Coclustering CD45 with CD4 or CD8 alters the phosphorylation and kinase activity of p56lck.

Antibody-mediated CD4 crosslinking results in increased tyrosine phosphorylation and tyrosine kinase activity of the associated p56lck. Treatment with anti-CD4 and anti-Ig also induced the phosphorylation of p56lck in a CD45- mutant cell line, indicating that the increase in phosphotyrosine content of p56lck is not the result of being sequestered from CD45 protein tyrosine phosphatase (PTPase). Antibody-mediated coclustering of CD45 with CD4 inhibited the anti-CD4-induced phosphorylation of p56lck on tyrosine and the concomitant increase in in vitro kinase activity. Similar results were obtained when CD45 was coclustered with CD8 on cytotoxic T cell lines. These observations provide strong evidence that p56lck is a substrate for CD45 in vivo and provide an assay to study the regulation and specificity of CD45 PTPase activity.

Negative regulation of CD45 protein tyrosine phosphatase activity by ionomycin in T cells.

CD45 is a leukocyte-specific, transmembrane protein tyrosine phosphatase (PTPase) required for T cell responsiveness. How the activity of PTPases is regulated in vivo is unclear. Treatment of murine thymocytes and a variety of murine T cell lines with the calcium ionophore ionomycin decreased CD45 PTPase activity. Ionomycin treatment also led to a decreased phosphorylation of serine residues in CD45. These results indicate that increased intracellular calcium modulates CD45 PTPase activity, demonstrating regulation of CD45 PTPase activity in vivo, and also implicate serine dephosphorylation as a possible mechanism.

Impaired drug uptake in methotrexate resistant Crithidia fasciculata without changes in dihydrofolate reductase activity or gene amplification.

Crithidia fasciculata cells grown in defined medium are sensitive to methotrexate (MTX), an inhibitor of dihydrofolate reductase (DHFR). When cells are challenged with 2-5 microM MTX, cell division ceases after 3-4 divisions and the cells become rounded and immotile for approximately 60 h, with a 40% decrease in cell viability occurring during this period. The cells then recover normal morphology and cell division resumes. Cells which undergo this treatment can be transferred directly into high levels of the drug (1-2 mM). The resistance phenotype is stable in the absence of the drug. Resistance correlates with impaired uptake of [3H]MTX, which in wild-type cells is taken up by a carrier-mediated process. There is no indication of gene amplification at the DNA level or at the level of DHFR activity, as occurs in the case of MTX-resistant Leishmania major. Several lines of MTX-resistant L. major which show gene amplification also exhibit impaired uptake of [3H]MTX.

Evidence for multiple lytic pathways used by cytotoxic T lymphocytes.

Previous data generated by ourselves and others questioned the role of degranulation as a mechanism to explain CTL-mediated cytotoxicity. In this report we examine this tissue in greater depth. CTL-mediated lysis was probed with three different inhibitors. 4,4'-diisothiocyano-2,2'-disulfonic acid stilbene inhibits degranulation in a wide range of cell types, including CTL. EGTA, through chelation of Ca2+, also inhibits degranulation processes in CTL, and would inhibit other events or processes dependent on extracellular Ca2+. We also used prolonged exposure to PMA to exhaust PKC activity in CTL. Using these inhibitors, we have defined three pathways of lysis used by CTL. One pathway requires Ca2+, is PMA sensitive, but does not depend on degranulation. The second pathway is independent of Ca2+, is not PMA sensitive, and also does not depend on degranulation. All primary CTL and cloned CTL lyse most target cells via pathway I. However, when confronted with certain target cells (which we have referred to previously as Ca2+-independent target cells), pathway II is induced. When pathway II is induced, pathway I apparently shuts down. We show here that pathway II does not depend on protein synthesis, and that it also leads to DNA solubilization in target cells. A limited number of cloned CTL use pathways I and II as just described, but use in addition, and simultaneously, a third pathway that appears to involve degranulation. This pathway is seen irregularly in most CTL clones, and may be influenced by levels of IL-2 in the culture medium.

