| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Gastroenterology Division, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA 02115;
Department of Immunology, Imperial College, London, U.K.; and
Institute for Immunology, Charité-Humboldt University, Berlin, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
), CEACAM1 containing either a long or a short cyt tail inhibited or costimulated, respectively, TCR/CD3 complex plus CD28 mediated activation with the inhibitory functions of the long cyt tail dominating. The inhibitory function of CEACAM1, was dependent upon src homology domain 2 containing phosphatase 1 activity, required both tyrosine residues within the immunoreceptor tyrosine-based inhibitory motif domains of the cyt tail and was mediated through the mitogen-activated protein kinase pathway. CEACAM1-mediated inhibition could be functionally reconstituted by incubation of PBMC with either a CEACAM1-specific mAb or CEACAM1-Fc fusion protein in the presence of an allogeneic or mitogenic stimulus, respectively. These studies indicate that the long and short cyt tails of CEACAM1 serve as inhibitory and costimulatory receptors, respectively, in T cell regulation. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In addition to expression on a wide variety of epithelial cell types, CEACAM1 is expressed by endothelial cells and a variety of hemapoietic cell types including granulocytes, NK cell, B cells, monocytes, and dendritic cells (18, 19, 20, 21). On several of these cell types, CEACAM1 expression requires the cell to be activated. This is consistent with the recent appreciation that CEACAM1 is an activation-induced Ag on the cell surface of human and mouse T cells (22, 23, 24, 25, 26, 27). The function of CEACAM1 on T lymphocytes is unclear. Although it is tempting to consider CEACAM1 as an inhibitory receptor given the presence of two ITIMs within the long cyt tail, ligation of CEACAM1 with either CEACAM1-specific Abs or bacterial products known to bind CEACAM1 heterophilically have resulted in either T cell activation (23, 26) or inhibition (24, 25, 27). One possible explanation for these contradictory results is that they reflect the consequences of ligating different CEACAM1 isoforms. Virtually all studies performed to date have examined primary mouse or human T cells, which are known to express various combinations of both the long and short cyt domain isoforms. More specifically, there have been no carefully performed functional assessments of the downstream pathways associated with the CEACAM1 long and short cyt tail domain-containing isoforms including their interactions in T cells and the specific role of the ITIM domains. Therefore, in this study, we sought to determine the functional behavior of CEACAM1 expressing three extracellular Ig-domains that differ only in expression of either a long (CEACAM1–3L) or short (CEACAM1–3S) cyt tail in the Jurkat cell line, which has been previously shown to be devoid of CEACAM1 expression even after activation (26). We found that the human CEACAM1–3L isoform has the properties of an ITIM-bearing inhibitory receptor in that inhibition of TCR/CD3 complex signaling is affected by this isoform and is dependent upon the presence of two functional tyrosine residues in the cyt tail. These two ITIMs are functionally linked to the inhibition of Th1 and Th2 cytokine secretion and, via SHP-1, to the specific inhibition of mitogen-activated protein kinases (MAPK) comprising the c-Jun N-terminal kinase (JNK) and extracellular response kinase (ERK). In contrast, the short cyt tail of human CEACAM1 behaves as an activating receptor whose function is dominated by the inhibitory function of the long cyt domain-containing isoform. These studies help to clarify the function of CEACAM1 in T lymphocytes and support consideration of the CEACAM1 long cyt tail-containing isoforms as inhibitory receptors and the short cyt tail-containing isoforms as costimulatory receptors.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Human PBMC were purified from whole blood of unrelated donors by centrifugation on a Ficoll/Hypaque gradient (Amersham Pharmacia, Piscataway, NJ) by standard methods. Jurkat cells stably transfected with SV40-large T Ag (Jurkat-T) have been previously described (28). COS-7 cells were obtained from the American Tissue Type Culture Collection (Manassas, VA) and cultured in complete medium containing DMEM.
Reagents and Abs
The mouse anti-human CEACAM1 mAbs, 5F4 and 26H7, are of the IgG1 isotype and generated as previously described (24). Mouse anti-human CD3 and CD28 mAbs and isotype control mouse IgG1
and IgG2a mAbs were purchased from BD PharMingen (San Diego, CA). Goat anti-mouse IgG (Fc) Ab was purchased from Pierce (Rockford, IL). The NFAT/AP1-luciferase (LUC) reporter plasmid was kindly provided by Dr. S. Burakoff (Dana-Farber Cancer Institute, Boston, MA) (29). Mouse anti-phosphorylated (p)-ERK, pJNK, and p-p38 as well as rabbit anti-ERK, JNK, and p38 were the products of Santa Cruz Biotechnology (Santa Cruz, CA).
