Recruitment of CD2 to the immunological synapse in response to antigen is dependent on its proline-rich cytoplasmic tail. A peptide from this region (CD2:322–339) isolated CMS (human CD2AP); a related protein, CIN85; and the actin capping protein, CAPZ from a T cell line. In BIAcore™ analyses, the N-terminal SH3 domains of CMS and CIN85 bound CD2:322–339 with similar dissociation constants (KD = ∼100 μm). CAPZ bound the C-terminal half of CMS and CIN85. Direct binding between CMS/CIN85 and CAPZ provides a link with the actin cytoskeleton. Overexpression of a fragment from the C-terminal half or the N-terminal SH3 domain of CD2AP in a mouse T cell hybridoma resulted in enhanced interleukin-2 production and reduced T cell receptor down-modulation in response to antigen. These adaptor proteins are important in T cell signaling consistent with a role for CD2 in regulating pathways initiated by CMS/CIN85 and CAPZ. Recruitment of CD2 to the immunological synapse in response to antigen is dependent on its proline-rich cytoplasmic tail. A peptide from this region (CD2:322–339) isolated CMS (human CD2AP); a related protein, CIN85; and the actin capping protein, CAPZ from a T cell line. In BIAcore™ analyses, the N-terminal SH3 domains of CMS and CIN85 bound CD2:322–339 with similar dissociation constants (KD = ∼100 μm). CAPZ bound the C-terminal half of CMS and CIN85. Direct binding between CMS/CIN85 and CAPZ provides a link with the actin cytoskeleton. Overexpression of a fragment from the C-terminal half or the N-terminal SH3 domain of CD2AP in a mouse T cell hybridoma resulted in enhanced interleukin-2 production and reduced T cell receptor down-modulation in response to antigen. These adaptor proteins are important in T cell signaling consistent with a role for CD2 in regulating pathways initiated by CMS/CIN85 and CAPZ. CD2 is a T and NK cell surface protein that mediates low affinity cell-cell interactions by binding to related immunoglobulin superfamily (IgSF) proteins, CD58 in humans and CD48 in rodents (reviewed in Ref. 1Davis S.J. Ikemizu S. Wild M.K. van der Merwe P.A. Immunol. Rev. 1998; 163: 217-236Crossref PubMed Scopus (117) Google Scholar). In antigen-specific T cell activation, the interaction between CD2 and its ligand contributes toward lowering the threshold of activation by TCR in antigen-specific responses (2Bachmann M.F. Barner M. Kopf M. J. Exp. Med. 1999; 190: 1383-1392Crossref PubMed Scopus (104) Google Scholar). Adhesion between the extracellular regions of CD2 and CD48 assists in the initial segregation of proteins in antigen-specific T cell activation (1Davis S.J. Ikemizu S. Wild M.K. van der Merwe P.A. Immunol. Rev. 1998; 163: 217-236Crossref PubMed Scopus (117) Google Scholar). However, recruitment of CD2 into the central contact zone during formation of an immunological synapse is dependent on the cytoplasmic tail of CD2 (3Dustin M.L. Olszowy M.W. Holdorf A.D. Li J. Bromley S. Desai N. Widder P. Rosenberger F. van der Merwe P.A. Allen P.M. Shaw A.S. Cell. 1998; 94: 667-677Abstract Full Text Full Text PDF PubMed Scopus (582) Google Scholar). Among the proteins to which CD2 is related in its extracellular region (4Wang N. Morra M. Wu C. Gullo C. Howie D. Coyle T. Engel P. Terhorst C. Immunogenetics. 2001; 53: 382-394Crossref PubMed Scopus (51) Google Scholar), CD2 has a unique mechanism of engagement of intracellular machinery based on proline-rich motifs. The membrane-distal region of the cytoplasmic tail contains the sequence HQQKGPPLPRPRVQPKPP, which is conserved in CD2 from many species. Deletion in this region impairs signaling events induced by CD2 mAbs 1The abbreviations used are: mAb, monoclonal antibody; CIN85-SH3, CIN85 SH3 domains 1–3; CD2BP1-SH3, CD2BP1 SH3 domain; -SH3d1, SH3 domain 1; -C, C-terminal region; -CC, coiled-coil region; pMAL, maltose binding protein; RU, response units; IL, interleukin; PBS, phosphate-buffered saline; MOPS, 4-morpholinepropanesulfonic acid; MALDI, matrix-assisted laser desorption ionization; EGFP, enhanced green fluorescence protein; TCR, T cell receptor. 