Identification of the CD45-associated 116-kDa and 80-kDa proteins as the alpha- and beta-subunits of alpha-glucosidase II.

CD45 is an abundant, highly glycosylated transmembrane protein-tyrosine phosphatase expressed on hematopoietic cells. Herein we demonstrate that two proteins of 116 kDa and 80 kDa copurify with CD45 from mouse T cells. Microsequence analysis of the 116-kDa protein revealed high similarity to an incomplete human open reading frame that has been suggested to correspond to the catalytic alpha-subunit of glucosidase II. We determined the nucleotide sequence of the mouse cDNA and observed that it encodes a protein product nearly identical to its human homologue and shares an active site consensus sequence with Family 31 glucosidases. Amino acid sequencing of the 80-kDa protein, followed by molecular cloning, revealed high homology to human and bovine cDNAs postulated to encode the beta-subunit of glucosidase II. Antisera developed to the mouse beta-subunit allowed us to demonstrate that the interaction between CD45 and glucosidase II can be reconstituted in vitro in an endoglycosidase H-sensitive manner. The strong interaction between glucosidase II and CD45 may provide a paradigm for investigating novel aspects of the biology of these proteins.

T cell receptor engagement induces tyrosine phosphorylation of FAK and Pyk2 and their association with Lck.

Stimulation through the TCR is known to induce tyrosine phosphorylation of a number of proteins, which leads to functional activation of T cells. Identification of the substrates that become phosphorylated and defining their interactions with other signaling molecules will provide insight into the mechanisms controlling T cell activation. Focal adhesion kinase (FAK) and the recently described Pyk2 kinase are homologous members of a non-receptor protein tyrosine kinase family. FAK has been shown to become phosphorylated upon TCR stimulation, but its role, if any, in T cell activation remains to be defined. Although Pyk2 has been shown to play a role in neuronal cell activation stimulated through G-protein-coupled receptors, a role in T cell activation has not been described. In this study we show that FAK and Pyk2 are two of the major 115-to-120-kDa proteins that become tyrosine phosphorylated in T cells following TCR complex stimulation. Furthermore, coincident with the increase in tyrosine phosphorylation, we show an association of these kinases with the SH2 domain of the tyrosine kinase Lck in vivo. The increase in tyrosine phosphorylation of both FAK and Pyk2, however, occurs in Lck-deficient cells suggesting that phosphorylation of both of these kinases does not require Lck. Taken together, these results suggest that FAK and Pyk2, perhaps in coordination with Lck, play a role in T cell activation.

Two distinct domains of the beta-subunit of glucosidase II interact with the catalytic alpha-subunit.

Recent purification and cDNA cloning of the endoplasmic reticulum processing enzyme glucosidase II have revealed that it is composed of two soluble proteins: a catalytic alpha-subunit and a beta-subunit of unknown function, both of which are highly conserved in mammals. Since the beta-subunit, which contains a C-terminal His-Asp-Glu-Leu (HDEL) motif, may function to link the catalytic subunit to the KDEL receptor as a retrieval mechanism, we sought to map the regions of the mouse beta-subunit protein responsible for mediating the association with the alpha-subunit. By screening a panel of recombinant beta-subunit glutathione S-transferase fusion proteins for the ability to precipitate glucosidase II activity, we have identified two non-overlapping interaction domains (ID1 and ID2) within the beta-subunit. ID1 encompasses 118 amino acids at the N-terminus of the mature polypeptide, spanning the cysteine-rich element in this region. ID2, located near the C-terminus, is contained within amino acids 273-400, a region occupied in part by a stretch of acidic residues. Variable usage of 7 alternatively spliced amino acids within ID2 was found not to influence the association of the two sub-units. We theorize that the catalytic subunit of glucosidase II binds synergistically to ID1 and ID2, explaining the high associative stability of the enzyme complex.

