Phosphoprotein associated with glycosphingolipid-enriched microdomains (PAG), a novel ubiquitously expressed transmembrane adaptor protein, binds the protein tyrosine kinase csk and is involved in regulation of T cell activation

Abstract

According to a recently proposed hypothesis, initiation of signal transduction via immunoreceptors depends on interactions of the engaged immunoreceptor with glycosphingolipid-enriched membrane microdomains (GEMs). In this study, we describe a novel GEM-associated transmembrane adaptor protein, termed phosphoprotein associated with GEMs (PAG). PAG comprises a short extracellular domain of 16 amino acids and a 397-amino acid cytoplasmic tail containing ten tyrosine residues that are likely phosphorylated by Src family kinases. In lymphoid cell lines and in resting peripheral blood alpha/beta T cells, PAG is expressed as a constitutively tyrosine-phosphorylated protein and binds the major negative regulator of Src kinases, the tyrosine kinase Csk. After activation of peripheral blood alpha/beta T cells, PAG becomes rapidly dephosphorylated and dissociates from Csk. Expression of PAG in COS cells results in recruitment of endogenous Csk, altered Src kinase activity, and impaired phosphorylation of Src-specific substrates. Moreover, overexpression of PAG in Jurkat cells downregulates T cell receptor-mediated activation of the transcription factor nuclear factor of activated T cells. These findings collectively suggest that in the absence of external stimuli, the PAG-Csk complex transmits negative regulatory signals and thus may help to keep resting T cells in a quiescent state.

Figure 1

pp80 accumulates in GEMs and associates with Fyn in peripheral blood T cells. (A) GEMs immunoprecipitated from NP-40 lysates of Raji and Jurkat cells by anti-CD59 or control (Ctr) mAb were subjected to in vitro kinase assay followed by SDS-PAGE and autoradiography. (B) A Fyn immunoprecipitate obtained from peripheral blood T cells was subjected to in vitro phosphorylation and two-dimensional electrophoresis; the previously identified and so far unidentified components (p43 and p85) are indicated. (C) Jurkat cells were solubilized in a 1% NP-40 lysis buffer, the suspension was subjected to sucrose gradient ultracentrifugation, and the fractions (numbered 1–8 from top to bottom) were analyzed by SDS-PAGE followed by anti–P-Tyr Western blotting.

Figure 1

pp80 accumulates in GEMs and associates with Fyn in peripheral blood T cells. (A) GEMs immunoprecipitated from NP-40 lysates of Raji and Jurkat cells by anti-CD59 or control (Ctr) mAb were subjected to in vitro kinase assay followed by SDS-PAGE and autoradiography. (B) A Fyn immunoprecipitate obtained from peripheral blood T cells was subjected to in vitro phosphorylation and two-dimensional electrophoresis; the previously identified and so far unidentified components (p43 and p85) are indicated. (C) Jurkat cells were solubilized in a 1% NP-40 lysis buffer, the suspension was subjected to sucrose gradient ultracentrifugation, and the fractions (numbered 1–8 from top to bottom) were analyzed by SDS-PAGE followed by anti–P-Tyr Western blotting.

Figure 1

pp80 accumulates in GEMs and associates with Fyn in peripheral blood T cells. (A) GEMs immunoprecipitated from NP-40 lysates of Raji and Jurkat cells by anti-CD59 or control (Ctr) mAb were subjected to in vitro kinase assay followed by SDS-PAGE and autoradiography. (B) A Fyn immunoprecipitate obtained from peripheral blood T cells was subjected to in vitro phosphorylation and two-dimensional electrophoresis; the previously identified and so far unidentified components (p43 and p85) are indicated. (C) Jurkat cells were solubilized in a 1% NP-40 lysis buffer, the suspension was subjected to sucrose gradient ultracentrifugation, and the fractions (numbered 1–8 from top to bottom) were analyzed by SDS-PAGE followed by anti–P-Tyr Western blotting.