Construction of plasmids
The human CEACAM1–3L and -3S expression constructs were generated in pEF6/V5-HisB vector (Invitrogen, Carlsbad, CA) by PCR amplification of the template pPAN3.1 DNA (23) with the following primers. The 5' and 3' sequences of the primer used were 5'-CGAATTCACCATGGGGCACCTCTCA-3' (primer A), 5'-GGTCTAGATACTGCTTTTTTACTTC-3' (primer B), 5'-ATCTAGACATTGGAGTGGTCCTGAGCTGCCGGTCTTCCCGAAAT-3' (primer C). Primers A and B and A and C were used for the construction of the CEACAM1–3L and -3S expression vectors, respectively. EcoRI and XbaI restriction sites were introduced at the 5' and 3' ends of the primers for site-specific cloning. The CEACAM1–3LY459A mutant containing an alanine substitution of Y459 was generated by PCR-directed mutagenesis of the CEACAM1–3L template DNA using the following primers: 5'-GATGAATGAAGTTACTGCTTCTACCCTGAAC-3' (forward primer), 5'-GTTCAGGGTAGAAGCAGTAACTTCATTCATC-3' (reverse primer) with the QuickChange Site-Directed Mutagenesis kit (Invitrogen). cDNAs encoding CEACAM1–3LY486A containing an alanine substitution of Y486 and CEACAM1–3LY459AY486A containing alanine substitutions of Y459 and Y486 were generated by PCR amplification of the CEACAM1–3L and CEACAM1–3LY459A templates, respectively, with the following primers: 5'-GATGAATGAAGTTACTGCTTCTACCCTGAAC-3' (forward primer), and 5'-CCTCTAGATGCTGCTTTTTTACTTCTGAAGCAATTATTTC-3' (reverse primer).
Generation of CEACAM1-Fc-fusion protein
The ScFvExpress-sec vector, which can induce the synthesis and secretion of proteins in eukaryotic cells using an EF-BOS promoter in front of the human IgG1 secretory leader peptide sequence (30), was digested with SalI and NotI, dephosphorylated with shrimp alkaline phosphatase (Amersham Pharmacia), and ligated to a SalI-NotI fragment (693 bp) containing the coding sequence of the human Fc (kindly provided by Dr. W. Lencer, Children’s Hospital Medical Center, Boston, MA). The extracellular domain of the human BGPb (CEACAM1–3L) coding sequence was amplified from pTPAN1, a pTZ19 vector with an XbaI fragment (1.3 kbp) containing the cDNA sequence for human BGP (24). The primers used for this amplification were PANSAL (5'-CTGATCCTGTCGACCAGCTCACTACTGAATCCATGC-3') and ADAPSABA (5'-AGTTTTGTCGACTGGACCGCCCGAACCGCCACCGGATCCACTATTATTCACAG-3'). The amplified fragment was digested with SalI, gel purified using NAC45 paper (Schleicher & Schuell, Kenne, NH), ligated to the dephosphorylated SalI-linearized pScHuFc3 vector and transformed into competent E. coli cells. One of these plasmids (pBGPFc1) was used for transfection of CHO cells using Effectine (Qiagen, Valencia, CA) for production of CEACAM1-Fc fusion protein.
MLR and PHA stimulation
Allogeneic MLR and PHA stimulation of PBMCs was performed by standard methods as previously described (31). Responder PBMCs were stimulated with equal numbers of irradiated (3000 rad, 13 min) stimulator PBMCs at concentrations of 106 cells/ml for 24 h. PBMCs were exposed to PHA (2 µg/ml) for 24 h.
Transfection studies
COS-7 cells were transfected with 2 µg of each plasmid DNA per well in a six-well tissue culture plate using LipofectAmine Plus reagent (Life Technologies) according to the manufacturer’s instructions and Western blotting performed by standard methods with an anti-CEACAM1 mAb, 5F4, at a concentration of 10 µg/ml using an enhanced chemiluminescence reagent (New England Nuclear, Boston, MA) for detection.
