Research ArticleImmunology

Function of the Nucleotide Exchange Activity of Vav1 in T Cell Development and Activation

See allHide authors and affiliations

Science Signaling  15 Dec 2009:
Vol. 2, Issue 101, pp. ra83
DOI: 10.1126/scisignal.2000420

Figures

  • Fig. 1

    Generation of mice expressing GEF-inactive Vav1. (A) Vav1 GEF mutants were originally designed on the basis of the structure of the complex between the DH and PH domains of Tiam1 and Rac1 [Protein Data Bank (PDB) ID 1FOE] (25) and of the DH domain of Vav1 (PDB ID 1F5X) (26) and confirmed on the basis of the structure of the complex between the DH-PH-C1 region of Vav1 and Rac1 (PDB ID 2VRW) (3). The left panel shows a structural overlap of these complexes based on least-squares superposition of the respective DH domains with the PH and C1 domains of Vav1 and the PH domain of Tiam1 removed for clarity. The DH domain of Vav1 is shown in cyan and Rac1 is colored blue. The switch II region of Rac1 is highlighted in red, and the DH and Rac1 components of the Tiam1 structure are shown in gray. The right panel shows a magnified view of the DH-Rac1 interactions around switch II. The side chains of Leu334 and Lys335 of Vav1 and of Leu1194 and Lys1195 of Tiam1 are shown in stick representation. In both complexes, these residues make crucial nonpolar and hydrogen-bonding interactions with switch II of the small GTPase but appear to make no structurally important interactions with the remainder of the DH domain. (B) The graph on the left shows the CD spectra of wild-type (WT) or mutated (AA) recombinant DH domains of Vav1 at 25°C. The two curves are so similar that it is hard to see them apart, indicating that the WT and mutant domains have indistinguishable secondary structure. The graph on the right shows the molar CD at 220 nm of the DH domains as a function of temperature. The change in CD as the temperature is increased is a measure of the thermal stability of the domain. (C) The time-dependent change in normalized fluorescence of mant-GDP, which had been preloaded onto recombinant Rac1. Release of mant-GDP results in a decrease in fluorescence. Rac1 was either incubated alone or in the presence of the WT or the AA mutant DH domain of Vav1. (D) Diagram showing part of the WT Vav1 genomic locus before gene targeting (Vav1WT), the Vav1 gene after insertion of loxP-flanked neo gene into intron 9 and of the Leu334→Ala,Lys335→Ala (LK334AA) mutation into exon 10 (Vav1AAneo), and of the final targeted Vav1 allele after Cre-mediated removal of the neo gene, leaving behind a single loxP site in intron 9 and the LK334AA mutation in exon 10 (Vav1AA). Black boxes indicate Vav1 exons, black triangles indicate loxP sites, and arrows indicate the positions of oligonucleotides used to identify the mutant allele by PCR. Dotted lines indicate the extent of homology used in the targeting vector. Restriction sites: R, Eco RI; S, Spe I; X, Xho I. (E) Electrophoretogram showing the PCR products that were used to identify Vav1WT [WT, 496 base pairs (bp)] and Vav1AA (AA, 571 bp) alleles from tail DNA of mice of the indicated genotypes. The sizes of marker (M) bands are indicated. (F) Predicted DNA and protein sequence around the introduced AA mutation. Altered nucleotides and amino acids are indicated in red. In addition to the nucleotide changes required to mutate amino acids Leu334 and Lys335, three further silent nucleotide changes were made to introduce an Afl III site and to disrupt a stable hairpin in the oligonucleotide used for mutagenesis. The graph below shows the fluorescence trace and deduced sequence from around the site of mutation in DNA amplified by PCR from the tail of a Vav1AA/AA mouse. Nucleotides mutated from the WT are underlined in red.

  • Fig. 2

    Quantitation and phosphorylation of Vav1 in Vav1AA/AA T cells. (A) Western blots of total cytoplasmic extracts of naïve CD4+CD44lo lymph node T cells from mice of the indicated genotypes incubated with antibodies against Vav1, α-tubulin, LAT, and ERK2. The graph shows the mean ± SEM amount of Vav1 in naïve CD4+ lymph node T cells from mice of the indicated genotypes relative to those of α-tubulin, ERK2, and LAT normalized to that in wild-type (WT) T cells, which was set to 1. (B) Western blot analysis of immunoprecipated Vav1 from naïve CD4+ lymph node T cells from mice of the indicated genotypes stimulated through CD3ɛ and CD28 for the indicated times. Western blots were incubated with antibodies against phosphotyrosine (to detect pVav1) and Vav1. The graph shows the relative abundance of pVav1, determined from a ratio of the intensity of pVav1 to that of total Vav1.

