ITIM-containing receptors play an essential role in modulating immune responses. Leukocyte-associated inhibitory receptor (LAIR)-1, also known as CD305, is an ITIM-containing inhibitory receptor, expressed by all leukocytes, that binds collagens. In this article, we investigate the effect of a conservative R65K mutation on LAIR-1 ligand binding and function. Compared with LAIR-1 wild-type (wt)-expressing cells, LAIR-1 R65K cells show markedly reduced binding to collagen, which correlates with a reduced level of LAIR-1 polarization to the site of interaction with collagens. Both LAIR-1 wt and R65K cells can generate intracellular signals when ligated by anti-LAIR-1 mAb, but only LAIR-1 wt cells respond to collagens or matrigel. In agreement, surface plasmon resonance analyses showed that LAIR-1 R65K protein has markedly reduced avidity for collagen type I compared with LAIR-1 wt. Likewise, LAIR-1 R65K protein has decreased avidity for cells expressing transmembrane collagen XVII. Thus, a single residue, Arg65, is critical for the interaction of LAIR-1 with collagens.
An adequate immune response is the result of a fine balance between activating and inhibitory signals that promotes elimination of the invading agent while suppressing hyperresponsiveness that could damage the host. Many mechanisms exist to accomplish this task, including the expression of both activating and inhibitory receptors by immune cells. A group of inhibitory receptors are characterized by the presence of one or more ITIM in their cytoplasmic tail. After interaction with their ligands, the tyrosine residues in the ITIM are phosphorylated by Src family tyrosine kinases. These phosphorylated tyrosine residues serve as docking sites for phosphatases, such as Src homology 2 domain-containing protein tyrosine phosphatase (SHP)4-1, -2, and SHIP, which then become activated and initiate the propagation of the inhibitory signal (1, 2). The leukocyte associated inhibitory receptor (LAIR)-1, also known as CD305, possesses two ITIM in its cytoplasmic tail that mediate its inhibitory capacity through interaction with SHP-1, SHP-2, and/or C-terminal Src kinase (3, 4, 5). This receptor is expressed by all leukocytes, NK cells, T cells, B cells, dendritic cells, monocytes, neutrophils, and hematopoietic stem cells (3, 6, 7, 8, 9). The ligands for LAIR-1, and for the highly homologous, secreted protein LAIR-2, are collagens, the most abundant type of protein in the body (10, 11). The interaction between LAIR-1 and collagens is of high affinity and is dependent on the conserved glycine-proline-hydroxyproline (GPO) repeating sequence that is characteristic of all collagen molecules (10, 11). The broad expression pattern of this inhibitory receptor and the possibility that immune cells are able to interact with collagens at many places during their trafficking through the body suggest that LAIR-1 may be an important receptor for modulating immune responses. Multiple studies have shown that the ligation of LAIR-1 by mAb or collagens is able to inhibit activation signals (3, 6, 7, 8, 10, 12, 13).
Although the function of LAIR-1 in vivo is currently unknown, it is known that the engagement of LAIR-1 with specific mAb is able to inhibit proliferation of human myeloid leukemic cell lines by inducing programmed cell death independent of Fas/Fas ligand interaction (14). Also, the engagement of LAIR-1 expressed on acute myeloid leukemia blasts, isolated from peripheral blood or bone marrow from patients, inhibits GM-CSF-induced proliferation leading to apoptosis, possibly by a mechanism involving inhibition of GM-CSF-induced AKT activation (15). Moreover, there is a correlation in patients with high-risk B cell chronic lymphocytic leukemia with the absence of LAIR-1, suggesting that the absence of this receptor may be involved in the proliferation of leukemic cells (16). During the course of HIV infection, LAIR-1 expression is abnormally expressed both in naive B cells (17) and in a unique tissue like memory B cell subset that is expanded in HIV-infected patients (18). Finally, the levels of LAIR-2 have been shown to be elevated in the synovial fluid of patients with rheumatoid arthritis (11). Altogether, these data suggest that LAIR-1, and also LAIR-2, may be involved in the pathogenesis and natural history of several diseases.