Jurkat-T cells (106) were transfected with the 1 µg of plasmids encoding mutated or nonmutated forms of CEACAM1 cloned into pEF6/V5-His B or empty vector alone as control in combination with 0.5 µg of NFAT/AP1-LUC reporter plasmid using GenePorter transfection reagent (Gene Therapy System, San Diego, CA). Twenty-four hours after transfection, cells (5 x 105/condition) were stimulated with anti-human CD3 mAb (0.4–10 µg/ml) and anti-human CD28 mAb (0.08–10 µg/ml) or isotype-matched control Abs followed by goat anti-mouse Fc Ab (3 µg/ml) for 6 h at 37°C. For the control mAbs for anti-CD3 and anti-CD28, mouse IgG1 and mouse IgG2a were used, respectively. Six hours after stimulation, cells were solubilized with 100 µl of cell culture lysis reagent (Promega, Madison, WI). Cell extracts were combined with Promega Luciferase Assay Reagent and luciferase activity determined using a TD-20/20 luminometer (Turner Designs, Sunnyvale, CA). The protein concentration of the lysates was determined by bicinchoninic acid protein assay reagent (Pierce) and the luciferase results normalized for the protein content. Alternatively, the cells were collected for Western blot analysis at various times following transfection or the supernatants of the transfectants removed for ELISA (BD PharMingen) based cytokine production assays at either days 1, 2, or 3 following transfection.
Western blotting
The cells were washed twice in PBS and lysed in 200 µl of 1% Triton X-100. The protein content was determined using a bicinchoninic acid protein assay kit (Pierce). Proteins (20 µg) were separated by SDS-PAGE on 10% polyacrylamide gels and then transferred to nitrocellulose membrane. Following blocking of nonspecific binding by 5% nonfat milk, the membranes were treated with pERK, pJNK, or p-p38 mAb for 1 h at room temperature. The membranes were then washed and incubated with HRP-conjugated, goat anti-mouse IgG (1:2000; DAKO, Carpinteria, CA) and developed using Western blot chemiluminescence reagent (NEN Life Science Products, Boston, MA). Following demonstration of pJNK, p-p38, or pERK, the nitrocellulose membranes were stripped by treatment with 250 mM glycine-HCl, pH 2.0, 1% SDS for 30 min and total MAPK levels were detected in the same membranes using rabbit anti-ERK, JNK, or p38 Abs, respectively. Densitometry was performed for standardization of the phosphorylated MAPK species relative to the total MAPK fractions.
SHP-1 gene silencing by means of small interfering (si) RNA
To selectively silence human SHP-1 gene expression in Jurkat-T cells, a specific siRNA approach was developed. The cDNA sequence was targeted at 5'-AATGCGGCTGACATTGAGAAC-3' (654–675; GenBank no. AF178946). siRNA duplexes were generated consisting of a sense strand (5'-UGCGGCUGACAUUGAGAACdTdT-3') and an anti-sense strand (3'-dTdTACGCCGACUGUAACUCUUG-5') (Xeragon, Germantown MD). siRNA (1 µg) was used to transfect 5 x 105 Jurkat-T cells using the TransMessenger Transfect kit (Qiagen) following the manufacturer’s instructions. To confirm the posttranscriptional gene silencing effect of SHP-1 siRNA, transfected Jurkat-T cells were harvested 48 h after transfection. Total cellular RNA was isolated using TRIzol (Invitrogen) following the manufacturer’s instructions. RNA (100 ng) was subjected to reverse-transcription using the Advantage RT-for-PCR Kit (Clontech Laboratories, Palo Alto, CA). The sense and antisense primers for human SHP-1 were 5'-GGAGAAGTTTGCGACTCTGA-3' and 5'-GGGAGAATGTTCTTGTAGCG-3' (634 bp). The conditions for the PCR were as follows: after initial denaturation at 95°C for 5 min, the thermocycler (MJ Research, Watertown, MA) was programmed for 35 cycles: 1 min at 95°C, 1 min at 55°C and 2 min at 72°C. The reaction was concluded with a final extension step at 72°C for 7 min. PCR products were separated and visualized with 1.5% agarose gel containing 0.01% ethidium bromide. OD of the cDNA bands were determined by the computerized image-analysis system and normalized to RT-PCR products of 
-actin.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous studies have shown that the Jurkat cell line cannot be induced to express CEACAM1 making it a useful tool to examine the function of the human (h)CEACAM1–3L and hCEACAM1–3S isoforms in human T cells (26). As shown in Fig. 1A, this was confirmed by the fact that Jurkat-T cells did not express CEACAM1 either before or after (3–24 h) activation with a variety of stimulatory conditions but did so after transfection with either the CEACAM1–3S or CEACAM1–3L isoforms. COS-7 cells, as a control (Fig. 1B), or the Jurkat-T cell line (Fig. 1C) were thus transfected with either an empty vector, as control, or plasmids containing DNA fragments encoding the CEACAM1–3L and/or CEACAM1–3S isoforms. As can be seen in Fig. 1C, confirming previous studies with Jurkat cells (26), Jurkat-T cells failed to express CEACAM1 unless transfected with the CEACAM1-encoding plasmids. Moreover, an evaluation of the time course of CEACAM1 protein expression after transient transfection in the Jurkat-T cells showed detectable evidence of CEACAM1 after 24 h with peak expression at 48–72 h for both the CEACAM1–3L and CEACAM1–3S isoforms (Fig. 1D). These studies (Fig. 1, B--D) show that the CEACAM1–3L- and CEACAM1–3S-encoding plasmids were equally efficient at driving CEACAM1 protein synthesis.