Figure 2

Purification and cDNA cloning of PAG. (A) GEMs prepared from 7 × 10⁸ Raji cells were separated by two-dimensional electrophoresis followed by silver staining of the gel. The arrow indicates the position of pp80, which was verified by anti–P-Tyr Western blotting (not shown). The silver-stained spot was excised and subjected to Nano ES MS/MS. (B) Sequencing of the p85 protein by Nano ES MS/MS. Top: a part of the spectrum of in-gel tryptic digest of the p85 protein. The peptide ions designated with * are autolysis products of trypsin. Ions of tryptic peptides originating from the p85 protein are designated with T. These ions were in turn isolated by the quadrupole mass filter of a triple quadrupole tandem mass spectrometer and fragmented in the collision cell, and their tandem mass spectra were acquired. The tandem mass spectrum acquired from the doubly charged peptide precursor ion having m/z 673.0 is shown (bottom). A short stretch of the peptide sequence was determined by considering precise mass differences between the adjacent fragment ions containing the COOH terminus (y-ions; reference 62). Note that the NH₂-terminal aa of the peptide is at the right-hand side. The peptide sequence tag (reference 63) was assembled using the deduced peptide sequence, masses of the correspondent fragment ions, and the mass of the intact peptide, and was used for searching comprehensive protein and EST databases. Searching against a protein sequence database produced no hit. Searching against an EST database hit the peptide sequence LLDENENLQEK that is present in the clone AA622656 and in several homologous clones. The match was verified by comparing the masses of fragment ions calculated for the retrieved peptide sequence and the masses of peptide ions observed in the tandem mass spectrum. The COOH-terminal part of the peptide sequence is covered by continuous series of y-ions (in bold), and the NH₂-terminal part of the sequence is covered by series of b-ions (reference 62). Thus database searching provided confident identification, although only a single peptide matched the sequence of the EST clone. Similarly, tandem mass spectrum acquired from the precursor ion with m/z 416.2 hit the peptide sequence FSSLSYK present in the EST clone AA220201 (data not shown). After a full-length sequence of the protein was obtained from the cDNA nucleotide sequence, more tandem mass spectra were retrospectively matched to correspondent tryptic peptides: T₁, IPPESAVDTMLTAR; T₄, ELPGPQTEGK. A stretch of the peptide sequence similar to the sequence of T₁ (…PPESAVD…) was deduced from tandem mass spectra acquired from peptide ions T₂ and T₃. (C) aa sequence of human (upper) and mouse (lower) PAG. The putative transmembrane region is boxed. All tyrosines are in bold, and the putative tyrosine-based signaling motifs are overlined in bold. The peptides sequenced by Nano ES MS/MS are indicated by a thin line over the corresponding sequences. These sequence data are available from EMBL/GenBank/DDBJ under accession nos. AF240634 (human) and A5250192 (mouse). (D) Jurkat cells were transfected with cDNA encoding PAG, or were mock transfected (−). The cell lysates were analyzed by SDS PAGE and anti-PAG Western blotting. The blot was restained by antiserum to β subunits of heterotrimeric G proteins (loading control), and both films were overlaid.