As stated previously, LAIR-1 binds collagens with high avidity; however, nothing is known about the structural basis of this interaction. In this study, we show that a single residue in LAIR-1, Arg65, is critical for the functional interaction of this receptor with several types of collagens.
Materials and Methods
In a manner analogous to previous studies (19), we generated human LAIR-1/CD3ζ wild-type (wt) and LAIR-1/CD3ζ R65K constructs. Briefly, RNA was isolated from human T cells using an RNAqueous-4PCR kit (Ambion). cDNA was obtained by an iScript cDNA Synthesis kit (Bio-Rad). cDNAs corresponding to human LAIR-1 extracellular domain and CD3ζ transmembrane and cytoplasmic domains were then amplified using the following primers: LEC-F, GAATTCGCCGCCACCATGTCTCCCCACCCCACCG; LEC-R, GGATCCTCAGCTTTCAGGCCTTGGGAAG; CD3ZTM-F, GGATCCCAAACTCTGCTACCTG; and CD3ZTM-R, TCTAGAGCGTCTGCAGGTCTGGCC. The PCR products were digested and cloned into the pcDNA3.1(+) plasmid (Invitrogen). The same strategy was followed to make the KIR2DL2/CD3ζ construct, except for using NK cells as a source of mRNA. The LAIR-1/Fc wt construct was a gift from Dr. L. Meyaard (University Medical Center, Utrecht, The Netherlands). The LAIR-1-EGFP wt construct was made by cloning LAIR-1 in the pEGFP-N3 plasmid vector (BD Clontech). The LAIR-1/Fc R65K and LAIR-1-EGFP R65K mutants were made by using a QuikChange Site-Directed Mutagenesis kit (Stratagene). All constructs were sequenced to confirm their identities.
Cells and reagents
The LAIR-1/Fc wt and LAIR-1/Fc R65K fusion proteins were isolated from the culture supernatants of transiently transfected 293T cells using protein A-Sepharose Fast Flow columns (Invitrogen). K562 cells were transiently transfected with LAIR-1-EGFP and LAIR-1-EGFP R65K constructs using the Amaxa nucleofection system (Amaxa). K562 cells were stably transfected by electroporation with the LAIR-1 wt or LAIR-1 R65K constructs, or a collagen XVII construct provided by Dr. H. Notbohm (University of Lübeck, Lübeck, Germany). Detection of transmembrane collagen XVII on transfected K562 cells was done by flow cytometry with the NC16a-3 Ab provided by Dr. L. Bruckner-Tuderman (University of Freiburg, Freiburg, Germany).
Reporter cell assay and binding assays
The reporter cell line BWZ.36 (20) was stably transfected by electroporation with the LAIR-1/CD3ζ wt or LAIR-1/CD3ζ R65K constructs. This strategy provides a facile means for detecting receptor ligation. In this system, if LAIR-1/CD3ζ expressed by the BWZ.36 cells interacts with a ligand, the ITAM in the CD3ζ become phosphorylated and, as a consequence, cytoplasmic NFAT proteins translocate to the nucleus and bind tandem NFAT binding sites in the promoter of the LacZ gene. This induces β-galactosidase expression, whose level of induction can be quantitated through its enzyme activity. Specifically, BWZ.36 reporter cells (1 × 105) were added to 24-well plates coated with 5 μg of anti-LAIR-1 mAb clone DX26 (BD Pharmingen), anti-KIR2DL2 mAb (Beckman Coulter), or isotype control Ig (Beckman Coulter); or to 20–50 μg of matrigel (BD Biosciences), collagen type I, collagen type IV, or laminin 24-well biocoat plates from BD Biosciences. After 24–48 h of incubation, the cells were harvested, and LacZ activity was