1The abbreviations used are: mAb, monoclonal antibody; CIN85-SH3, CIN85 SH3 domains 1–3; CD2BP1-SH3, CD2BP1 SH3 domain; -SH3d1, SH3 domain 1; -C, C-terminal region; -CC, coiled-coil region; pMAL, maltose binding protein; RU, response units; IL, interleukin; PBS, phosphate-buffered saline; MOPS, 4-morpholinepropanesulfonic acid; MALDI, matrix-assisted laser desorption ionization; EGFP, enhanced green fluorescence protein; TCR, T cell receptor. (5King P.D. Sadra A. Teng J.M. Bell G.M. Dupont B. Int. Immunol. 1998; 10: 1009-1016Crossref PubMed Scopus (13) Google Scholar). CD2AP isolated in a yeast two-hybrid screen with the cytoplasmic tail of CD2 contains three SH3 domains (3Dustin M.L. Olszowy M.W. Holdorf A.D. Li J. Bromley S. Desai N. Widder P. Rosenberger F. van der Merwe P.A. Allen P.M. Shaw A.S. Cell. 1998; 94: 667-677Abstract Full Text Full Text PDF PubMed Scopus (582) Google Scholar). The N-terminal SH3 domain appeared to have a relatively high specificity and avidity for this conserved region of the CD2 cytoplasmic region (3Dustin M.L. Olszowy M.W. Holdorf A.D. Li J. Bromley S. Desai N. Widder P. Rosenberger F. van der Merwe P.A. Allen P.M. Shaw A.S. Cell. 1998; 94: 667-677Abstract Full Text Full Text PDF PubMed Scopus (582) Google Scholar). In cells, a truncated protein containing only the two most N-terminal SH3 domains of CD2AP disrupted T cell polarization, implicating CD2 in cytoskeletal rearrangement (3Dustin M.L. Olszowy M.W. Holdorf A.D. Li J. Bromley S. Desai N. Widder P. Rosenberger F. van der Merwe P.A. Allen P.M. Shaw A.S. Cell. 1998; 94: 667-677Abstract Full Text Full Text PDF PubMed Scopus (582) Google Scholar). Antigen-specific engagement was required for proper cluster formation by CD2, leading to speculation as to whether CD2 regulates the cytoskeleton or vice versa (3Dustin M.L. Olszowy M.W. Holdorf A.D. Li J. Bromley S. Desai N. Widder P. Rosenberger F. van der Merwe P.A. Allen P.M. Shaw A.S. Cell. 1998; 94: 667-677Abstract Full Text Full Text PDF PubMed Scopus (582) Google Scholar). There are data showing that other SH3 domains have the potential to bind the distal conserved region of CD2 (3Dustin M.L. Olszowy M.W. Holdorf A.D. Li J. Bromley S. Desai N. Widder P. Rosenberger F. van der Merwe P.A. Allen P.M. Shaw A.S. Cell. 1998; 94: 667-677Abstract Full Text Full Text PDF PubMed Scopus (582) Google Scholar, 6Bell G.M. Fargnoli J. Bolen J.B. Kish L. Imboden J.B. J. Exp. Med. 1996; 183: 169-178Crossref PubMed Scopus (80) Google Scholar, 7Li J. Nishizawa K. An W. Hussey R.E. Lialios F.E. Salgia R. Sunder-Plassmann R. Reinherz E.L. EMBO J. 1998; 17: 7320-7336Crossref PubMed Scopus (94) Google Scholar). CD2BP2, a protein containing a novel GYF domain, is a candidate for interacting with a more membrane-proximal conserved motif PPPPGH, which is repeated exactly in human (8Freund C. Dotsch V. Nishizawa K. Reinherz E.L. Wagner G. Nat. Struct. Biol. 1999; 6: 656-660Crossref PubMed Scopus (78) Google Scholar). A molecular interaction has not yet been defined to explain the role of the distal end of the CD2 tail in enhancing the avidity for its cell surface ligand (9Hahn W.C. Rosenstein Y. Calvo V. Burakoff S.J. Bierer B.E. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 7179-7183Crossref PubMed Scopus (49) Google Scholar). We identified proteins that interact with the CD2 cytoplasmic region and show that there is a direct link through CMS and CIN85 with the actin capping protein, CAPZ. We show these proteins have a role in regulating signaling in T cells. Peptides—The peptides were synthesized with an N-terminal biotin (Table I). CD2 peptide sequences are numbered according to Swissprot: P06729. CD2:322–339 *This work was supported by the Medical Research Council. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. was used in peptide pull-down experiments. Position 14 differs from Swissprot:P06729. No difference in specificity of binding between CD2:322–339 *This work was supported by the Medical Research Council. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. (not shown) and CD2:322–339 (see Figs. 3 and 4) was observed in BIAcore™ experiments.Table IPeptidesPeptide namePeptide sequenceManufacturerCD2:322-339*Biotin-QKGPPLPRPRVQPAPPHG-COOHSigma-GenosysCD2:322-339Biotin-QKGPPLPRPRVQPKPPHG-COOHInteractivaCD2:281-305Biotin-HPPPPPGHRSQAPSHRPPPPGHRVQ-COOHInteractivaICOS-PBiotin-DPNGEYMFMRAVNTAKKSRLTDVTL-COOHInteractiva Open table in a new tab Fig. 4CAPZ does not bind SH3 domains but does bind the C-terminal region of CMS. A, CIN85-SH3 (0.5 μm) did not bind immobilized cCAPZ (solid line; 2507 RU) or control streptavidin (dashed line; 1423 RU) but does bind CD2:322–339 (not shown). The inset shows CIN85-SH3 (2 μm) binds to immobilized CD2:322–339 (solid line; 711 RU) not CD2:281–305 (dashed line; 868 RU) or ICOS-P (dotted line; 699 RU). B, CMS-C (1.3 μm) bound to immobilized cCAPZ (1742 RU) but not to CD2:322–339 (150 RU) or to a blank flow cell. C and D, CMS-C (C) and CIN85-C (D) (1.3 μm) bound immobilized hCAPZ (4265 RU) not CD2:322–339 (640 RU), whereas (C, inset) CMS-CC (1.1 μm) did not bind immobilized hCAPZ (solid line; 2425 RU) or CD2:322–339 (dashed line; 640 RU). E and F, hCAPZ (0.42 μm) bound immobilized CMS-C (E, solid line; 1330 RU) and CIN85-C (F, solid line; 924 RU) but not CD2:322–339 (E and F, dashed lines; 664 RU).View Large Image Figure ViewerDownload (PPT) Pull-down Experiments—The human T cell line Jurkat cells were washed twice in PBS and lysed at 108/ml in 10 mm Tris-HCl, pH 7.4, 140 mm NaCl, 1 mm EDTA, 10% glycerol, and 1% Brij-96 detergent. Protease inhibitors, 1 mm phenylmethylsulfonyl fluoride, 20 μg/ml leupeptin, 1 μg/ml aprotinin, 1 μg/ml pepstatin, 50 mm benzamidine, 1 mm sodium vanadate, 1 mm NaF were added immediately prior to use. The lysates were rotated at 4 °C for 60 min and then clarified at 10,000 × g (13,000 rpm) in a Microfuge or 2,000 × g (3,000 rpm) in a Beckman benchtop GPR centrifuge at 4 °C for 10 min and then filtered (0.45 μm). Streptavidin-coated Dynabeads (Dynal A.S., Oslo, Norway) were washed once in PBS before being saturated with biotinylated peptides. At least 20-fold excess peptide (16 μg) over the bead (25 μl at 10 mg/ml) capacity was used. After 30 min at 4 °C beads were washed twice in PBS. 25 μl of beads and 2.5 × 108 cell equivalents of lysate were used for each lane of the gel. The control beads were saturated with d-Biotin in at least 100-fold excess. The beads were rotated for 2 h at 4 °C in the lysate, transferred to new tubes, and washed three times with 200 μl of PBS in a Dynal MPC-E magnetic particle concentrator with a change of tube during each wash. Washed beads were resuspended in 10 μl of NuPAGE SDS sample buffer (Novex, San Diego, CA), reduced with 1% (w/v) dithiothreitol, vortexed, and incubated at 70 °C for 10 min. The beads were pelleted in a microcentrifuge, and supernatant was run on NuPAGE 4–12% Bis-Tris gels (Novex) in an X-Cell II gel tank (Novex) using MOPS SDS running buffer (Novex). The gels were stained with Coomassie, and the protein bands were excised from the gel and analyzed by trypsin digestion and mass spectrometry (10Shevchenko A. Wilm M. Vorm O. Mann M. Anal. Chem. 1996; 68: 850-858Crossref PubMed Scopus (7736) Google Scholar). After overnight trypsin digestion, the peptides were extracted twice with 30 μl of 50% acetonitrile, 5% trifluoroacetic acid. Finally, the dried peptide solution was resuspended in 10 μl of H2O. MALDI Mass Spectrometry—0.5 μl of saturated 2,5-dihydroxybenzoic acid in H2O was mixed on-target with 