mAb. After 72 h, CFSE fluorescence was analyzed by flow cytometry. Indicated is the percentage of cells at successive cell division numbers, D0 through D4. D, Purified splenic T cells from littermate wild type and SH2D4A-deficient mice were stimulated with CD3 and CD28 mAb plus IL-2 for 48 h and then cultured in serum-containing (filled histogram) or serum-free (bold histogram) medium for a further 24 h. The percentage of Annexin-V-positive cells in cultures was then determined by flow cytometry. Shown are results from two different wild-type and SH2D4A-deficient mice.

FIGURE 6

Listeria infection of SH2D4A-deficient mice. Wild type or SH2D4A-deficient mice were challenged i.p. with 5 × 10⁵ CFU of Listeria monocytogenes. At the indicated times post-infection, the number of CFU in spleen and liver was determined. Points represent data from individual mice. Means are indicated with a bar. Differences in CFU between mice within an organ are not statistically significant at either time point.

FIGURE 7

Effect of siRNA knockdown of SH2D4A expression in human T cells. A, Human PBMC were transfected with SH2D4A siRNA or mock transfected by nucleofection. Cells were then stimulated with CD3 and CD28 mAb for 24 h or 48 h and lysed. SH2D4A expression was determined by Western blotting using an anti-SH2D4A antiserum. The blot was reprobed with an ERK antibody to demonstrate equal loading. B, SH2D4A siRNA-transfected and mock-transfected PBMC were stimulated or not with CD3 and CD28 mAb for 24 or 48 h. Concentrations of IL-2 and IFN-γ in culture supernatants were determined by ELISA. Data are represented as means +/− 1 SE of triplicate determinations. C, SH2D4A siRNA-transfected and mock-transfected PBMC were labeled with CFSE and stimulated or not with CD3 and CD28 mAb. After 96 h, the extent of cell division was determined by flow cytometry. For each cell type and condition of stimulation is shown is the percentage of cells that have undergone the indicated number of divisions. D, SH2D4A siRNA-transfected and mock-transfected PBMC were cultured in IL-2 for 48 h, washed and stimulated with CD3 and CD28 mAb for the indicated times. Protein tyrosine phosphorylation was then determined by Western blotting of whole cell lysates. Phosphotyrosine bands corresponding to the expected positions of the LAT phosphoprotein, Src-family and ZAP-70 PTK are indicated. Blots were reprobed with a GAPDH antibody to demonstrate equal protein loading.