|
5-fold increase in luciferase activity in contrast to the CEACAM1–3L isoform, which lacked any stimulatory function (Fig. 2A, left panel). However, the CEACAM1–3L isoform was not inert in that it significantly inhibited the baseline and spontaneous CEACAM1–3S isoform-stimulated luciferase activity when transfected at an equimolar DNA ratio to the CEACAM1–3S isoform (Fig. 2, left panel). When the Jurkat-T cells were exposed to TCR/CD3 complex plus CD28 stimulation under cross-linking conditions, luciferase activity was observed to increase 5-fold above baseline levels (Fig. 2A, right panel). Consistent with the results observed in the absence of TCR/CD3 complex stimulation, transfection of the CEACAM1–3L isoform significantly inhibited the TCR/CD3 complex plus CD28-stimulated activation of the NFAT-AP1-directed luciferase reporter activity in the absence or presence of the CEACAM1–3S isoform (Fig. 2A, right panel). Similar observations were made when proliferation of Jurkat-T cells was used as the experimental endpoint (data not shown).
|
Inhibitory function of the human CEACAM1–3L isoform in T cells is dependent upon both functional ITIM domains
The human CEACAM1–3L cyt tail contains two tyrosine residues as part of ITIMs (5, 7). Previous studies have directly linked these ITIMs to inhibitory functions in mouse epithelial cells (32) and human B lymphocytes (19) and indirectly to T cells due to the ability of the CEACAM1 cyt tail to associate with SHP-1 and SHP-2 in human and mouse T cells (25, 27). However, the role of the two ITIMs in T cell function has not been directly examined. To determine whether the two ITIMs are important in the inhibitory function of the CEACAM1–3L isoform, alanine substitutions were made at the Y459 and/or Y486 residues in the cyt tail of the CEACAM1–3L isoform. These three CEACAM1–3L mutants (3L-Y459A, 3L-Y486A, and 3L-Y459A/Y486A) were examined for their ability to inhibit TCR/CD3 complex-mediated stimulation of T cell. As can be seen in Fig. 3A, deletion of both the Y459 and Y486 residues but not either alone resulted in abrogation of the ability of the CEACAM1–3L isoform to inhibit spontaneous (left panel) and anti-CD3 plus anti-CD28 stimulated (right panel) activation of NFAT-AP1 luciferase reporter activity. The biologic significance of this inhibition was shown by the ability of the CEACAM1–3L isoform to inhibit IL-2 secretion induced by anti-CD3 plus anti-CD28 stimulation of the Jurkat-T cell line, the loss of this inhibition in the context of deletion of both the Y459 and Y486 residues in the cyt tail (YY) and the opposing effects of the -L and -S isoforms on IL-2 secretion when transfected together (LS). (Fig. 3B). These studies indicate that the two tyrosine residues of the cyt tail of human CEACAM1–3L are functional in T cells and likely to be part of ITIMs given that their elimination leads to a loss of inhibitory function.