Characterization of intercellular adhesion molecule-1 (ICAM-1)-augmented degranulation by cytotoxic T cells. ICAM-1 and anti-CD3 must be co-localized for optimal adhesion and stimulation.

Purified intercellular adhesion molecule-1 (ICAM-1), a ligand of the LFA-1, was used to analyze the contribution of ICAM-1 to the activation of CTL. ICAM-1 facilitates degranulation when co-immobilized with substimulatory amounts of anti-CD3. This facilitated response is most likely mediated through LFA-1, since Abs to this molecule significantly inhibit the response, Interestingly, when ICAM-1 and anti-CD3 are immobilized on separate beads and presented to the CTL, no ICAM-1-enhanced degranulation is observed. The ICAM-1 and anti-CD3 must be immobilized on the same surface to augment the response, suggesting that ICAM-1 either does not transmit signals into the cell or it transmits a very localized signal, since the ICAM-1 and anti-CD3 must be juxtaposed. Consistent with this finding, we demonstrate that ICAM-1 does not induce tyrosine phosphorylation or a Ca(2+)-flux in the CTL clone, but does potentiate these responses when co-immobilized with substimulatory anti-CD3. Finally ICAM-1 and anti-CD3 must be immobilized on the same bead for stable adhesion of CTL to ICAM-1. When ICAM-1 and anti-CD3 are immobilized on separate beads, there is only a transient, low level of adhesion to the ICAM-1 beads. Taken together, these results suggest that LFA-1 is acting principally as an adhesion molecule, with respect to ICAM-1, in CTL and that this adhesion is regulated through the TCR complex.

Fibronectin induces phosphorylation of a 120-kDa protein and synergizes with the T cell receptor to activate cytotoxic T cell clones.

Fibronectin (FN) has been shown to act as a costimulator in both CD4+ and CD8+ T cell activation through the T cell receptor (TcR). Consistent with previous studies, we found that FN is able to both enhance the maximal amount of TcR-triggered degranulation and lower the threshold for activation. The density of immobilized anti-CD3 or anti-TcR required to induce degranulation and tyrosine phosphorylation of cellular proteins by several cytotoxic T lymphocyte clones is quantitatively about tenfold lower in the presence of FN. We further demonstrate that FN alone stimulates transient tyrosine phosphorylation of a 120-kDa protein (pp120) in CD8+ T cells and when FN is coimmobilized with substimulatory amounts of anti-CD3 or anti-TcR there is a synergistic response, resulting in prolonged and enhanced phosphorylation of pp120. To determine if FN acts as a costimulator in CD8+ cells solely through mediating adhesion events or if it also transduces signals in T cells we conducted remote stimulation experiments. Degranulation was induced when FN and sub-stimulatory anti-CD3 were presented on separate surfaces, indicating that FN induces independent transmembrane signals capable of augmenting TcR-induced signals resulting in a functional response. Both FN plus TcR-induced tyrosine phosphorylation of pp120 and degranulation are inhibited by RGD-containing peptides, implying that an RGD-dependent FN receptor is mediating phosphorylation of pp120 and enhancing TcR-mediated degranulation.

CD45 protein-tyrosine phosphatase is specifically associated with a 116-kDa tyrosine-phosphorylated glycoprotein.

CD45 is a protein-tyrosine phosphatase expressed on all cells of hematopoietic origin. In an attempt to further characterize CD45 function, we set out to identify molecule(s) that specifically associate with CD45. A 116-kDa protein was detected in immunoprecipitates from CD45+ cells but not CD45- cells. The association between CD45 and this 116-kDa protein can be reconstituted by mixing lysates from CD45- cell lines with purified CD45. p116 appears to associate with CD45 through the external, transmembrane, or membrane-proximal region of CD45 since p116 is associated with a mutant form of CD45 possessing a truncated cytoplasmic domain. The association of p116 with CD45 is not isoform-specific as p116 associates equally well with various CD45 isoforms. We have determined that p116 is a tyrosine-phosphorylated glycoprotein and that it is associated with CD45 in all hematopoietic cells examined. Because of its broad distribution, it is possible that identification of p116 will provide additional insight into the function of CD45 in lymphoid as well as non-lymphoid hematopoietic cells.

