|
|
||||||||
Department of Immunobiology, DNAX Research Institute of Molecular and Cellular Biology, Palo Alto, CA 94304
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
One of the most interesting characteristics of adhesion molecules is their dynamic role in adhesion and detachment. Lymphocytes rapidly interconvert between a nonadherent state in circulation and an adherent state in tissues. Although the regulation of adhesion is not well understood, lymphocyte activation and structural features of the cytoplasmic domain of adhesion molecules play important roles in triggering and/or facilitating intercellular binding. For example, interaction of LFA-1 with its ligands, the ICAMs, is augmented by cell activation initiated by signals from TCR (5, 10). Upon stimulation, LFA-1 avidity for ICAM-1 is transiently increased over a period of minutes. Structural studies have determined that the three contiguous threonines within the cytoplasmic domain of the ß subunit of LFA-1 are critical for binding to ICAM-1 (11, 12).
We have recently described an adhesion molecule, DNAX accessory molecule-1 (DNAM-1),3 which is expressed on the majority of T cells and NK cells (13). DNAM-1 is a type 1 transmembrane glycoprotein and a member of the Ig supergene family. Cross-linking DNAM-1 with anti-DNAM-1 mAb induces cytolysis mediated by CTL and NK cells and also results in tyrosine phosphorylation of the DNAM-1 molecule, indicating that DNAM-1 transduces an activation signal. COS-7 cells transfected with DNAM-1 bind to a colon carcinoma cell line, Colo-205, and this binding is specifically blocked by anti-DNAM-1 mAb, suggesting that Colo-205 expresses a cell surface ligand (DNAM-1L). We have now investigated the structural features of the DNAM-1 receptor that are involved in adhesion and signal transduction. We report here that PKC phosphorylates Ser329 of DNAM-1 and plays a critical role for both DNAM-1 adhesion and signaling.
| Materials and Methods |
|---|
|
|
|---|
P815 and BW5147 are mouse mastocytoma and thymoma cell lines, respectively. Colo-205 is a human colon carcinoma cell line. PMA, paraformaldehyde, cyclohexamide, and actinomycin D were purchased from Sigma (St. Louis, MO). PKC inhibitor GF109203X was purchased from Calbiochem (San Diego, CA). PMA, GF109203X, and actinomycin D were dissolved in DMSO (which did not exceed 0.1% in the assay medium). A 1% paraformaldehyde solution was prepared in PBS (pH 7.4).
NK clones
NK clones were established from peripheral blood of healthy donors using the culture conditions described previously (14) with 100 IU/ml rIL-2 as a growth factor.
Cytotoxicity assay
Tumor cell lines (
1.5 x 106 cells) were
labeled with 100 µCi 51Cr for 2 h, washed, and used
as targets in a 4-h radioisotope release cytotoxicity assay, as
described (15). Data are expressed as the means of
triplicate cultures, and spontaneous radioisotope release was typically
<10% of total 51Cr release. Percent cytolysis was
calculated as:
![]() |
mAbs and flow cytometry
Control IgG was generously provided by Becton Dickinson Immunocytometry Systems (San Jose, CA). The anti-monomorphic HLA class I mAb (DX17 mAb) (16) and anti-DNAM-1 mAb (DX11 mAb) (13) were generated as described. Methods of immunofluorescent staining and flow cytometry have been described previously (17).
Establishment of transfectants of BW5147, which stably express DNAM-1
Wild-type and several deletion mutant DNAM-1 cDNA were produced by PCR using the DNAM-1 cDNA plasmid LL378 (13) as a template. To generate a site-specific DNAM-1 mutant at residue 329, an antisense PCR primer, which contained a codon for Phe329 (TTT) instead of Ser329 (TCT), was designed. The PCR products were subcloned into a retroviral vector pMX-neo (kindly provided by Dr. Toshio Kitamura) with cloning sites of BamHI (5') and Not1 (3'). BOSC23 packaging cells (18) were transfected with the DNAM-1 cDNA in the retroviral vector using lipofectamine (Life Technologies, Gaithersburg, MD) as described previously (19, 20). BW5147 cells were infected with the DNAM-1 retrovirus stock. Two days after infection, the cells were analyzed and DNAM-1-expressing cells were cloned by flow cytometry. All mutant cDNAs were verified by sequencing.