|
The fact that the stimulatory and inhibitory functions of the CEACAM1–3S and CEACAM1–3L isoforms, respectively, were elicited in the absence of addition of any exogenous factors suggested that the newly expressed CEACAM1–3S or CEACAM1–3L isoforms on the transfected Jurkat-T cell line were being ligated homophilically in a homotypic interaction between the transfected Jurkat-T cells. Moreover, given the fact that human T cells have been previously shown to simultaneously express CEACAM1 with both a long and short cyt tail (26), given the results described above, it would be predicted that the functional consequence of CEACAM1 ligation would be predominantly inhibitory in a T cell. In addition, it is well known that CEACAM1 is characterized by homophilic interactions in other cell types such as epithelia (1, 2, 3). Therefore, to determine whether this was indeed the case and whether this was functionally linked to T cell inhibition, the following studies were performed. First, it was observed that the human CEACAM1-specific mAb, 5F4, significantly inhibited an allogeneic MLR in a dose-dependent manner (Fig. 4A). Similar observations and concentration dependencies were observed with another human CEACAM1-specific mAb, 26H7 (data not shown). Second, we constructed a CEACAM1-Fc fusion protein that consisted of the extracellular portion of the CEACAM1–3L isoform linked to the Fc-domain of human IgG1. This CEACAM1-Fc fusion protein significantly inhibited the activation of human PBMCs stimulated by either PHA (Fig. 4B) or an allogeneic stimulus in a dose-dependent fashion (data not shown). Moreover, the inhibition elicited by the CEACAM1-Fc fusion protein was nearly as potent as that observed with a CTLA4-Ig fusion protein. Taken together with previous observations that CEACAM1 is involved in homophilic interactions (1, 2, 3) and that T cells do not express any of the known heterophilic ligands for CEACAM1, namely other CEACAM-related molecules (23, 24), these studies suggest that the mechanism for CEACAM1 ligation on human T cells is likely through homophilic adhesion and that this interaction is predominantly inhibitory.
|
Given that the CEACAM1–3L isoform exhibited direct inhibitory effects on TCR/CD3 complex function, we next sought to define the mechanism of the observed inhibition. Since CEACAM1 has been recently shown to regulate Th1 cytokine production in untransfected spleen T cells (41) and that Th differentiation is influenced by MAPK pathways (33), we investigated the effects of CEACAM1 on MAPK activation induced by TCR/CD3 complex stimulation and CD28 costimulation. Resting Jurkat-T cells exhibited negligible levels of pJNK and pERK that rose substantively after anti-CD3 plus anti-CD28 stimulation (Fig. 5, A and B, respectively). Whereas, transfection of Jurkat-T cells with the CEACAM1–3S isoform enhanced slightly the levels of pJNK (Fig. 5A) and pERK (Fig. 5B) when stimulated by anti-CD3 plus anti-CD28 stimulation consistent perhaps with the costimulatory function of this isoform as shown above, transfection with the CEACAM1–3L isoform significantly inhibited these levels relative to that observed with the vector control. In contrast, CEACAM1 transfection had no significant effects on the levels of p-p38 (Fig. 5C). Cotransfection of Jurkat-T cells with equimolar quantities of the CEACAM1–3S and CECAM1–3L isoforms resulted in decreased inhibition imposed by the CEACAM1–3L isoform on the levels of pJNK and pERK consistent with the opposing functional effects of the CEACAM1–3L and CEACAM1–3S isoforms as described above.
|
and IL-4 levels were all below 50 pg/ml/106 cells in the various groups analyzed (Fig. 6, A and B, respectively). In contrast, in the presence of anti-CD3/CD28 Ab stimulation, the Jurakat-T cells exhibited an 3-fold increase in the secretion of IFN- (Fig. 6A) and IL-4 (Fig. 6B). As noted above, transfection with the CEACAM1–3S isoform caused a further slight up-regulation in secretion of these two cytokines above the vector control. In contrast, transfection with the CEACAM1–3L isoform resulted in significant inhibition of both IFN- and IL-4 secretion with a decrease in the inhibition when the two ITIM domains were deleted by mutation of the Y459 and Y486 residues (YY) and opposing effects of the -L and -S isoforms (LS) when transfected together. These studies suggest that CEACAM1–3L, by virtue of the two ITIM domains in the cyt tail, inhibits pJNK and pERK MAPK activity and the downstream activation of both Th1 and Th2 cytokine secretion when examined in a transfected cellular system as shown here.