Developmentally regulated changes in glucosidase II association with, and carbohydrate content of, the protein tyrosine phosphatase CD45.

Glucosidase II (GII) stably interacts with the external domain of CD45 in a carbohydrate-dependent manner. We have found that the association occurs in immature cells, but is significantly reduced in mature T cells. Using mannose-binding protein (MBP), in both FACS analysis and pull-down assays, we find that MBP can specifically recognize cell surface CD45 from immature, but not mature T cells. Analysis of thymocytes reveals increased MBP binding and GII association with CD45 in double-positive thymocytes compared with either double-negative or single-positive thymocytes. As well, the same pool of CD45 recognized by MBP can also associate with GII. Initial analysis of the basis of the interaction between CD45 and MBP suggests MBP binds two different glycoforms of CD45 based on the differential competition with glucose. Finally, inhibition of GII activity in cells that do not normally express MBP ligands results in significant increases in cell surface MBP ligands, including CD45. Taken together, these data suggest that the glucose content of the cell surface CD45 changes as thymocytes undergo maturation to mature T cells, and may be regulated by GII interactions. Such changes in the cell surface carbohydrate on CD45 may affect the development of thymocytes, perhaps via binding of CD45 on thymocytes to lectins on stromal cells.

Immobilized antibodies to CD45 induce rapid morphologic changes and increased tyrosine phosphorylation of p56lck-associated proteins in T cells.

CD45 is a transmembrane protein tyrosine phosphatase required for signal transduction through the Ag receptor complexes of T and B lymphocytes. Herein, we demonstrate that immobilized mAbs to the external domain of CD45 induce rapid and dramatic morphologic changes in a variety of T cell lines, including CD8+ cytotoxic clones. CD45-induced morphologic changes can be inhibited by the cytoskeletal inhibitors cytochalasin D and E and by the protein tyrosine kinase inhibitor herbimycin A. Consistent with the requirement for tyrosine kinase activity, tyrosine phosphorylation of proteins at about 60 to 75 kDa and 115 to 130 kDa is increased upon engagement with immobilized anti-CD45 mAb with kinetics paralleling the observed changes in morphology. The phosphorylation of these proteins is inhibited by tyrosine kinase inhibitors at concentrations that also inhibit changes in morphology. The phosphoproteins induced when cells are added to immobilized anti-CD45 are co-immunoprecipitated with p56lck, suggesting that this tyrosine kinase might play a role in the phosphorylation of these proteins. Consistent with this, there is no increase in the phosphorylation of these proteins in p56lck-deficient CTLL-2 cells in response to immobilized anti-CD45 mAb. An important role for p56lck in the morphologic pathway is further supported by the observation that p56lck-deficient human J.CAM1.6 cells, in contrast to the parental Jurkat line, cannot be induced to undergo morphologic changes. Taken together, these results suggest a possible role for CD45 in coordinating a cytoskeletal remodeling cascade that may be important in cell activation.

Evidence for protein kinase C-dependent and -independent activation of mitogen-activated protein kinase in T cells: potential role of additional diacylglycerol binding proteins.