0.5 μl of sample and dried. Peptide mass fingerprints were acquired on a REFLEX III mass spectrometer (Bruker Daltonics, Coventry, UK). The spectra were internally or externally calibrated (Calibration Mixture 2 from a Sequazyme kit, Applied Biosystems, Warrington, UK). Data interpretation was carried out using the Prospector Suite of search programs (prospector.ucsf.edu/). Peak lists from MALDI spectra were input into MS-Fit, and the NCBI nonredundant protein data base was searched. Constructs—Maltose-binding protein fusion proteins were constructed using the pMAL-c2X vector (www.neb.com). Templates CMS (11Kirsch K.H. Georgescu M.M. Ishimaru S. Hanafusa H. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 6211-6216Crossref PubMed Scopus (131) Google Scholar) and CIN85 (12Take H. Watanabe S. Takeda K. Yu Z.X. Iwata N. Kajigaya S. Biochem. Biophys. Res. Commun. 2000; 268: 321-328Crossref PubMed Scopus (136) Google Scholar) for amplification by PCR using cloned Pfu polymerase were provided by K. Kirsch and S. Kaijigaya, respectively. The fragments were cloned in-frame into vector cut with BamHI and SalI. SH3 domains with 5′ BamHI and 3′ SalI sites lacked the initiator Met and contained a stop codon after the following C-terminal sequences CIN85-SH3, FVKLLPP; CMS-SH3d1, KEIKRE; and CIN85-SH3d1, REIKKE. C-terminal regions were cloned using an engineered 5′ BamHI site for CIN85-C, a natural BglII site for CMS-C, and a XhoI site after the stop codon producing the following N-terminal sequences after the junction with pMAL: CMS-CC, IVEALK; CMS-C, SGTVYP; and CIN85-C, GALPPR. CD2BP1-SH3 was amplified from human PBL cDNA and fused to pMAL, the sequence codons for SPAQEY following the BamHI site and with a SalI site after the stop codon. In the T cell hybridoma, PLEGFP-C1 (www.clontech.com) was used for expression of EGFP fused to CD2AP-C, a HindIII (1078 bp)-XhoI (2426 bp) restriction fragment of CD2AP (M. Dustin and A. Shaw, St. Louis, MO), CD2AP-SH3d1-CC (CMS and CD2AP SH3d1 are identical, CD2AP terminology is used in the mouse experiments) constructed in two steps using a 3′ primer containing adjacent BglII and HindIII sites. The intermediate vector was cut with BglII and SalI, and the CMS-CC fragment was inserted, which produced the joining sequence ERSIVE. CMS-CC was also inserted into pLEGFP-C1 to provide a dimeric EGFP control. A mouse CD2 pBabe construct was made by Anna Cambiaggi. Proteins—pMAL fusion proteins were expressed and affinity-purified using standard procedures with amylose resin (www.neb.com). SH3 constructs were further purified by gel filtration using Superdex 75 (Amersham Biosciences). Chicken CAPZ α1β1, which was the first CAPZ available to us was expressed using a construct provided by T. Obinata (13Soeno Y. Abe H. Kimura S. Maruyama K. Obinata T. J. Muscle Res. Cell Motil. 1998; 19: 639-646Crossref PubMed Scopus (53) Google Scholar) and purified from bacterial lysates containing 0.15 m NaCl by ion exchange on Q Sepharose using a binding buffer containing 10 mm Tris-HCl, pH 8.0, 0.05 m NaCl and eluted with the same buffer and increasing NaCl concentration, with CAPZ eluting at about 0.4 m NaCl. A single peak containing the heterodimeric CAPZ was isolated by gel filtration on Superdex 200. Human erythrocyte CAPZ was provided by P. Khulman (14Kuhlman P.A. Fowler V.M. Biochemistry. 