|
Since CEACAM1 has been shown to associate with SHP-1 in epithelial cells through direct interactions with the ITIM domains of the long cyt tail and SHP-1 is known to negatively regulate MAPK pathways (35), we reasoned that the inhibition of TCR-mediated activation of MAPK pathways by CEACAM1–3L may occur through SHP-1. This possibility was assessed by analyzing the ability of the CEACAM1–3L isoform to inhibit the activation of JNK and ERK as a consequence of TCR/CD3 plus CD28 activation in the presence and absence of SHP-1. To do so, SHP-1 transcription was silenced by exposure of Jurkat-T cells to control and SHP-1 specific RNA specific oligomers. As shown in Fig. 7A, whereas the basal levels of SHP-1 mRNA were inhibited >90% by the SHP-1 specific oligomer, there was no effect of the control inhibitory oligomer. Using gene silencing to assess the role of SHP-1 on inhibition of MAPK activation by CEACAM1, it was observed that whereas CEACAM1 could inhibit the generation of pJNK and pERK as a consequence of anti-CD3 plus anti-CD28 activation in the presence of the control oligomer and thus in the presence of SHP-1, CEACAM1 could not inhibit pJNK and pERK generation in either the absence of SHP-1 as caused by gene silencing with the SHP-1 specific oligomer or in the absence of the two ITIM domains as caused by alanine substitutions of the Y459 and Y486 residues (Fig. 7B). Silencing of SHP-1 expression in the Jurkat-T cell line also reversed the inhibitory function of CEACAM1–3L isoform in modulating IL-2 (Fig. 3B), IFN- (Fig. 6A) and IL-4 (Fig. 6B) secretion (compare Lc and Li). These studies prove that CEACAM1–3L inhibition of MAPK activation after ligation of the TCR/CD3 complex in the presence of CD28 costimulation is due to ITIM-dependent regulation of SHP-1.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous studies that have characterized CEACAM1 isoform expression in human and mouse T cells have used RT-PCR amplification, cloning, and sequencing since there are no mAbs available that are able to discriminate CEACAM1 isoforms. These studies have shown that human and mouse T cells express the BGP a, b, c, and d (or CEACAM1–4L, 3L, 4S and 3S, respectively) splice variants (26, 37). In the limited data available, the long cyt domain-containing isoforms appear to be expressed in excess over the short cyt domain-containing isoforms. Given our results in the Jurkat system that the inhibitory functions of the long cyt tail are dominant over the stimulatory functions of the short cyt tail, it might be predicted that the outcome of CEACAM1 expression on a T cell is a dampening effect on TCR/CD3 complex function. Moreover, this inhibitory effect is likely to be stimulated homophilically based upon results with the Jurkat transfection system and the human CEACAM1-Fc fusion protein as shown here, which draws comparisons to CD22, a similarly functioning molecule (38). To further confirm that the mechanism of inhibition is homophilic ligation of CEACAM1 and stimulation of an inhibitory pathway as opposed to inhibition of an activating pathway, future studies should be aimed at examining the effects of monovalent reagents such as Fab that would bind directly to the homophilic binding site without cross-linking. Whether there are circumstances in which the costimulatory functions of the short cyt tail domain-containing isoforms dominate remains to be determined but would be predicted to be the case under certain conditions and may suffice to explain previous stimulatory functions ascribed to this molecule.
Like the inhibitory functions of CEACAM1 observed in mouse intestinal epithelial cells (7, 32) and human B cells (19), the tyrosine residues within the ITIMs appear to be important for CEACAM1–3L function in T cells. However, there are interesting differences which suggest that the mechanism(s) by which CEACAM1–3L isoform functions in human T cells may be distinct. First, in contrast to the long cyt tail in mouse intestinal epithelial cells, which requires the short cyt tail-containing isoform for inhibitory function (7) the inhibitory effect of the long cyt tail in T cells on TCR/CD3 complex function can be detected independently of the short cyt tail. The function of the long cyt tail has also been examined in human B cell lines by expressing a chimera of the extracellular domain of FcRIIb and the long cyt tail of CEACAM1 (19). In these studies, the long cyt tail of CEACAM1 inhibited calcium mobilization in B cells initiated by stimulation of the B cell receptor complex. This inhibition was dependent upon either SHP-1 or SHP-2 and the Y459 residue with the Y486 residue being dispensable. In contrast, our studies in T cells showed that both the Y459 and Y486 contribute significantly toward the inhibitory function of CEACAM1. The role of the two ITIMs in controlling the inhibitory function of CEACAM1 is further supported by our observation that the mouse CEACAM1–2L splice variant controls the effector function of T cells in the CD4+CD45RBhigh transfer model but only in the presence of the ITIMs and SHP-1 (T. Nagaishi and R. S. Blumberg, unpublished observations).