Activation of mitogen-activated protein kinases (MAPK) is a critical signal transduction event for CTL activation, but the signaling mechanisms responsible are not fully characterized. Protein kinase C (PKC) is thought to contribute to MAPK activation following TCR stimulation. We have found that dependence on PKC varies with the method used to stimulate the T cells. Extracellular signal-regulated kinase (ERK) activation in CTL stimulated with soluble cross-linked anti-CD3 is completely inhibited by the PKC inhibitor bisindolylmaleimide (BIM). In contrast, only the later time points in the course of ERK activation are sensitive to BIM when CTL are stimulated with immobilized anti-CD3, a condition that stimulates CTL degranulation. Surprisingly, MAPK activation in response to immobilized anti-CD3 is strongly inhibited at all time points by the diacylglycerol (DAG)-binding domain inhibitor calphostin C implicating the contribution of a DAG-dependent but PKC-independent pathway in the activation of ERK in CTL clones. Chronic exposure to phorbol ester down-regulates the expression of DAG-responsive PKC isoforms; however, this treatment of CTL clones does not inhibit anti-CD3-induced activation of MAPK. Phorbol ester-treated cells have reduced expression of several isoforms of PKC but still express the recently described DAG-binding Ras guanylnucleotide-releasing protein. These results indicate that the late phase of MAPK activation in CTL clones in response to immobilized anti-CD3 stimulation requires PKC while the early phase requires a DAG-dependent, BIM-resistant component.

Specific isoforms of the resident endoplasmic reticulum protein glucosidase II associate with the CD45 protein-tyrosine phosphatase via a lectin-like interaction.

We have previously demonstrated that CD45 physically associates with the endoplasmic reticulum processing enzyme glucosidase II (GII). GII consists of the catalytic alpha-chain and an associated beta-chain. To gain insight into the basis of the association between CD45 and GII, we examined the biochemical requirements for the interaction. We show that the alpha-subunit is essential for the interaction. Interestingly, only a higher molecular weight form of GIIalpha is capable of associating with CD45 in a competitive situation where multiple GIIalpha isoforms are expressed. Further, transfection studies demonstrate that only isoforms containing the alternatively spliced sequence Box A1 are capable of binding CD45, although all isoforms are catalytically active. The interaction between CD45 and GII is dependent on the active site of GII, is mediated through the carbohydrate on CD45, and can be inhibited with mannose. Taken together, these results suggest that GIIalpha acts as a lectin and binds to CD45 in an exon-dependent manner. This lectin activity of GII may be a novel mechanism for the regulation of CD45 biology and play a role in immune function, possibly by regulating CD45 glycosylation.

Alternative splicing of transcripts encoding the alpha- and beta-subunits of mouse glucosidase II in T lymphocytes.

Glucosidase II is a processing enzyme of the endoplasmic reticulum that functions to hydrolyze two glucose residues in immature N -linked oligosaccharides attached to newly synthesized polypeptides. We previously reported the cDNA cloning of the alpha- and beta-subunits of mouse glucosidase II from T cells following copurification of these proteins with the highly glycosylated transmembrane protein-tyrosine phosphatase CD45. Subsequent examination of additional cDNA clones, coupled with partial genomic DNA sequencing, has revealed that both subunits are encoded by gene products that undergo alternative splicing in T lymphocytes. The catalytic alpha-subunit possesses two variably expressed segments, box Alpha1, consisting of 22 amino acids located proximal to the amino-terminus, and box Alpha2, composed of 9 amino acids situated between the amino-terminus and the putative catalytic site in the central region of the molecule. Box Beta1, a variably expressed 7 amino acid segment in the beta-subunit of glucosidase II, is located immediately downstream of an acidic stretch near the carboxyl-terminus. Screening of reverse transcribed RNA by polymerase chain reaction confirms the variable inclusion of each of these segments in transcripts obtained from a panel of T-lymphocyte cell lines. Thus, distinct isoforms of glucosidase II exist that may perform specialized functions.

Cytotoxic T lymphocytes express a beta3 integrin which can induce the phosphorylation of focal adhesion kinase and the related PYK-2.

Fibronectin has been shown to stimulate tyrosine phosphorylation of a number of proteins in the 115-125 kDa range and facilitate degranulation by alloantigen-specific cytotoxic T lymphocyte (CTL) clones in response to substimulatory amounts of anti-CD3 or anti-T cell receptor (TCR). The current study was initiated to further characterize integrin expression and usage by these CTL clones. We demonstrate that vitronectin and fibrinogen, but not laminin or collagen, are also able to both facilitate degranulation in the presence of substimulatory anti-CD3 and stimulate tyrosine phosphorylation of these 115-125-kDa proteins, with a 115-kDa protein being the most prominently phosphorylated. These results implicate the expression and usage of the vitronectin receptor, alpha beta3 integrin, by these CTL clones. We demonstrate by both flow cytometry and immunoprecipitation that CTL clones do in fact express beta3 integrin. Immobilized antibody to beta3 stimulates the phosphorylation of the 115-125-kDa proteins, suggesting that engagement of beta3 transmits the same signal into these cells as fibronectin or vitronectin. The fibronectin and vitronectin-induced phosphorylation as well as adhesion to either fibronectin or vitronectin can be significantly inhibited with antibodies to beta3 integrins. Finally, we are able to immunoprecipitate 115-kDa proteins with antiserum to focal adhesion kinase and a related kinase, called PYK-2, that becomes phosphorylated in response to vitronectin or immobilized anti-beta3. Taken together, these results demonstrate that CTL express and use beta3-integrins as signaling molecules which can augment TCR-mediated stimulation.

Cytotoxic T lymphocyte mediated lysis without release of serine esterase.

Cloned cytotoxic T lymphocytes (CTL) can lyse target cells in a Ca2+-dependent or Ca2+-independent fashion, depending on the target cell used. We have used these Ca2+-free assay conditions to determine whether there is serine esterase release in the absence of extracellular Ca2+. We find that under conditions where there is significant target cell killing in the absence of Ca2+, there is no detectable serine esterase release. To the extent that serine esterase and perforin-like molecules reside in the same granules, these results also suggest that target cell killing may occur in the absence of degranulation. Our results provide an example of target cell killing without serine esterase release, and the first indication that lysis and degranulation can be functionally dissociated.

Sustained TCR signaling is required for mitogen-activated protein kinase activation and degranulation by cytotoxic T lymphocytes.

Requirements for T cell activation are not fully established. One model is that receptor occupancy and down-regulation are essential for activation, and another, not necessarily mutually exclusive, model is that sustained signals are important. Here we examine the importance of signal duration in T cell activation. First, we demonstrate that immobilized, but not soluble cross-linked, Abs to CD3 stimulate degranulation by CTL. The cross-linked Abs are not deficient in their ability to signal since they stimulate the same tyrosine phosphorylation pattern as immobilized Ab, but it is very transient relative to that stimulated by immobilized Ab. Furthermore, novel decreased migratory forms of Lck occur to a significant extent only after stimulation with immobilized Abs. A dramatic difference in the duration of signals is very evident when mitogen-activated protein kinase (MAPK) activity is examined. Immobilized anti-CD3 stimulates very high levels of MAPK activation that is still detectable 1 h after stimulation. In contrast, cross-linked Ab stimulates only transient and incomplete activation of MAPK. Taken together, these results suggest that TCR engagement and induction of tyrosine phosphorylation alone are not sufficient for T cell activation and that the duration of TCR-stimulated signals is critical to attain a functional response.

Mutational analysis of CD45. A leukocyte-specific protein tyrosine phosphatase.

The cytoplasmic domain of murine CD45 has been expressed using an in vitro transcription/translation system. The recombinant protein was isolated by immunoprecipitation with a specific rabbit antiserum and was shown to have protein tyrosine phosphatase (PTPase) activity. Oligonucleotide-directed mutagenesis was then used to investigate the structural requirements for PTPase activity. Although the cysteine crucial for PTPase activity resides in domain I, this domain was not active alone. Both PTPase domains of CD45 and the membrane proximal region of 77 amino acids were required for enzymatic activity. Deletion of 78 residues at the carboxyl terminus of the cytoplasmic region did not influence activity, but an additional deletion of 13 amino acids from PTPase domain II totally abolished activity. Excision of the 21-residue acidic insert in the second PTPase domain resulted in a decrease of activity of approximately 4-fold. Nine conserved residues around the critical cysteine in the first domain were found to be important for activity. Of the 3 conserved tyrosine residues in domain I, only Tyr729 was specifically required for activity.