Figure 2

Purification and cDNA cloning of PAG. (A) GEMs prepared from 7 × 10⁸ Raji cells were separated by two-dimensional electrophoresis followed by silver staining of the gel. The arrow indicates the position of pp80, which was verified by anti–P-Tyr Western blotting (not shown). The silver-stained spot was excised and subjected to Nano ES MS/MS. (B) Sequencing of the p85 protein by Nano ES MS/MS. Top: a part of the spectrum of in-gel tryptic digest of the p85 protein. The peptide ions designated with * are autolysis products of trypsin. Ions of tryptic peptides originating from the p85 protein are designated with T. These ions were in turn isolated by the quadrupole mass filter of a triple quadrupole tandem mass spectrometer and fragmented in the collision cell, and their tandem mass spectra were acquired. The tandem mass spectrum acquired from the doubly charged peptide precursor ion having m/z 673.0 is shown (bottom). A short stretch of the peptide sequence was determined by considering precise mass differences between the adjacent fragment ions containing the COOH terminus (y-ions; reference 62). Note that the NH₂-terminal aa of the peptide is at the right-hand side. The peptide sequence tag (reference 63) was assembled using the deduced peptide sequence, masses of the correspondent fragment ions, and the mass of the intact peptide, and was used for searching comprehensive protein and EST databases. Searching against a protein sequence database produced no hit. Searching against an EST database hit the peptide sequence LLDENENLQEK that is present in the clone AA622656 and in several homologous clones. The match was verified by comparing the masses of fragment ions calculated for the retrieved peptide sequence and the masses of peptide ions observed in the tandem mass spectrum. The COOH-terminal part of the peptide sequence is covered by continuous series of y-ions (in bold), and the NH₂-terminal part of the sequence is covered by series of b-ions (reference 62). Thus database searching provided confident identification, although only a single peptide matched the sequence of the EST clone. Similarly, tandem mass spectrum acquired from the precursor ion with m/z 416.2 hit the peptide sequence FSSLSYK present in the EST clone AA220201 (data not shown). After a full-length sequence of the protein was obtained from the cDNA nucleotide sequence, more tandem mass spectra were retrospectively matched to correspondent tryptic peptides: T₁, IPPESAVDTMLTAR; T₄, ELPGPQTEGK. A stretch of the peptide sequence similar to the sequence of T₁ (…PPESAVD…) was deduced from tandem mass spectra acquired from peptide ions T₂ and T₃. (C) aa sequence of human (upper) and mouse (lower) PAG. The putative transmembrane region is boxed. All tyrosines are in bold, and the putative tyrosine-based signaling motifs are overlined in bold. The peptides sequenced by Nano ES MS/MS are indicated by a thin line over the corresponding sequences. These sequence data are available from EMBL/GenBank/DDBJ under accession nos. AF240634 (human) and A5250192 (mouse). (D) Jurkat cells were transfected with cDNA encoding PAG, or were mock transfected (−). The cell lysates were analyzed by SDS PAGE and anti-PAG Western blotting. The blot was restained by antiserum to β subunits of heterotrimeric G proteins (loading control), and both films were overlaid.

Figure 3

PAG is identical to pp80. (A) GEM-associated phosphoproteins were obtained from Raji cells by CD59 immunoprecipitation followed by in vitro kinase assay (IVK). Phosphorylated proteins were then subjected to reprecipitation using anti-PAG or control (Ctr) mAbs, and were analyzed by SDS-PAGE followed by autoradiography. (B) Phosphoproteins obtained by immunoprecipitation and in vitro kinase assay of a Fyn immunoprecipitate prepared from peripheral blood T cells (left) were reprecipitated using a polyclonal mouse antiserum to PAG (center) or a polyclonal rabbit antiserum to SKAP55 (right), and were analyzed by two-dimensional electrophoresis followed by autoradiography. (C) Laurylmaltoside lysates of peripheral blood α/β T cells were subjected to preclearing by anti-PAG or control (Ctr) immunosorbents, followed by SDS-PAGE and analysis by anti–P-Tyr (top) or anti-PAG (bottom) Western blotting. L, untreated lysate.

Figure 3

PAG is identical to pp80. (A) GEM-associated phosphoproteins were obtained from Raji cells by CD59 immunoprecipitation followed by in vitro kinase assay (IVK). Phosphorylated proteins were then subjected to reprecipitation using anti-PAG or control (Ctr) mAbs, and were analyzed by SDS-PAGE followed by autoradiography. (B) Phosphoproteins obtained by immunoprecipitation and in vitro kinase assay of a Fyn immunoprecipitate prepared from peripheral blood T cells (left) were reprecipitated using a polyclonal mouse antiserum to PAG (center) or a polyclonal rabbit antiserum to SKAP55 (right), and were analyzed by two-dimensional electrophoresis followed by autoradiography. (C) Laurylmaltoside lysates of peripheral blood α/β T cells were subjected to preclearing by anti-PAG or control (Ctr) immunosorbents, followed by SDS-PAGE and analysis by anti–P-Tyr (top) or anti-PAG (bottom) Western blotting. L, untreated lysate.

Figure 3

PAG is identical to pp80. (A) GEM-associated phosphoproteins were obtained from Raji cells by CD59 immunoprecipitation followed by in vitro kinase assay (IVK). Phosphorylated proteins were then subjected to reprecipitation using anti-PAG or control (Ctr) mAbs, and were analyzed by SDS-PAGE followed by autoradiography. (B) Phosphoproteins obtained by immunoprecipitation and in vitro kinase assay of a Fyn immunoprecipitate prepared from peripheral blood T cells (left) were reprecipitated using a polyclonal mouse antiserum to PAG (center) or a polyclonal rabbit antiserum to SKAP55 (right), and were analyzed by two-dimensional electrophoresis followed by autoradiography. (C) Laurylmaltoside lysates of peripheral blood α/β T cells were subjected to preclearing by anti-PAG or control (Ctr) immunosorbents, followed by SDS-PAGE and analysis by anti–P-Tyr (top) or anti-PAG (bottom) Western blotting. L, untreated lysate.