  • Fig. 3

    The exchange activity of Vav1 is required for normal T cell development. (A) Contour plots showing separation of CD4CD8 double-negative (DN) thymocytes into DN1 (CD25CD44+), DN2/3 (CD25+CD44+/−), and DN4 (CD25CD44) subsets in mice of the indicated Vav1 genotypes. Numbers on these and subsequent plots indicate the percentages of cells that fell into the gates. Graphs show mean ± SEM numbers of total thymocytes and of cells in the DN subsets. Numbers here and in subsequent figures indicate millions of cells. (B) Contour plots showing CD4 and CD8 on thymocytes from mice of the indicated genotypes. Gates identify DN (CD4CD8), DP (CD4+CD8+), and CD4+ or CD8+ single-positive (CD4SP and CD8SP) thymocytes. For quantification of CD4SP and CD8SP cells, only TCRβhi cells were taken into account. Graph shows mean ± SEM numbers of cells in the indicated subsets. (C) Histograms showing CD5 on DP thymocytes. (D) Contour plots of CD4 and CD8 in the lymph nodes of mice of indicated genotypes. Gates identify CD4+ and CD8+ T cells. Graph shows mean ± SEM numbers of CD4+ and CD8+ T cells in the spleen and lymph nodes (LN). (E) Contour plots showing CD44 and CD25 on CD4+ T cells from the lymph nodes of mice of the indicated genotypes. Gates identify populations of naïve (CD25CD44lo), activated or memory (CD25CD44hi), and regulatory (CD25+CD44+) CD4+ T cells. Graph shows mean ± SEM numbers of the corresponding T cell populations. (F) Histograms showing CD44 on CD8+ T cells from the lymph nodes of mice of the indicated genotypes. Gates identify populations of naïve (CD44lo) and activated or memory (CD44hi) CD8+ T cells. Graph shows mean ± SEM numbers of the corresponding T cell populations. (G) Histograms showing TCRβ on naïve T cells from the lymph nodes of mice of the indicated genotypes. Colors used are as described for (C). (H) Graphs showing mean ± SEM numbers of CD4SP and CD8SP thymocytes and of CD4+ and CD8+ T cells in the lymph nodes of wild-type (WT) and Vav1+/− mice. n = 6 mice for all parts of this figure. *P < 0.05; **P < 0.01.

  • Fig. 4

    The exchange activity of Vav1 is required for positive and negative selection of thymocytes. (A) Contour plots of CD4 and CD8 on thymocytes from mice carrying the F5 TCR transgene that were also Rag1−/− and had the indicated Vav1 genotype. Numbers above the plots show mean ± SEM thymic cellularity (×106). Graphs show mean ± SEM numbers of CD8SP thymocytes and CD8+ lymph node (LN) T cells, and the ratio of CD8SP/DP thymocytes. (B) Contour plots of CD4 and CD8 on thymocytes from mice carrying the TCR7 TCR transgene. Graphs show CD4SP and CD8SP thymocyte numbers and the ratio of CD4SP/DP cells. (C) Contour plots showing CD4 and CD8 in the lymph nodes of mice described in (B). Gates identify CD4+ and CD8+ T cells. Graphs show mean ± SEM numbers of these T cells in the spleen and lymph nodes. (D) Graph shows the mean ± SEM percentages of CD4SP thymocytes expressing a TCR with the indicated Vβ chain in mice of the indicated genotypes. n ≥ 8 mice for all parts of this figure. *P < 0.05; **P < 0.01.

  • Fig. 5

    The absence of the exchange activity of Vav1 results in severely compromised activation and proliferation of T cells. Naïve CD4+CD44lo lymph node T cells from mice of the indicated genotypes were prelabeled with CFSE and stimulated for 72 hours with the indicated doses of plate-bound antibody against CD3ɛ in the presence or absence of soluble antibody against CD28 (1 μg/ml). Histograms show forward scatter (FSC) as a measure of cell size, staining of CD4 and CD25, and CFSE fluorescence. Graphs show the mean percentage of cells in triplicate samples that had high FSC, high abundance of CD4, or were positive for CD25. Percentage of cells in the initial population that had divided at least once by the end of culture period was determined by analysis of CFSE dilution as described in Materials and Methods. Comparison of wild-type (WT) and Vav1AA/AA T cells with (A and B) Vav1−/− or (C) Vav1+/− T cells. Experiments shown in (A) and (B) were carried out with plates with different binding capacities for antibodies and are therefore not directly comparable.