Cell adhesion assay
The BW5147 transfectants were incubated in PBS containing 5(6)-carboxyfluorescein-N-hydroxysuccinimide ester (Boehringer Mannheim Biochemica, Mannheim, Germany) at the concentration of 0.4 µg/ml for 1 h at room temperature and then washed with PBS three times, so that BW5147 transfectants fluoresced green. Colo-205 cells were stained with phycoerythrin (PE)-conjugated anti-human HLA class I (DX17 mAb) and then washed with PBS three times, so that Colo-205 cells fluoresce red. A total of 5 x 105 BW5147 transfectants and 1 x 105 Colo-205 cells were suspended in 200 µl of RPMI 1640 medium with 10% FCS containing or lacking EGTA and/or EDTA (each concentration at 0.5 mM), or in 200 µl HEPES buffer (0.13 M NaCl, 10 mM HEPES, 10 mM glucose, and 0.5% BSA) containing or lacking both calcium and magnesium (each concentration at 1.0 mM), and mixed in a 12- x 75-mm polystyrene tube (Becton Dickinson Labware, Franklin Lakes, NJ). The cell mixture was centrifuged at 500 rpm (40 x g) for 3 min and incubated at 37°C, room temperature, or 4°C. The cells were then gently resuspended and analyzed by flow cytometry.
In some experiments, green fluorochrome-labeled BW5147 transfectants
were pretreated with DMSO alone (0.1%), cyclohexamide (10 µg/ml),
actinomycin D (10 µg/ml), PMA (50 ng/ml), or GF109203X at the
indicated concentrations for 2 h at 37°C. In experiments to
determine the metabolic requirements for DNAM-1 binding, BW5147
transfectants or Colo-205 cells were fixed with 1% paraformaldehyde
PBS for 1 h at 4°C and washed extensively with PBS
(21). The transfectants were cocultured with Colo-205
(stained with PE-conjugated anti-HLA class I mAb) and analyzed by
flow cytometry, as described above. The percentage of Colo-205 cells
binding to the BW5147 transfectants was calculated as follows:
![]() |
![]() |
Biochemistry
Cells were labeled with [32P]orthophosphate (Amersham Corp., Arlington Heights, IL) and stimulated or not with PMA (50 ng/ml, 2 h at 37°C). Cells were then lysed in Tris-buffered saline (50 mM Tris, 15 mM NaCl, pH 8.0) containing 1% Nonidet P-40 and protease inhibitors (1 mM PMSF and 20 Kallikrein inhibitor U/ml aprotinin) and phosphatase inhibitors (1 mM EGTA, 10 mM NaF, 1 mM Na4P2O7, 0.1 mM ß-glycerophosphate, and 1 mM Na3VO4). DNAM-1 Ag was immunoprecipitated using the method described previously (22). Samples were analyzed by SDS-PAGE and radioisotopes detected using a PhosphorImager (Molecular Dynamics, Sunnyvale, CA).
| Results |
|---|
|
|
|---|
In order to study the regulation of DNAM-1 adhesion, we
established a transfectant of the mouse T cell line BW5147 which stably
expresses human DNAM-1 (Fig. 1
A). Using this transfectant
and the colon carcinoma Colo-205, which expresses a DNAM-1L
(13), we developed a cell-cell adhesion assay to determine
the requirements for DNAM-1 binding. As shown in Figure 1
B,
54% of Colo-205 cells bound to the DNAM-1 BW5147 transfectant after
incubation for 60 min at 37°C, whereas <10% of Colo-205 cells bound
to untransfected BW5147 cells. The binding of Colo-205 to
DNAM-1+ BW5147 cells was specifically inhibited when the
transfectants were preincubated with DX11 mAb (Fig. 1
B),
demonstrating specificity of the interaction. Equivalent binding was
observed in medium containing or lacking both calcium and magnesium
(each concentration at 1.0 mM). Additionally, DNAM-1 binding was not
affected when EDTA or EGTA were present in the medium (not shown).