Figure 4

Posttranslational modifications of PAG. (A) Lysates of pervanadate-treated (+) or untreated (−) Jurkat cells were analyzed by anti-PAG Western blotting. (B) COS cells were transiently transfected with the depicted cDNA constructs. Lysates corresponding to 10% of the transfectants were analyzed by P-Tyr Western blotting (top left). The blot was stripped, and expression of the individual constructs was assessed using FLAG (FLAG-PAG and FLAG-Syk), Lck, Fyn, and MYC (MYC-ZAP70) Abs, respectively (right). The remaining 90% of the lysates were subjected to anti-FLAG immunoprecipitation and analyzed by P-Tyr Western blotting (bottom left). Arrowheads indicate the positions of phosphorylated PAG. (C) Anti-PAG (left) and anti–P-Tyr (right) Western blots of total cell lysates of wild-type Jurkat cells, the Lck-negative mutant J.CaM1.6, and the ZAP70/Syk-negative mutant P116. (D) Lysates of Jurkat cells treated for 1 min with 10 μM inhibitor of Src family PTKs PP1 (left) or untreated controls (Ctr; right) were analyzed by anti–P-Tyr blot. The blots were stripped and reincubated with mAb to PAG (bottom).

Figure 5

Tissue distribution of PAG. (A) Multiple-tissue Northern blot analysis. Positions of size markers (kb) are shown on the left. (B) Peripheral blood leukocytes were permeabilized, stained by mAbs to PAG and irrelevant negative control (Ctr) followed by fluorescein-labeled goat anti–mouse IgG, and analyzed by cytofluorometry. The results are shown as contour plots (x-axis: intensity of fluorescence in logarithmic scale; y-axis: side scatter). Different leukocyte populations are marked: N, neutrophils; M, monocytes; L, lymphocytes.

Figure 5

Tissue distribution of PAG. (A) Multiple-tissue Northern blot analysis. Positions of size markers (kb) are shown on the left. (B) Peripheral blood leukocytes were permeabilized, stained by mAbs to PAG and irrelevant negative control (Ctr) followed by fluorescein-labeled goat anti–mouse IgG, and analyzed by cytofluorometry. The results are shown as contour plots (x-axis: intensity of fluorescence in logarithmic scale; y-axis: side scatter). Different leukocyte populations are marked: N, neutrophils; M, monocytes; L, lymphocytes.

Figure 6

Subcellular localization of PAG. (A) Raji cells were permeabilized and immunostained using PAG mAb MEM-250 (left) followed by secondary Cy2-labeled anti–mouse IgG (green fluorescence). Nuclei are visualized by propidium iodide (red fluorescence). In the negative control (right), PAG mAb was omitted. (B) Lysates of Jurkat cells solubilized in 1% Brij58 or laurylmaltoside (LM) were subjected to sucrose gradient ultracentrifugation, and the fractions (numbered 1–8 from top to bottom) were analyzed by anti-PAG or anti-LAT Western blotting as indicated. LAT was used as a well-established control marker of GEMs. S, sediment. (C) Raji cells were biosynthetically labeled with [³H]palmitate solubilized in laurylamltoside-containing buffer, and PAG or control (Ctr) immunoprecipitates were analyzed by SDS-PAGE followed by fluorography.