measured by β-galactosidase assay kits (CPRG; Gene Therapy Systems). For cell binding assays, LAIR-1 wt or LAIR-1 R65K-transfected K562 cells were labeled with 2 μg/ml Cell Tracker Green CMFDA (Invitrogen) in RPMI 1640 containing 1% FCS for 30 min at 37°C. Then, cells were washed twice with PBS containing 5% FCS and resuspended at 1.5 × 106/ml. One hundred microliters of cells were added to 96-well plates (in triplicates), uncoated (for the input), or coated with collagen type I. After 2–3 h of incubation at 37°C, plates were flicked and washed four times with PBS containing 5% FCS, and fluorescence was read with a plate reader. The binding percentage was calculated according to the following formula: (problem fluorescence − medium fluorescence/input fluorescence − medium fluorescence) × 100. The fusion protein binding assay was performed as previously described (21), except that the binding of the fusion protein was detected by FITC-conjugated mouse anti-human Fc (Jackson ImmunoResearch Laboratories). For conjugate formation, 1 × 106/ml untransfected K562 cells, LAIR-1 wt, and LAIR-1 R65K transfected K562 cells were added to 24-well plates along with 100 μl of a solution that contained 1200 μl of RPMI 1640 medium with 12 μl of collagen I-labeled microspheres, size 1.0 μm, yellow-green fluorescent (505/515) (Invitrogen). After 3 h of incubation at 37°C, cells are harvested and conjugate formation was analyzed in a FACSCalibur cytometer (BD Biosciences).
Western blot and immunoprecipitations analysis
LAIR-1 wt or LAIR-1 R65K-transfected K562 cells were treated with collagen type I at 1 μg/107cells/ml in medium without serum at 37°C for 20 min. For pervanadate treatment, cells were incubated for 20 min with freshly prepared sodium pervanadate (0.1 mM sodium orthovanadate and 10 mM hydrogen peroxide from Sigma-Aldrich) in medium without serum at 37°C. This treatment inhibits tyrosine phosphatases and thereby promotes detection of phosphorylated proteins. Cells were immediately lysed in cold lysis buffer (0.5% Triton X-100, 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 2 mM EDTA, protease, and phosphatase inhibitors) (Sigma-Aldrich) for 30 min on ice. After centrifugation for 30 min at 15,000 rpm, supernatants were first precleared by mixing with protein G-Sepharose beads (Sigma-Aldrich) and then immunoprecipitated with anti-human LAIR-1 (clone DX26) prebound to protein G-Sepharose beads. After gentle rotation at 4°C for 12 h, the beads were washed five times with cold lysis buffer, and bound proteins were eluted with Laemmli buffer, resolved by SDS-PAGE on a 4–12% gradient NuPAGE polyacrylamide gel (Invitrogen), and transferred electrophoretically to nitrocellulose membranes (Invitrogen). Membranes were then blocked with 3% (w/v) BSA in TBS-T (20 mM Tris-HCl (pH 8), 150 mM NaCl, and 0.05% Tween 20) for 2 h and then probed with mouse anti-phosphotyrosine mAb (clone 4G10; Millipore). After extensive washing in TBS-T, the membranes were incubated with HRP-labeled goat anti-mouse Ig Ab (Amersham Biosciences), and immunoreactivity was visualized by using the ECL system (Amersham Biosciences). For loading controls, membranes were stripped and reprobed with mouse anti-human LAIR-1 mAb (clone 14/LAIR-1 from BD Transduction Laboratories).