1997; 36: 13461-13472Crossref PubMed Scopus (48) Google Scholar). Attempts to produce full-length CMS and CIN85 in bacteria as pMAL fusion proteins and in insect cells with a His-biotin tag were hampered by low level expression and degradation and not pursued. Confirmatory sequencing of all constructs was obtained using Big-Dye™ (ABI) Sanger dideoxynucleotide method on a Prism 377 DNA or 3100 Genetic Analyzer sequencer (ABI). Purified proteins were analyzed by SDS-PAGE and Coomassie staining. The concentrations were estimated by absorption at 280 nm in a 1-cm flow cell using theoretical extinction coefficients for the pMAL fusion proteins: CMS-SH3d1 = 74,250 m–1 cm–1 and CIN85-SH3d1 = 77,380 m–1 cm–1. Experimentally determined extinction coefficients determined in triplicate for CMS-SH3d1 (76,973, 79,501, and 88,107 m–1 cm–1) are similar to the theoretical value. BIAcore Analysis—The experiments were carried out using a BIAcore 2000 at 25 and at 37 °C (15van der Merwe P.A. Bodian D.L. Daenke S. Linsley P. Davis S.J. J. Exp. Med. 1997; 185: 393-403Crossref PubMed Scopus (404) Google Scholar, 16Brown M.H. Boles K. van der Merwe P.A. Kumar V. Mathew P.A. Barclay A.N. J. Exp. Med. 1998; 188: 2083-2090Crossref PubMed Scopus (347) Google Scholar) for affinity measurements with SH3d1 proteins using buffers, anti-mouse Fcγ, and amine coupling to research grade CM5 chips (BIAcore AB) using flow rates of 10 μl/min and for equilibrium binding 20 μl/min. Chicken CAPZ mAbs 5B12.3 and 1E5.25.4 were obtained from Development Studies Hybridoma Bank (www.uiowa.edu/~dshbwww). Biotinylated peptide solutions were injected in HEPES-buffered saline, pH 7.4, at 400 ng/ml over immobilized streptavidin (3000–4000 RU). CMS, CIN85, and cCAPZ were directly immobilized by amine coupling in 10 mm NaAc buffer, pH 5.0. The response units immobilized are in the relevant figure legends. An extra washing step after blocking with ethanolamine was avoided to avoid possible inactivation of immobilized material. For affinity measurements, 5-μl injections of increasing and decreasing concentrations of monomeric SH3d1 proteins were passed over immobilized peptides at 20 μl/min at 37 °C, and the data were analyzed as described (15van der Merwe P.A. Bodian D.L. Daenke S. Linsley P. Davis S.J. J. Exp. Med. 1997; 185: 393-403Crossref PubMed Scopus (404) Google Scholar, 16Brown M.H. Boles K. van der Merwe P.A. Kumar V. Mathew P.A. Barclay A.N. J. Exp. Med. 1998; 188: 2083-2090Crossref PubMed Scopus (347) Google Scholar). Antigen-specific T Cell Hybridoma assays—2B4 mouse T cell hybridoma and Chinese hamster ovary cells expressing mouse I-Ek were used as model T and antigen presenting cells, respectively, as described (17Wild M.K. Cambiaggi A. Brown M.H. Davies E.A. Ohno H. Saito T. van der Merwe P.A. J. Exp. Med. 1999; 190: 31-41Crossref PubMed Scopus (99) Google Scholar). CD2+ 2B4 T cell hybridoma cells were made by Anna Cambiaggi (Oxford, UK) using calcium phosphate transfection of the BOSC 293T packaging cell line, retroviral transduction, and selection using puromycin at 1 μg/ml. Several clones were pooled to produce a polyclonal population. PLEGFP-C1 constructs were introduced into the CD2+ and CD2– T cell hybridomas by retroviral transduction, with recombinant virus being produced by transfection using FuGENE (Roche Applied Science) into the Phoenix packaging cell line. Antigen-specific IL-2 production and TCR down-modulation experiments were essentially as previously described (17Wild M.K. Cambiaggi A. Brown M.H. Davies E.A. Ohno H. Saito T. van der Merwe P.A. J. Exp. Med. 1999; 190: 31-41Crossref PubMed Scopus (99) Google Scholar) with the following differences. For both antigen stimulation and IL-2 production, round bottomed plates were used in the initial experiments (i.e. Fig. 6), and subsequently, flat bottomed plates were used. To measure IL-2 production, 8,000 HT2 cells (50 μl) were incubated with supernatants (50 μl: usually 1:1 and 1:5) for 16 h followed by a 6-h pulse with 0.8–1 μCi of [3H]thymidine. IL-2 production of triplicate or quadruplicate cultures was calculated using a standard of recombinant human IL-2 and curve fitting in PRISM. Down-modulation experiments were initially carried out in plates, with triplicate cultures being pooled into one fluorescence-activated cell sorter tube (Fig. 7B), but subsequently the whole assay was performed in fluorescence-activated cell sorter tubes (Fig. 7C). Optimal staining was obtained using a three-step staining procedure with KT3 (anti-mCD3), biotinylated anti-rat Ig (Jackson Labs), and Streptavidin-Quantum Red (Sigma). The stained cells were fixed with 2% paraformaldehyde and read the following day. The median fluorescence was determined from cells gated for the same levels of EGFP.Fig. 7Interactions of the N-terminal SH3 domain of CD2AP are important in regulating T cell signaling, and the mechanism involves TCR down-modulation. A, IL-2 production in response to antigen (30 nm moth cytochrome c) by cells expressing CD2AP-C, CD2AP-SH3D1-CC, and CMS-CC. B and C, TCR down-modulation by cells expressing CD2AP-C, CD2AP-SH3D1-CC, and CMS-CC in response to antigen (1 μm moth cytochrome c). Two (of >5) representative experiments are shown.View Large Image Figure ViewerDownload (PPT) CMS, CIN85, and CAPZ Were Isolated from Jurkat Cell Lysates by CD2 Peptide 322–339 —To identify proteins interacting with the cytoplasmic tail of CD2, synthetic peptides were coupled to magnetic beads and used as an affinity matrix in "pull-down" experiments to purify associated proteins from extracts of the Jurkat T cell line. Four bands seen in analysis by SDS-PAGE and Coomassie Brilliant Blue staining (Fig. 1) with peptide CD2:322–339 but not with other peptides including CD2:281–305 (data not shown) were identified (Table II). Band 2 represents CMS, the human homologue of CD2AP (11Kirsch K.H. Georgescu M.M. Ishimaru S. Hanafusa H. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 6211-6216Crossref PubMed Scopus (131) Google Scholar). Band 1 is a closely related protein CIN85 (12Take H. Watanabe S. Takeda K. Yu Z.X. Iwata N. Kajigaya S. Biochem. Biophys. Res. Commun. 2000; 268: 321-328Crossref PubMed Scopus (136) Google Scholar), which has the same overall structure as CMS. Bands 3 and 4 contained the α and β subunits of the actin capping protein, CAPZ. CAPZα exists in two closely related isoforms (α1 and α2) that differ at the C termini (18Hart M.C. Korshunova Y.O. Cooper J.A. Cell Motil. Cytoskelet. 1997; 38: 120-132Crossref PubMed Scopus (55) Google Scholar). Band 3 contained both CapZ α1 and α2 isoforms, because peptides unique to α1 and α2 were present (Table II). Peptide data from band 4 matched the human CAPZβ2 subunit (19Barron-Casella E.A. Torres M.A. Scherer S.W. Heng H.H. Tsui L.C. Casella J.F. J. Biol. Chem. 1995; 270: 21472-21479Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar) (Table II), which is the dominant isoform in nonmuscle cells (20Schafer D.A. Korshunova Y.O. Schroer T.A. Cooper J.A. J. Cell Biol. 1994; 127: 453-465Crossref PubMed Scopus (122) Google Scholar).Table IIIdentities of bands 1–4 in Fig. 1BandProteinGenBank™ accession numberTheoretical molecular massCoveragekDa%1CIN85 and glycinin7188749 and 571219973 and 6147 and 372CMS75122771583CAPZα1 and CAPZα25453597 and 545359933 and 3372 and 464CAPZβ248266593159 Open table in a new tab CMS and CIN85 but Not CAPZ Bind Directly to CD2:322–339 —Direct interaction between the mouse homologue of CMS, CD2AP, and the cytoplasmic region of CD2 contained in CD2: 322–339 has been previously demonstrated (3Dustin M.L. Olszowy M.W. Holdorf A.D. Li J. Bromley S. Desai N. Widder P. Rosenberger F. van der Merwe P.A. Allen P.M. Shaw A.S. Cell. 1998; 94: 667-677Abstract Full Text Full Text PDF PubMed Scopus (582) Google Scholar). The N-terminal SH3 domain of CD2AP mediated binding. Both this SH3 domain and the CD2:322–339 peptide have identical sequences in mice and humans. To confirm that CMS was binding directly to CD2:322–339 via its N-terminal SH3 domain and to ask whether CIN85 had comparable specificity, we tested binding of recombinant forms of CMS-SH3d1 and CIN85-SH3d1 (Fig. 2) to peptides using a BIAcore™. The N-terminal SH3 domains of CMS and CIN85 are the most similar, being 