Figure 7

Fyn and and Csk associate with PAG. (A) In vitro pull-down assay. Lysates of untreated (–) or pervanadate (PV)-treated (+) Jurkat cells were subjected to precipitation using the depicted recombinant GST–SH2 domain fusion proteins. Subsequently, anti-PAG or anti-TRIM Western blotting was performed. (B) Resting peripheral blood T cells (3 × 10⁷) were left untreated or were treated with the Src family kinase inhibitor PP1 (10 μM) for 2 min. Subsequently, PAG immunoprecipitates obtained from laurylmaltoside lysates were subjected to sequential anti–P-Tyr, anti-Fyn, and anti-Csk immunoblotting. (C) Untreated Raji, Jurkat, or peripheral blood α/β T cells were solubilized by 1% laurylmaltoside followed by PAG or control immunoprecipitation and Csk Western blotting. L, original lysates. (D) The blots shown in Fig. 4 B (corresponding to both total lysates [bottom] and anti-FLAG immunoprecipitates [middle]) were stripped and incubated with a polyclonal antiserum directed against Csk. (E) COS cells (expressing endogenous Csk) were transfected with cDNA constructs encoding tyrosine mutants of FLAG-PAG and the PTK Fyn. Anti-FLAG immunoprecipitates were analyzed by Western blotting for the presence of Csk and Fyn. The numbers at the top identify the Tyr residues mutated to Phe. The expression levels of Fyn and Csk in total lysates were determined in parallel (bottom strips). None of the investigated PAG mutants showed any gross alterations of the overall level of tyrosine phosphorylation (not shown).

Figure 7

Fyn and and Csk associate with PAG. (A) In vitro pull-down assay. Lysates of untreated (–) or pervanadate (PV)-treated (+) Jurkat cells were subjected to precipitation using the depicted recombinant GST–SH2 domain fusion proteins. Subsequently, anti-PAG or anti-TRIM Western blotting was performed. (B) Resting peripheral blood T cells (3 × 10⁷) were left untreated or were treated with the Src family kinase inhibitor PP1 (10 μM) for 2 min. Subsequently, PAG immunoprecipitates obtained from laurylmaltoside lysates were subjected to sequential anti–P-Tyr, anti-Fyn, and anti-Csk immunoblotting. (C) Untreated Raji, Jurkat, or peripheral blood α/β T cells were solubilized by 1% laurylmaltoside followed by PAG or control immunoprecipitation and Csk Western blotting. L, original lysates. (D) The blots shown in Fig. 4 B (corresponding to both total lysates [bottom] and anti-FLAG immunoprecipitates [middle]) were stripped and incubated with a polyclonal antiserum directed against Csk. (E) COS cells (expressing endogenous Csk) were transfected with cDNA constructs encoding tyrosine mutants of FLAG-PAG and the PTK Fyn. Anti-FLAG immunoprecipitates were analyzed by Western blotting for the presence of Csk and Fyn. The numbers at the top identify the Tyr residues mutated to Phe. The expression levels of Fyn and Csk in total lysates were determined in parallel (bottom strips). None of the investigated PAG mutants showed any gross alterations of the overall level of tyrosine phosphorylation (not shown).

Figure 8

PAG is involved in regulation of T cell responses. (A) COS cells were transfected with the depicted cDNAs. 40 h later, anti-CD25 (Tac) or anti-FLAG (FLAG-PAG) immunoprecipitates (IP) were prepared and subjected to anti–P-Tyr, anti-ζ, or anti-Csk Western blotting (top). In parallel, total lysates were analyzed by anti-PAG and anti-Fyn Western blotting (bottom). (B) Right: COS cells were transfected with the indicated cDNA constructs followed by lysis in 1% laurylamaltoside, Fyn immunoprecipitation, in vitro kinase assay, SDS-PAGE, and autoradiography. The expression levels of Fyn in total lysates were determined in parallel. Left: corresponding densitometric analysis. (C) Jurkat T cells were transfected with cDNA constructs coding for Fyn, PAG, and FLAG–SLP-76, respectively, and reporter constructs coding for a triplicated NF-AT binding site of the human IL-2 promotor and for Renilla luciferase. 16 h after transfection, 7 × 10⁴ cells per well were stimulated for additional 6 h using anti-TCR mAb C305, a combination of C305 and CD28 mAb, or PMA (10⁻⁹ M) and ionomycin (1 μg/ml). Subsequently, cells were lysed and luciferase activity was determined. Values represent means of quadruplicates. Right: anti-PAG, FLAG (SLP-76), Fyn, and mitogen-activated protein kinase (MAPK) Western blotting analysis of the transfectants. Note that overexpression of PAG alone also inhibited TCR- and TCR plus CD28–mediated induction of NF-AT activity, although to a lesser extent (∼40%; not shown). This might be due to incomplete tyrosine phosphorylation of PAG in the cells not overexpressing Fyn.