Our studies also revealed that the ITIM-dependent inhibitory functions of the CEACAM1–3L isoform are linked to MAPK pathways and specifically ERK and JNK but not p38. ERK was first identified as a downstream mediator of the Ras oncogene (33). Furthermore, this pathway is required for Th2 differentiation as shown by Yamashita and colleagues using ERK activation-compromised H-Ras transgenic mice (39). In contrast, JNK significantly regulates Th1 differentiation, as revealed by the fact that T cells from JNK-/- mice produce significantly less IFN- (40) and, as a corollary, JNK1-deficient mice exhibit an exaggerated Th2 response (34). Our findings that the CEACAM1–3L isoform significantly suppresses both JNK and ERK activation and regulates both Th1 and Th2 cytokine secretion after TCR/CD3 complex mediated activation is consistent with these observations. These results also extend recent observations obtained from rat pheochromocytoma PC12 cells which showed that anti-CEACAM Abs were able to activate ERK (37). Moreover, they are likely to be physiologically relevant because we have observed similar inhibition of JNK and ERK activation by ligation of CEACAM1 either homophilically or heterophilically in an adoptive transfer model of colitis and a hapten-mediated colitis model, respectively, in association with diminished immunopathology and a specific decrease in Th1 cytokine production (Ref. 41 , and unpublished observations). The inhibition of both Th1 and Th2 cytokines by CEACAM1 as shown here in this in vitro model system suggests that overexpression of CEACAM1 can inhibit secretion of both types of cytokoines.
Ligation of CEACAM1 on human CD4+ T cells by the Opa52 Ag of Neisseria gonorrhea leads to association of SHP-1 and SHP-2 with CEACAM1 (25). In addition, the long cyt tail of mouse CEACAM1 can associate with SHP-1 in T cells when phosphorylated but not when nonphosphorylated using CEACAM1 fusion proteins in pull-down assays (27). Together with this and the observation that deletion of the two ITIMs within the cyt tail of the CEACAM1–3L isoform abrogated the inhibition of MAPK pathways, we assessed the role of SHP-1 in mediating the inhibitory effects of the CEACAM1–3L isoform using siRNA mediated gene silencing. siRNA represents double stranded RNA containing 21–23 nucleotides which induces mRNA degradation in a sequence-specific fashion (42). While it is increasingly evident that siRNA acts as a potent method for posttranscriptional gene silencing in mammalian cells, its effect on Src homology 2-containing phosphatases has not been previously explored. As there are no specific chemical inhibitors of SHP-1 function, the proven efficacy of siRNA as reported herein indicates that it may be an important tool for manipulation of downstream signaling pathways in T cells. It was specifically observed that cotransfection with SHP-1 specific siRNA significantly decreased SHP-1 gene expression in human T cells and in turn dramatically abrogated the inhibitory action of the CEACAM1–3L isoform on the activation of JNK and ERK as induced by the TCR/CD3 complex. These results are consistent with reports that JNK is significantly increased in SHP-1 deficient B cells (43) and that SHP-1 positively regulates B cell apoptosis by negatively modulating JNK activation (44). Taken together, these studies predict that in contrast to B cells, both tyrosine residues of CEACAM1 are involved in the inhibitory function as mediated by SHP-1 independently of any contribution of the short cyt tail containing isoforms.
In summary, by an examination of CEACAM1 isoforms in Jurkat cells, which do not normally express CEACAM1, we have shown that distinct stimulatory and inhibitory functions can be assigned to a short cyt domain- and long cyt domain-containing isoforms of CEACAM1, respectively. Although the mechanism for stimulation by CEACAM1–3S remain to be defined, the inhibitory function of CEACAM1–3L in T cells appears to be mediated by the participation of the two ITIMs contained within the long cyt tail which direct the SHP-1 mediated inhibition of the JNK and ERK MAPKs. These studies draw clear functional similarities between CEACAM1 splice variants and KIRs and KARs. This is further supported by evidence that CEACAM1 expression on activated human T cells, NK and decidual NKT cells is associated with inhibition of cytotoxic T cell and NK activity, respectively (45). Defining the regulation of CEACAM1 splice product generation and the mechanisms that underlie these functional attributes of CEACAM1 will be crucial future areas worthy of investigation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 D.C., H.I., and T.N. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Richard S. Blumberg, Laboratory of Mucosal Immunology, Gastroenterology Division, Department of Medicine, Brigham and Women’s Hospital/Harvard Medical School, 75 Francis Street, Thorn 1419, Boston, MA 02115. E-mail address: rblumberg{at}partners.org
4 Abbreviations used in this paper: CEACAM1, carcinoembryonic Ag-related cellular adhesion molecule 1; BGP, biliary glycoprotein; cyt, cytoplasmic; ITIM, immunoreceptor tyrosine-based inhibitory motif; SHP-1, src homology domain 2 containing phosphatase 1; MAPK, mitogen-activated protein kinases; JNK, c-Jun N-terminal kinase; ERK, extracellular response kinase; si, small interference; p, phosphorylated; KIR, killer inhibitory receptor; KAR, killer activating receptor.