K562-transfected cells were deposited on glass coverslips coated with collagen type I (BD Biosciences), poly-lysine (Sigma-Aldrich), anti-LAIR-1 mAb, or anti-KIR2DL2 mAb for 45–60 min at 37°C. The cells attached to the coverslips were then fixed with 4% paraformaldehyde for 10 min at room temperature and washed with PBS before mounting on glass slides with mounting agent containing mowiol (EMD Biosciences). Images were acquired with a Leica SP2 AOBS confocal imaging system (Leica Microsystems Heidelberg GmbH). Images of ∼0.3-μm-thick sections of cells expressing LAIR-1-EGFP, LAIR-1-EGFP R65K, or KIR2DL2-EGFP were acquired. The z stacks were analyzed for green pixel intensity. This was done using stack profile, an algorithm that is contained within the Leica LCS Lite software. The stack profile shows mean pixel intensity (MPI) from channel 1 (EGFP) plotted against the section number from the top to the bottom of the cell. To quantify the polarization of LAIR-1-EGFP wt or LAIR-1 EGFP R65K, the MPI of channel 1 (EGFP) inside a region of interest (ROI), drawn around the cell in each z-stack section was obtained and then, a ratio between MPI from the bottom section and middle sections was calculated.
Surface plasmon resonance (SPR) analysis
Binding of LAIR-1/Fc wt, LAIR-1/Fc R65K, or ICAM-1/Fc, as a negative control, to rat collagen type I was analyzed by surface plasmon resonance (SPR) using a BIAcore 2000. Collagen was covalently coupled to a CM-5 sensor chip by standard methods (22) to either 2600 or 575 resonance units, and samples were injected over the coupled surfaces in modified HBST, which contains 10 mM HEPES (pH 7.2), 125 mM NaCl, 2 mM CaCl2, and 0.005% Tween 20, without EDTA, at a flow rate of 50 μl/min. Washout was accomplished with the same buffer devoid of analyte. Data were collected at a rate of 5 s−1 and were analyzed using BIAeval 3.2 using models for simultaneous kinetics of a 1:1 Langmuir binding reaction.
Molecular graphics comparisons
Arg65 is essential for functional recognition of collagens by LAIR-1
To study the functional interaction between LAIR-1 and its ligands, we used the BWZ.36 T cell mouse hybridoma reporter cell line transfected with a chimeric receptor consisting of the human LAIR-1 extracellular portion and human CD3ζ transmembrane and intracytoplasmic portions (BWZ.36 LAIR-1/CD3ζ cells). In Fig. 1,A, upper panel, we confirm previous results showing that LAIR-1 is able to functionally interact with collagens; specifically, collagen type I, a prototypic fibril-forming collagen that is present in noncartilaginous connective tissues, and the network forming collagen type IV, which is an important component of basement membranes (26). As a specificity control, we used BWZ.36 KIR2DL2/CD3ζ cells that only are able to express β-galactosidase activity when KIR2DL2 is ligated (Fig. 1,A, lower panel). We next examined if LAIR-1 can recognize collagen in the context of other proteins that are expressed in the extracellular matrix. In Fig. 1,B, we show that LAIR-1 is able to functionally interact with matrigel, a commercial preparation of basement membranes that contains collagen type IV. As the epitope recognized by LAIR-1 is unique to collagens (26), it is unlikely that other components within matrigel, such as laminin (see Fig. 1 B), heparan sulfate, proteoglycans, or entactin, can bind to LAIR-1.
Although a simple triple helix (GPO)10 characteristic of collagens is sufficient for binding to LAIR-1 (10), it is unknown what residue(s) in LAIR-1 are important for the interaction with collagens. During the construction of LAIR-1/CD3ζ wt, we fortuitously generated the LAIR-1/CD3ζ R65K mutant construct and decided to investigate if this relatively conservative mutation significantly impacts LAIR-1 ligand recognition. We found that Arg65 is essential for the functional recognition of collagens by LAIR-1 expressed in the BWZ.36 reporter system (Fig. 1,C). As a positive control, we show that ligation of the LAIR-1/CD3ζ R65K expressed in BWZ.36 cells with anti-LAIR-1 mAb induces the same amount of β-galatosidase activity as LAIR-1/CD3ζ expressed in these cells. This indicates that LAIR-1/CD3ζ R65K is capable of signaling and that the binding sites for collagen and anti-LAIR-1 mAb to LAIR-1 are probably different. Next, we examined the effect of ligation on the phosphorylation status of LAIR-1 wt and LAIR-1 R65K. To do this experiment, we used K562 cells transfected with plasmids encoding LAIR-1 wt and LAIR-1 R65K. Fig. 2 shows that after interaction with collagen type I LAIR-1 wt is markedly phosphorylated relative to LAIR-1 R65K (Fig. 2, upper panel). As a control, we show that both LAIR-1 wt and LAIR-1 R65K are phosphorylated when cells are treated with pervanadate (Fig. 2, lower panel). Altogether, these results indicate that Arg65 has an essential role in the functional interaction of LAIR-1 with collagens.