Figure 8

PAG is involved in regulation of T cell responses. (A) COS cells were transfected with the depicted cDNAs. 40 h later, anti-CD25 (Tac) or anti-FLAG (FLAG-PAG) immunoprecipitates (IP) were prepared and subjected to anti–P-Tyr, anti-ζ, or anti-Csk Western blotting (top). In parallel, total lysates were analyzed by anti-PAG and anti-Fyn Western blotting (bottom). (B) Right: COS cells were transfected with the indicated cDNA constructs followed by lysis in 1% laurylamaltoside, Fyn immunoprecipitation, in vitro kinase assay, SDS-PAGE, and autoradiography. The expression levels of Fyn in total lysates were determined in parallel. Left: corresponding densitometric analysis. (C) Jurkat T cells were transfected with cDNA constructs coding for Fyn, PAG, and FLAG–SLP-76, respectively, and reporter constructs coding for a triplicated NF-AT binding site of the human IL-2 promotor and for Renilla luciferase. 16 h after transfection, 7 × 10⁴ cells per well were stimulated for additional 6 h using anti-TCR mAb C305, a combination of C305 and CD28 mAb, or PMA (10⁻⁹ M) and ionomycin (1 μg/ml). Subsequently, cells were lysed and luciferase activity was determined. Values represent means of quadruplicates. Right: anti-PAG, FLAG (SLP-76), Fyn, and mitogen-activated protein kinase (MAPK) Western blotting analysis of the transfectants. Note that overexpression of PAG alone also inhibited TCR- and TCR plus CD28–mediated induction of NF-AT activity, although to a lesser extent (∼40%; not shown). This might be due to incomplete tyrosine phosphorylation of PAG in the cells not overexpressing Fyn.

Figure 8

PAG is involved in regulation of T cell responses. (A) COS cells were transfected with the depicted cDNAs. 40 h later, anti-CD25 (Tac) or anti-FLAG (FLAG-PAG) immunoprecipitates (IP) were prepared and subjected to anti–P-Tyr, anti-ζ, or anti-Csk Western blotting (top). In parallel, total lysates were analyzed by anti-PAG and anti-Fyn Western blotting (bottom). (B) Right: COS cells were transfected with the indicated cDNA constructs followed by lysis in 1% laurylamaltoside, Fyn immunoprecipitation, in vitro kinase assay, SDS-PAGE, and autoradiography. The expression levels of Fyn in total lysates were determined in parallel. Left: corresponding densitometric analysis. (C) Jurkat T cells were transfected with cDNA constructs coding for Fyn, PAG, and FLAG–SLP-76, respectively, and reporter constructs coding for a triplicated NF-AT binding site of the human IL-2 promotor and for Renilla luciferase. 16 h after transfection, 7 × 10⁴ cells per well were stimulated for additional 6 h using anti-TCR mAb C305, a combination of C305 and CD28 mAb, or PMA (10⁻⁹ M) and ionomycin (1 μg/ml). Subsequently, cells were lysed and luciferase activity was determined. Values represent means of quadruplicates. Right: anti-PAG, FLAG (SLP-76), Fyn, and mitogen-activated protein kinase (MAPK) Western blotting analysis of the transfectants. Note that overexpression of PAG alone also inhibited TCR- and TCR plus CD28–mediated induction of NF-AT activity, although to a lesser extent (∼40%; not shown). This might be due to incomplete tyrosine phosphorylation of PAG in the cells not overexpressing Fyn.

Figure 9

Decrease of PAG tyrosine phosphorylation and Csk association with PAG after stimulation of T cells. (A) Peripheral blood α/β T cells were stimulated for 1 min by an IgM anti-CD3 mAb (+) or were left untreated (−), and postnuclear lysates were analyzed by anti–P-Tyr Western blotting. (B) Lysates of α/β T cells stimulated (+) or unstimulated (−) with a mixture of IgM mAbs to CD3 and CD28 were subjected to anti-PAG or control immunoprecipitation followed by sequential anti–P-Tyr, Csk, PAG, and Fyn Western blotting. Left: quantitative evaluation of the data shown at right.