These results indicate that DNAM-1L ligand binding does not require
divalent cations. When Colo-205 and the DNAM-1 transfectant were
cocultured at room temperature or 4°C, the binding was significantly
less than at 37°C (Fig. 1
C). These results suggest that
efficient DNAM-1 adhesion to DNAM-1L may require a metabolic change of
either DNAM-1, DNAM-1L, or both molecules. To determine which molecule
was affected, Colo-205 and/or the DNAM-1 transfectant were fixed with
paraformaldehyde and then cocultured at 37°C. Colo-205 binding to
DNAM-1 transfectants was significantly inhibited when the DNAM-1
transfectants, but not Colo-205 cells, were fixed with paraformaldehyde
(Fig. 1
C), suggesting that the DNAM-1 receptor is either
sensitive to fixation or may undergo a conformational alteration to
permit ligand binding. Binding was not affected by pretreatment of
DNAM-1 transfectants with either cyclohexamide (inhibitor of protein
synthesis) or actinomycin D (inhibitor of transcription) (data not
shown).
|
We have previously reported that anti-DNAM-1 mAb DX11
activated the cytolytic activity of CTL when the CTL were cocultured
with the murine FcR-bearing target P815 mastocytoma (13).
DX11 mAb, as well as anti-CD16 (Fc
RIII), induced redirected
cytolysis by NK clones against P815. Since NK cell-mediated
cytotoxicity, but not Ab-dependent cellular cytotoxicity mediated by
CD16, is dependent on PKC (23, 24), we examined whether
DNAM-1-mediated cytolytic activation of NK cells requires the PKC
pathway. As demonstrated in Figure 2
,
DX11 mAb-induced redirected cytolysis mediated by NK clones was
inhibited in a dose-dependent manner by the pretreatment of NK cells
with the PKC inhibitor GF109203X. In contrast, consistent with prior
findings (23, 24), the PKC inhibitor did not affect
anti-CD16-induced redirected cytolysis. These results indicate that
the PKC pathway is involved in DNAM-1-mediated cytolytic activation of
NK cells.
|
Since PKC activation plays an important role in DNAM-1-mediated
cytotoxicity, we examined the effect of PKC inhibitor GF109203X on
DNAM-1 adhesion. As demonstrated in Figure 3
, A and B,
GF109203X specifically inhibited in a dose-dependent manner the binding
of Colo-205 to the DNAM-1 transfectants. Moreover, pretreatment of the
DNAM-1 transfectants with PMA accelerated the binding of the
transfectants to Colo-205 (Fig. 3
B). Expression of DNAM-1 on
the cell surface of the transfectants was not affected by either PMA or
GF109203X, as determined by flow cytometry (not shown). These results
suggest that PKC is involved in DNAM-1 adhesion.
|
DNAM-1 contains two putative PKC-binding sites at residues 293 and
329 in the cytoplasmic domain (13). To determine whether
elements in the cytoplasmic domain are necessary for DNAM-1 binding,
BW5147 transfectants stably expressing several DNAM-1 truncation
mutants were established (designated
287,
298,
308, and
327) (Fig. 4
, A and
B). Although the amount of DNAM-1 glycoprotein expressed on
the cell surface of these transfectants was comparable with the
wild-type DNAM-1 transfectant (Figs. 1
A and 4B),
Colo-205 did not bind to any of these transfectants (Fig. 4
C). Furthermore, pretreatment of these transfectants with
PMA did not augment binding (Fig. 4
C), suggesting that the
amino acids responsible for PMA-induced DNAM-1 adhesion might be
located in the cytoplasmic domain (Fig. 4
A). Because the
DNAM-1 mutant containing a stop codon at residue 327 failed to bind, we
examined whether the putative PKC phosphorylation site at
Ser329 might be required for adhesion. Therefore, we
created a BW transfectant expressing a mutant DNAM-1 molecule with a
site-directed mutation at residue 329, converting Ser329 to
Phe329 (designated S-F329) (Fig. 4
, A and B). The Phe329 DNAM-1
transfectant was unable to bind Colo-205, even after the pretreatment
with PMA (Fig. 4
C). These results suggested that
Ser329 might be the residue that serves as a substrate for
PKC and is responsible for DNAM-1 adhesion. To test this
hypothesis, the wild-type and mutant DNAM-1 BW transfectants
were labeled with [32P]orthophosphate, stimulated with
PMA, and DNAM-1 proteins were immunoprecipitated with DX11 mAb. As
shown in Figure 5
, the wild-type DNAM-1,
but neither the deletion mutant (
327) nor Phe329-DNAM-1
mutant (S-F329), was phosphorylated after stimulation with
PMA. Thus, Ser329 appears to be the site of serine
phosphorylation by activated PKC and this event is critical for DNAM-1
adhesion.