68% identical. Care was taken to conduct experiments with monomeric material within hours of elution from a gel filtration column. CMS-SH3d1 was injected over immobilized peptide, and the data are shown for five concentrations in Fig. 3A. CD2:281–305, which did not bind pMAL fusion proteins containing all three SH3 domains of CMS (not shown) or CIN85 (Figs. 2 and 4A, inset), was used as a negative control peptide. CMS-SH3d1 bound to CD2:322–339 and not to CD2:281–305 (Fig. 3A). CIN85-SH3d1 bound in a similar fashion (Fig. 3B) The signals for CMS-SH3d1 and CIN85-SH3d1 passing over the control CD2:281–305 are due simply to the high concentration of protein necessary for detection of weak interactions. Binding was characteristic of a low affinity monomeric interaction with rapid kinetics enabling a quantitative comparison between the two proteins to be made by calculating the KD from equilibrium binding values over a range of concentrations of soluble CMS or CIN85 SH3d1 (Fig. 3, C and D). Fitting binding data to a nonlinear hyperbole and Scatchard analysis (Fig. 3, C and D, insets) showed that CMS-SH3d1 and CIN85-SH3d1 bound CD2:322–339 according to a Langmuir model and with similar dissociation constants, 105 and 76 μm, i.e. in the order of 100 μm. Duplicate sets of data gave comparable results: CMS, 76 μm and CIN85, 72 μm. Slightly lower levels of immobilized peptide resulted in KD values of 129 and 127 μm for 127 and 76 and 77 μm for CIN85. We did not detect in our pull-down experiments another SH3 domain containing protein, CD2BP1, previously reported to interact with the region of CD2 covered by CD2:322–339 in either the absence or the presence of presence of 200 μm Zn2+ ions (7Li J. Nishizawa K. An W. Hussey R.E. Lialios F.E. Salgia R. Sunder-Plassmann R. Reinherz E.L. EMBO J. 1998; 17: 7320-7336Crossref PubMed Scopus (94) Google Scholar). This may have been due to the sensitivity of the assay, and this putative interaction was tested using purified monomeric pMAL fusion protein containing the SH3 domain of CD2BP1, CD2BP1-SH3 (Fig. 2B, lane 4). An interaction was detected at 37 °C, but levels of binding were low despite the high levels of peptide immobilized, and the interaction showed less specificity than CMS or CIN85 because CD2BP1 bound to both CD2:322–339 and CD2:281–309 peptides (Fig. 3E). In experiments using immobilized peptide levels that gave clear CMS-SH3 binding at 4 μm, no binding was detectable with CD2BP1-SH3 at 40 μm even in the presence of 2 mm MgCl2 (data not shown). CAPZ does not contain domains that are known to bind proline-rich sequences (21Kay B.K. Williamson M.P. Sudol M. FASEB J. 2000; 14: 231-241Crossref PubMed Scopus (1031) Google Scholar) and was not predicted to bind directly to the CD2 cytoplasmic tail. To distinguish between direct and indirect interactions, recombinant chicken CAPZ α1β1, which was available to us (cCAPZ; Fig. 2B, lane 9), CAPZ being highly conserved among species, was tested for interaction with CD2 peptides. Chicken CAPZ (46 μm) (Fig. 2) did not bind CD2 peptides (Fig. 3F) but was antigenically active because it bound mAbs specific for either chicken CAPZ α1 and β1 (Fig. 3F, inset). CAPZ Binds the C-terminal Region of CMS and CIN85 but Not the Coiled-coil Region of CMS—The lack of binding of CD2 peptides by cCAPZ was consistent with CAPZ being associated with the CD2 peptide via CMS and/or CIN85. The lack of proline-rich sequence in CAPZ suggested the mechanism of interaction between CAPZ and CIN85 or CMS would not involve the SH3 domains. This prediction was correct because CIN85-SH3d1–3 did not bind to immobilized cCAPZ (Fig. 4A), whereas CIN85-SH3d1–3 bound CD2:322–339 peptide (inset). A C-terminal fragment of CMS (CMS-C; Fig. 2, A and B, lane 6) was tested for interaction with immobilized cCAPZ. It bound avidly (Fig. 4B) to cCAPZ but not to CD2:322–339. This result with cCAPZ α1β1 heterodimer shows that the interaction is not CAPZ β isoform-specific because the human CAPZ (hCAPZ) identified in the pull-down experiments contained the α1β2 heterodimer. The avid binding is consistent with CMS-C being dimeric as shown by gel filtration (data not shown and Ref. 11Kirsch K.H. Georgescu M.M. Ishimaru S. Hanafusa H. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 6211-6216Crossref PubMed Scopus (131) Google Scholar). CMS-C also bound directly to human CAPZ α1β2 heterodimer (Figs. 4C and 2B, lane 8) and not to CD2:322–339. A motif (VEALK) within the coil-coil region of CMS known to selfassociate (11Kirsch K.H. Georgescu M.M. Ishimaru S. Hanafusa H. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 6211-6216Crossref PubMed Scopus (131) Google Scholar) is also present in the C terminus of human CAPZ β2, raising the possibility that this region mediated binding to CAPZ. However, the coiled-coil region of CMS, CMS-CC (Fig. 2, A and B, lane 5), which formed a tetramer as described for CIN85 (22Watanabe S. Take H. Takeda K. Yu Z.X. Iwata N. Kajigaya S. Biochem. Biophys. Res. Commun. 2000; 278: 167-174Crossref PubMed Scopus (69) Google Scholar) demonstrating a capability to form higher order complexes, did not bind to CAPZ (Fig. 4C, inset). CIN85-C (Fig. 2, A and B, lane 7) was also dimeric (data not shown) and bound immobilized hCAPZ with high avidity (Fig. 4D). In the reverse orientation, hCAPZ bound to immobilized CMS and CIN85, showing that binding was not an artifact of CAPZ immobilization (Fig. 4, E and F). Thus CAPZ interacts through a discrete fragment of the CMS/CIN85 C-terminal region, which is distinct from the upstream putative endophilin-binding sites (23Petrelli A. Gilestro G.F. Lanzardo S. Comoglio P.M. Migone N. Giordano S. Nature. 2002; 416: 187-190Crossref PubMed Scopus (371) Google Scholar) and three of the four putative actin binding sites in the coiled-coil region at the C terminus of CMS (11Kirsch K.H. Georgescu M.M. Ishimaru S. Hanafusa H. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 6211-6216Crossref PubMed Scopus (131) Google Scholar) (Fig. 5). Overexpression of the C-terminal Region of CD2AP Enhances Anti
In 40 years the analysis of the lymphocyte cell surface has been transformed with major leaps following technological advances. However, considerable progress was made in the early 1970s by the group of Alan Williams (Figure 12.1A) who joined the MRC Immunochemistry Unit (ICU) in 1970 and turned to ...
Journal Article Sequence of a rat MHC class II-associated invariant chain cDNA clone containing a 64 amino acid thyroglobulin-.like domain Get access Andrew J. McKnight, Andrew J. McKnight MRC Cellular Immunology Unit, Sir William Dunn School of Pathology, University of OxfordOxford OX1 3RE, UK Search for other works by this author on: Oxford Academic PubMed Google Scholar Don W. Mason, Don W. Mason MRC Cellular Immunology Unit, Sir William Dunn School of Pathology, University of OxfordOxford OX1 3RE, UK Search for other works by this author on: Oxford Academic PubMed Google Scholar A.Neil Barclay A.Neil Barclay MRC Cellular Immunology Unit, Sir William Dunn School of Pathology, University of OxfordOxford OX1 3RE, UK Search for other works by this author on: Oxford Academic PubMed Google Scholar Nucleic Acids Research, Volume 17, Issue 10, 25 May 1989, Pages 3983–3984, https://doi.org/10.1093/nar/17.10.3983 Published: 25 May 1989 Article history Received: 11 April 1989 Published: 25 May 1989
'%3e%3cpath fill='%23012169' d='M0 0v30h60V0z'/%3e%3cpath stroke='%23fff' stroke-width='6' d='m0 0 60 30m0-30L0 30'/%3e%3cpath stroke='%23C8102E' stroke-width='4' d='m0 0 60 30m0-30L0 30' clip-path='url(%23b)'/%3e%3cpath stroke='%23fff' stroke-width='10' d='M30 0v30M0 15h60'/%3e%3cpath stroke='%23C8102E' stroke-width='6' d='M30 0v30M0 15h60'/%3e%3c/g%3e%3c/svg%3e)