Received for publication June 24, 2003. Accepted for publication January 5, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. S. W. Lee, M. A. Ostrowski, and S. D. Gray-Owen CEACAM1 Dynamics during Neisseria gonorrhoeae Suppression of CD4+ T Lymphocyte Activation J. Immunol., May 15, 2008; 180(10): 6827 - 6835. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Chen, L. Chen, S.-W. Qiao, T. Nagaishi, and R. S. Blumberg Carcinoembryonic Antigen-Related Cell Adhesion Molecule 1 Inhibits Proximal TCR Signaling by Targeting ZAP-70 J. Immunol., May 1, 2008; 180(9): 6085 - 6093. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Monu and A. B. Frey Suppression of Proximal T Cell Receptor Signaling and Lytic Function in CD8+ Tumor-Infiltrating T Cells Cancer Res., December 1, 2007; 67(23): 11447 - 11454. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. S. W. Lee, I. C. Boulton, K. Reddin, H. Wong, D. Halliwell, O. Mandelboim, A. R. Gorringe, and S. D. Gray-Owen Neisserial Outer Membrane Vesicles Bind the Coinhibitory Receptor Carcinoembryonic Antigen-Related Cellular Adhesion Molecule 1 and Suppress CD4+ T Lymphocyte Function Infect. Immun., September 1, 2007; 75(9): 4449 - 4455. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Schmitter, S. Pils, V. Sakk, R. Frank, K.-D. Fischer, and C. R. Hauck The Granulocyte Receptor Carcinoembryonic Antigen-Related Cell Adhesion Molecule 3 (CEACAM3) Directly Associates with Vav to Promote Phagocytosis of Human Pathogens J. Immunol., March 15, 2007; 178(6): 3797 - 3805. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Yu, E. M. C. Chow, H. Wong, J. Gu, O. Mandelboim, S. D. Gray-Owen, and M. A. Ostrowski CEACAM1 (CD66a) Promotes Human Monocyte Survival via a Phosphatidylinositol 3-Kinase- and AKT-dependent Pathway J. Biol. Chem., December 22, 2006; 281(51): 39179 - 39193. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Albarran-Somoza, R. Franco-Topete, V. Delgado-Rizo, F. Cerda-Camacho, L. Acosta-Jimenez, M. Lopez-Botet, and A. Daneri-Navarro CEACAM1 in Cervical Cancer and Precursor Lesions: Association With Human Papillomavirus Infection J. Histochem. Cytochem., December 1, 2006; 54(12): 1393 - 1399. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Wang, Z. Li, R. Ding, G. D. Frank, T. Senbonmatsu, E. J. Landon, T. Inagami, and Z. J. Zhao Antagonism or Synergism: ROLE OF TYROSINE PHOSPHATASES SHP-1 AND SHP-2 IN GROWTH FACTOR SIGNALING J. Biol. Chem., August 4, 2006; 281(31): 21878 - 21883. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A.-M. Bamberger, J. Briese, J. Gotze, I. Erdmann, H. M. Schulte, C. Wagener, and P. Nollau Stimulation of CEACAM1 expression by 12-O-tetradecanoylphorbol-13-acetate (TPA) and calcium ionophore A23187 in endometrial carcinoma cells Carcinogenesis, March 1, 2006; 27(3): 483 - 490. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Azuz-Lieberman, G. Markel, S. Mizrahi, R. Gazit, J. Hanna, H. Achdout, R. Gruda, G. Katz, T. I. Arnon, S. Battat, et al. The involvement of NK cells in ankylosing spondylitis Int. Immunol., July 1, 2005; 17(7): 837 - 845. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Stern, G. Markel, T. I. Arnon, R. Gruda, H. Wong, S. D. Gray-Owen, and O. Mandelboim Carcinoembryonic Antigen (CEA) Inhibits NK Killing via Interaction with CEA-Related Cell Adhesion Molecule 1 J. Immunol., June 1, 2005; 174(11): 6692 - 6701. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
M) as crosslinker. After 6 h, luciferase activity was analyzed and expressed as a ratio of total protein content. *, p < 0.01. B, Jurkat-T cells were transfected with NFAT-AP1 luciferase reporter and either an empty vector or vector containing the CEACAM1–3S cDNA. After 24 h, the transfected cells were stimulated with anti-CD3 (2 µg/ml) and either control IgG1