|
|
| Discussion |
|---|
|
|
|---|
The consensus sequences for PKC phosphorylation sites are S or T
residue flanked on one or both sides by a basic amino acid, K or R,
sometimes separated by one or two "spacer" amino acids
(27). DNAM-1 contains two putative PKC phosphorylation
sites, T(Q)K and SRR at residues 293 and 329, respectively, in the
cytoplasmic domain (Fig. 4
A). We have demonstrated that
mutation of the phosphorylation site at residue 329 from S to F
completely inhibits DNAM-1 binding and the phosphorylation of DNAM-1
induced by PMA. These results indicate that S329
phosphorylation by PKC is required for DNAM-1 adhesion.
In many respects, DNAM-1 is similar to the hemopoietic cell-specific
integrin LFA-1. For example, LFA-1 and DNAM-1 are expressed on NK and T
cells, and mAbs against these molecules can partially, or totally,
inhibit cell-mediated cytotoxicity (13, 28, 29). Moreover,
intercellular binding of LFA-1 or DNAM-1 to cells bearing their ligands
is temperature dependent and enhanced by PMA activation of PKC
(5). However, in contrast to DNAM-1, the S residue in the
cytoplasmic domain of the LFA-1ß subunit that is the major substrate
for PKC phosphorylation is not necessary for LFA-1 adhesion
(11), although the cytoplasmic domain of LFA-1ß, but not
, is required for binding to ICAM-1 (12). Furthermore,
the binding of LFA-1, but not DNAM-1, requires the presence of divalent
cations such as Mg2+ and Ca2+ (5).
It takes a period of hours for DNAM-1 to reach maximal adhesion,
whereas binding of LFA-1 to its ligands is rapidly accelerated
following T cell activation, with maximal binding in minutes
(30, 31, 32, 33). DNAM-1 is a monomeric glycoprotein of the Ig
superfamily (11); LFA-1 is a heterodimer composed of two
noncovalently associated subunits of the integrin family
(34). While the ligands for LFA-1, the ICAMs, are members
of the Ig superfamily (25), the cellular ligand for DNAM-1
has not yet been identified.
We have presented evidence for a critical role of PKC for both DNAM-1
signaling and adhesion. PKC is activated by signals through the TCR/CD3
complex on T cells (5). Therefore adhesion mediated by
DNAM-1, as well as LFA-1, may be accelerated following recognition of
Ag presented by MHC on target cells by the TCR/CD3 complex on T cells.
It is possible that intercellular binding between T cells and APC is
mediated first by LFA-1/ICAMs adhesion within a period of minutes and
then sequentially enhanced by DNAM-1/DNAM-1L adhesion to permit
sustained signaling. PKC is also activated by signals through several
other leukocyte surface receptors, including CD2 (5).
Since essentially all CTL and NK cells express CD2, it is possible that
DNAM-1 adhesion to its ligand on target cells is facilitated by a
signal through CD2. Also, our studies demonstrated that specific
inhibitors of PKC prevented DNAM-1-mediated signaling for cytolytic
activation of NK cells, suggesting the possibility that DNAM-1
"outside-in" signaling may activate PKC and initiate DNAM-1
adhesion. This is consistent with the observation that the transfectant
with wild-type DNAM-1 significantly binds Colo-205 without stimulation
by PMA (Fig. 4
A).