LAIR-1 R65K shows decreased ability for binding collagen type I
To further investigate the interaction of LAIR-1 wt and LAIR-1 R65K with collagens, we used the LAIR-1-negative K562 cell line transfected with LAIR-1-EGFP wt or LAIR-1-EGFP R65K to study binding to plate-bound collagens. As shown in Fig. 3,A, there is extensive polarization of the LAIR-1-EGFP wt to the site of contact with collagen type I or anti-LAIR-1 mAb. LAIR-1-EGFP wt expressed in LAIR-1-positive Jurkat T cells showed similar polarization toward the site of contact with anti-LAIR-1 mAb and collagen type I (data not shown). As negative control, we observed a lack of polarization of LAIR-1-EGFP wt when the K562 cells are in contact with poly-l-lysine. KIR2DL2-EGFP expressed in K562 cells failed to polarize to the site of contact with collagen (Fig. 3,A) or poly-l-lysine (data not shown). These results indicate that polarization of LAIR-1 is specific to its interaction with ligand. Results in Fig. 3,B show that LAIR-1-EGFP R65K is still able to interact with collagen as shown by the polarization of the receptor toward the site of interaction with collagen type I. However, quantification of the polarization shows that LAIR-1 R65K polarizes less efficiently than LAIR-1 wt (Fig. 3,C). As expected, the extent of polarization of LAIR-1-EGFP R65K in response to ligation by anti-LAIR-1 mAb is virtually identical to that obtained with LAIR-1-EGFP wt (Fig. 3 B).
We also observed that the pattern of polarization of LAIR-1-EGFP R65K with collagen is different from that of LAIR-1-EGFP wt. As it is shown in Fig. 3 D, LAIR-1-EGFP wt accumulates with uniform distribution at the site of contact with collagen type I; on the other hand, LAIR-1-EGFP R65K accumulates in a ring-like pattern with an increased intensity in the periphery and less in the center of the contact site. Moreover, while cells expressing LAIR-1-EGFP wt tend to have straight protrusions (lamellapods) outside of the central body in the area in contact with collagen, cells that express LAIR-1- EGFP R65K have undulated protrusions. These observations indicate that, while both LAIR-1 receptors, wt and R65K, are able to polarize when interacting with collagen type I, the nature and dynamics of the polarization are quite different, probably reflecting the inability of LAIR-1 R65K to efficiently transmit intracellular signals.
To rule out the possibility that the tagging of EGFP to LAIR-1 wt and LAIR-1 R65K causes the differences observed in the polarization of the receptor toward collagen, we performed cell binding assays with K562 cells transfected with LAIR-1 wt or LAIR-1 R65K. Results presented in Fig. 4,A clearly show that the binding by the LAIR-1 R65K K562 cells is dramatically diminished compared with LAIR-1 wt K562 cells. In a similar experiment, we show that LAIR-1 wt cells form more conjugates with FITC labeled collagen type I beads than LAIR-1 R65K K562 cells (Fig. 4 B).
LAIR-1 R65K binds transmembrane collagen XVII with low efficiency
In the results presented above, the interaction of LAIR-1 wt and LAIR-1 R65K with collagens is in the context of the receptors expressed on cells. To confirm that the observed differences in collagen binding between LAIR-1 wt and LAIR-1 R65K were directly attributable to the R65K mutation and not to interactions with other molecules expressed on cell surfaces, we generated recombinant LAIR-1/Fc wt and LAIR-1/Fc R65K proteins for binding studies. First, we checked the binding of these proteins to K562 cells transfected with collagen XVII, a transmembrane form of collagen (27). Fig. 5 shows that LAIR-1/Fc R65K protein binds with much lower efficiency to K562 cells transfected with collagen XVII cells than the LAIR-1/Fc wt protein. Negative controls show that neither protein binds to untransfected K562 cells nor does an irrelevant Fc fusion protein (ICAM-1/Fc) bind to K562 cells transfected with collagen XVII.
LAIR-1 R65K binds collagen with very low avidity
To examine further the quantitative difference in the binding of LAIR-1/Fc wt and of LAIR-1/Fc R65K to collagen, we tested the interaction by SPR (Fig. 6). LAIR-1/Fc wt bound collagen with an apparent KD of 1.45 × 10−7 M (ka of 3.15 × 105 L/mol/s; kd of 0.0456 s−1) (Fig. 6,A). LAIR1/Fc R65K bound detectably, but to a much lower capacity, precluding reliable evaluation (Fig. 6,B). The profound differences in relative binding of LAIR-1/Fc wt and LAIR-1/Fc R65K to collagen are emphasized in the binding isotherms shown in Fig. 6, C and D, which depict binding to a highly coupled (Fig. 6,C) and low density coupled (Fig. 6 D) surfaces.
LAIR-1 is a major ITIM-containing receptor that is expressed on all hematopoietic cell types whose ligand is collagens. Elucidation of the mechanism of interaction of LAIR-1 with collagens is an essential step toward understanding the inhibitory signal mediated through this receptor. We show here that mutation of a single residue, Arg65, dramatically decreases the binding of LAIR-1 to collagens. More importantly, we show that Arg65 is a critical residue for the functional interaction of this receptor with its ligands.
Collagens are the most abundant protein in mammals, and they are important not only for the biomechanical properties of tissues and organs, but also they are involved in many cell functions such as cell adhesion and migration, morphogenesis, differentiation, and so on (28). The hallmark of all collagen molecules is a triple helix made up of repeating sequences of glycine-x-y repeats, where x and y, in many cases, are proline and hydroxyproline (26). There are several mammalian receptors known to bind collagens, including α1β1, α2β1, α10β1, and α11β1 integrins, discoidin domain receptor (DDR)1 and DDR2, GPVI, LAIR-1, and members of the mannose receptor family (28). Each receptor binds different types of collagens; for example, integrin α1β1 preferentially binds collagens type IV and VI, as well as fibril-forming collagens, whereas GPVI binds fibril-forming collagen and the synthetic collagen-related peptide (GPO)10 (28). The α1β1 integrin and DDR2 have been shown to bind a distinct sites within collagen type I and type II, respectively, and not (GPO)10 (28). LAIR-1 has been shown to bind fibril-forming collagens and transmembrane collagens (10). In addition, we show here that LAIR-1 also interacts with collagen type IV, including within the context of the extracellular matrix. Similar to GPVI (29), LAIR-1 binds (GPO)10, but not (GPP)10.
The structure of LAIR-1, alone or in complex with collagen, is unknown. Moreover, there are no studies that have described what residues in the LAIR-1 receptor are important for binding collagen. To gain insight into the structural location of the Arg65 in the LAIR-1 molecule, and its possible relationship to its capacity to bind collagen, we searched the protein structure database (23) for proteins showing homology to LAIR-1. Several molecules with significant amino acid sequence homology were identified, including the human monocyte-activating receptor, designated LILRA5/LIR9/ILT11 or CD85f (24) (42% identity over 96 residues) and the human platelet GPVI (25) (42% identity over 91 residues). LAIR-1 Arg65 aligns with Ser42 and Ser44 of LILRA5 and GPVI, respectively (supplemental Fig . S1A).5 Ribbon diagrams (supplemental Fig. S1, B and C)5 show the position of Ser42 and Ser44 of the mature LILRA5 and GPVI, respectively. The collagen-related peptide (GPO)n binding site of GPVI has been identified by molecular docking algorithms (25). Examination of surface representations of GPVI reveals that Ser44 lies within a very basic region (Supplemental Fig. S1D)5 and contributes to the floor of the collagen-related peptide binding site as illustrated in Supplemental Fig. S1E.5
The gene encoding GPVI is located in the leukocyte receptor cluster on human chromosome 19 along with other genes encoding receptors with one, two, or more Ig-like domains, such as KIR, LIR/ILT, as well as LAIR-1 and LAIR-2 themselves (30). LAIR-2 is a soluble receptor that binds collagen with high affinity, and it is able to antagonize the LAIR-1/collagen inhibitory interaction (11). The sequence homology of LAIR-1 and LAIR-2 is 84% (see Supplemental Fig. S1)5 and, very importantly, LAIR-2 also has an Arg in position 65, in agreement with our data showing the necessity of this residue for collagen binding. The GPVI D1 domain has a 12-residue deletion when compared with other members of the leukocyte receptor cluster. This deletion promotes the creation of a shallow hydrophobic groove, which has been shown by computational algorithms to accommodate the collagen-related peptide (GPO)n (25) (Supplemental Fig. S1, D and E).5 However, LAIR-1 lacks the 12-residue deletion (supplemental Fig. S1A),5 which raises the possibility that LAIR-1 and GPVI bind collagen differently. This is supported by the fact that LAIR-1 binds more types of collagens than GPVI does (10).
Another important question is how binding of collagen to LAIR-1 leads to intracellular signaling. It is known that multiple GPVI molecules can bind a single (GPO)n triple helix (25). This is consistent with the observation that collagen binding induces clustering of GPVI, resulting in a signaling cascade via the FcRγ chain adaptor protein (25). Although it is likely, it is unknown if more than one LAIR-1 molecule can bind to a single collagen triple helix, and it is also not known if LAIR-1 clustering is required for signaling transmission. Assuming that clustering occurs and it is involved in signal transmission, the role that Arg65 plays in this process will need to be determined. Another nonexclusive possibility is that R65K mutation abolishes the ability of collagen to induce conformational changes in the LAIR-1 molecule that result in signaling transmission.
We have shown that a very conservative substitution of Lys for Arg at position 65 is enough to dramatically affect the ability of LAIR-1 to bind several types of collagen. Even though LAIR-1 R65K can still bind collagens, SPR analysis indicates that the inherent avidity of the interaction of the mutant with collagen is considerably reduced compared with the wt molecule. The facts that LAIR-1/CD3ζ R65K expressing cells fail to transmit signals and that LAIR-1 R65K in expressing cells fails to become phosphorylated upon exposure to collagen indicate that the R65K mutation affects LAIR-1 function. Solving the crystal structure of LAIR-1 complexed with collagen, along with additional mutagenesis studies, will clarify the role of Arg65, and other residues, in the interaction of LAIR-1 with its ligands.
We thank Drs. E. O. Long and M. March for the ICAM-1/Fc fusion protein, Drs. E. O. Long and D. N. Burshtyn for the KIR2DL2-EGFP construct, Dr. N. Shastri for the BWZ.36 cells, and Dr. E. O. Long for the 293T cells. We also thank Dr. David Garboczi for critical review of the manuscript.
The authors have no financial conflict of interest.
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.
This work was supported by the intramural program of the National Institute of Allergy and Infectious Diseases.
Abbreviations used in this paper: SHP, Src homology 2 domain-containing protein tyrosine phosphatase; DDR, discoidin domain receptor; GPO, glycine-proline-hydroxyproline; GPVI, glycoprotein VI; LAIR, leukocyte-associated inhibitory receptor; MPI, mean pixel intensity; RU, resonance units; SPR, surface plasmon resonance; wt, wild type.
The online version of this article contains supplemental material.