  • Fig. 6

    Expression of selected genes is severely perturbed in T cells with compromised Vav1 exchange activity. (A) Graphs show mean ± SEM amounts of IL-2 protein in culture supernatants of CD4+ T cells that were either unstimulated (−) or activated with the indicated doses of plate-bound antibody against CD3ɛ (n = 4 samples). (B) Mean ± SEM amounts of mRNA for Il2, TNFα, Il2rα (CD25), and NFAT2 in CD4+ T cells of the indicated genotypes that were either unstimulated (−) or activated (+) for 6 hours with plate-bound antibody against CD3ɛ (220 ng per well) (n = 5 samples). (C) Mean ± SEM abundance of IFNγ mRNA in CD4+ T cells of the indicated genotypes that were activated for the indicated times with plate-bound antibody against CD3ɛ (1 μg per well) (n = 3 samples). In all experiments, stimulation with antibody against CD3ɛ was carried out in the presence of soluble antibody against CD28 (1 μg/ml). *P < 0.05; **P < 0.01.

  • Fig. 7

    TCR-induced Ca2+ flux is independent of Vav1 GEF activity. Graphs show ratio of Indo1 violet/blue fluorescence as a measure of the concentration of intracellular Ca2+ [Ca2+]i as a function of time in naïve (CD44lo) CD4+ or CD8+ lymph node T cells from mice of the indicated genotypes. Arrows indicate the times at which cells were stimulated with antibody against CD3ɛ. (A) Comparison of wild-type (WT) and Vav1−/− T cells with Vav1AA/AA cells or (B) Vav1+/− T cells.

  • Fig. 8

    The GEF activity of Vav1 is required for a subset of TCR-induced signaling pathways. (A and B) Western blots of total cytoplasmic extracts of naïve CD4+CD44lo lymph node T cells from mice of the indicated genotypes that were stimulated through CD3ɛ and CD28 for the indicated times. Western blots were incubated with antibodies against phosphorylated forms of ERK2, PKD1, Akt, and LAT and with antibodies against total ERK2, PKD1, Akt, and LAT. The graphs on the right show the relative abundance of phosphorylated forms of ERK2, PKD1, Akt, and LAT, as determined from a ratio of the intensity of the bands of phosphorylated proteins to those of total proteins. Comparison of Vav1AA/AA and Vav1−/− T cells with (A) wild-type (WT) or (B) Vav1+/− T cells. (C) Western blots of active Rac1-GTP pulled down with a GST-Pak1-RBD fusion protein and of total cytoplasmic lysates from CD4+ T cells stimulated through CD3ɛ and CD28 for the indicated times. Western blots were incubated with an antibody against Rac1. Graph shows mean ± SEM ratios of Rac1-GTP to total Rac1 of three independent experiments, normalized as described in Materials and Methods. (D) Western blots and graphs [presented as described for (A)] of WT naïve CD4+ lymph node T cells treated with either 20 or 40 μM Rac inhibitor EHT 1864 or the inactive control compound EHT 4063.

  • Fig. 9

    Formation of APC–T cell conjugates and polymerization of actin require the GEF activity of Vav1. (A) The graph in the left panel shows the mean ± SEM percentages of F5 CD8+ T cells found in conjugates with APCs as a function of the concentration of the NP68 peptide. The graph in the right panel shows the mean ± SEM percentages of F5 CD8+ T cells in conjugates with APC cells loaded with a nonspecific peptide (gag) at the highest dose tested for NP68 (5 nM). (B) Graph shows the mean ± SEM percentages of CD4+ T cells of the indicated genotypes that had adhered to plates coated with ICAM-1 after stimulation through CD3ɛ (n = 6 samples). Adhesion was normalized by setting the adhesion of cells seen in the absence of antibody against CD3ɛ to 0% and that in response to Mn2+ to 100%. (C) Confocal single optical sections of preactivated and rested F5 CD8+ T cells conjugated for 7 min with NP68 peptide–pulsed and CMTMR-stained APCs. Cells were stained for F-actin phalloidin; nuclei were visualized with Hoechst. Scale bar, 5 μm. (D) Graph of the mean ± SEM amounts of F-actin (arbitrary units) in F5 CD8+ T cells quantified from images such as those shown in (C) (n ≥ 32 conjugates analyzed). (E) Mean ± SEM polarization of F-actin and LFA-1 within F5 CD8+ T cells toward APCs quantified as described in Materials and Methods (n ≥ 27 cells analyzed). (F) Confocal single optical sections of DP thymocytes expressing the F5 TCR, conjugated for 20 min with NP68 peptide–pulsed and CMTMR-stained APCs. Cells were immunostained for α-tubulin and CD4 (to identify DP cells). Nuclei were stained with Hoechst. Arrows indicate position of the MTOC. Scale bar, 5 μm. (G) Schematic representation of the analysis of MTOC movement. (H) Graph shows distribution of the MTOC within DP thymocytes conjugated as described in (F) and classified as described in (G) according to whether the MTOC was located proximal or distal to the APC, or was in the middle of the thymocyte. Statistical differences in MTOC distribution between the proximal and middle and distal positions were determined by Fisher’s exact test (n ≥ 34 conjugates analyzed). *P < 0.05; **P < 0.01.

Stay Connected to Science Signaling

Navigate This Article