In summary, our studies have shown that PKC plays an important role for both DNAM-1-mediated signaling and adhesion. T and NK cells express various adhesion molecules including LFA-1, CD2, and DNAM-1. The role of the adhesion molecules in lymphocyte-APC interactions may be dictated by a dynamic process and influenced by signals generated through other membrane receptors. We have identified certain cell lines that specifically bind to DNAM-1 transfectants (13) and are presently pursuing these molecules. The identification and molecular characterization of DNAM-1L would contribute to understanding the structural requirements involved in DNAM-1 signaling and binding.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Address correspondence and reprint requests to Joseph H. Phillips, Department of Immunobiology, DNAX Research Institute of Molecular and Cellular Biology, 901 California Avenue, Palo Alto, CA 94304. ![]()
3 Abbreviations used in this paper: DNAM-1, DNAX accessory molecule-1; DNAM-1L, DNAM-1 ligand; PE, phycoerythrin; PKC, protein kinase C. ![]()
Received for publication November 20, 1997. Accepted for publication April 13, 1998.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
Y. T. Bryceson, H.-G. Ljunggren, and E. O. Long Minimal requirement for induction of natural cytotoxicity and intersection of activation signals by inhibitory receptors Blood, September 24, 2009; 114(13): 2657 - 2666. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Wang, K. Shibuya, Y. Yamashita, J. Shirakawa, K. Shibata, H. Kai, T. Yokosuka, T. Saito, S.-i. Honda, S. Tahara-Hanaoka, et al. LFA-1 decreases the antigen dose for T cell activation in vivo Int. Immunol., September 1, 2008; 20(9): 1119 - 1127. [Abstract] [Full Text] [PDF] |
||||
![]() |
J.-L. Jian, C.-S. Zhu, Z.-W. Xu, W.-M. Ouyang, D.-C. Ma, Y. Zhang, L.-J. Chen, A.-G. Yang, and B.-Q. Jin Identification and Characterization of the CD226 Gene Promoter J. Biol. Chem., September 29, 2006; 281(39): 28731 - 28736. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Shirakawa, Y. Wang, S. Tahara-Hanaoka, S.-i. Honda, K. Shibuya, and A. Shibuya LFA-1-dependent lipid raft recruitment of DNAM-1 (CD226) in CD4+ T cell Int. Immunol., June 1, 2006; 18(6): 951 - 957. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Tahara-Hanaoka, K. Shibuya, H. Kai, A. Miyamoto, Y. Morikawa, N. Ohkochi, S.-i. Honda, and A. Shibuya Tumor rejection by the poliovirus receptor family ligands of the DNAM-1 (CD226) receptor Blood, February 15, 2006; 107(4): 1491 - 1496. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. T. Bryceson, M. E. March, H.-G. Ljunggren, and E. O. Long Synergy among receptors on resting NK cells for the activation of natural cytotoxicity and cytokine secretion Blood, January 1, 2006; 107(1): 159 - 166. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. S. Boles, W. Barchet, T. Diacovo, M. Cella, and M. Colonna The tumor suppressor TSLC1/NECL-2 triggers NK-cell and CD8+ T-cell responses through the cell-surface receptor CRTAM Blood, August 1, 2005; 106(3): 779 - 786. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Shirakawa, K. Shibuya, and A. Shibuya Requirement of the serine at residue 329 for lipid raft recruitment of DNAM-1 (CD226) Int. Immunol., March 1, 2005; 17(3): 217 - 223. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. J. Ralston, S. L. Hird, X. Zhang, J. L. Scott, B. Jin, R. F. Thorne, M. C. Berndt, A. W. Boyd, and G. F. Burns The LFA-1-associated Molecule PTA-1 (CD226) on T Cells Forms a Dynamic Molecular Complex with Protein 4.1G and Human Discs Large J. Biol. Chem., August 6, 2004; 279(32): 33816 - 33828. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Tahara-Hanaoka, K. Shibuya, Y. Onoda, H. Zhang, S. Yamazaki, A. Miyamoto, S.-i. Honda, L. L. Lanier, and A. Shibuya Functional characterization of DNAM-1 (CD226) interaction with its ligands PVR (CD155) and nectin-2 (PRR-2/CD112) Int. Immunol., April 1, 2004; 16(4): 533 - 538. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Fuchs, M. Cella, E. Giurisato, A. S. Shaw, and M. Colonna Cutting Edge: CD96 (Tactile) Promotes NK Cell-Target Cell Adhesion by Interacting with the Poliovirus Receptor (CD155) J. Immunol., April 1, 2004; 172(7): 3994 - 3998. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Shibuya, J. Shirakawa, T. Kameyama, S.-i. Honda, S. Tahara-Hanaoka, A. Miyamoto, M. Onodera, T. Sumida, H. Nakauchi, H. Miyoshi, et al. CD226 (DNAM-1) Is Involved in Lymphocyte Function-associated Antigen 1 Costimulatory Signal for Naive T Cell Differentiation and Proliferation J. Exp. Med., December 15, 2003; 198(12): 1829 - 1839. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Kojima, H. Kanada, S. Shimizu, E. Kasama, K. Shibuya, H. Nakauchi, T. Nagasawa, and A. Shibuya CD226 Mediates Platelet and Megakaryocytic Cell Adhesion to Vascular Endothelial Cells J. Biol. Chem., September 19, 2003; 278(38): 36748 - 36753. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |