Identification and characterization of a human monoclonal antagonistic antibody AL-57 that preferentially binds the high-affinity form of lymphocyte function-associated antigen-1

Abstract

LFA-1 (alpha(L)beta(2)) mediates cell-cell and cell-extracellular matrix adhesions essential for immune and inflammatory responses. One critical mechanism regulating LFA-1 activity is the conformational change of the ligand-binding alpha(L) I domain from low-affinity (LA), closed form, to the high-affinity (HA), open form. Most known integrin antagonists bind both forms. Antagonists specific for the HA alpha(L) I domain have not been described. Here, we report the identification and characterization of a human antibody AL-57, which binds to the alpha(L) I domain in a HA but not LA conformation. AL-57 was discovered by selection from a human Fab-displaying library using a locked-open HA I domain as target. AL-57 Fab-phage bound HA I domain-expressing K562 cells (HA cells) in a Mg(2+)-dependent manner. AL-57 IgG also bound HA cells and PBMCs, activated by Mg(2+)/EGTA, PMA, or DTT. The binding profile of AL-57 IgG on PBMCs was the same as that of ICAM-1, the main ligand of LFA-1. In contrast, an anti-alpha(L) murine mAb MHM24 did not distinguish between the HA and LA forms. Moreover, AL-57 IgG blocked ICAM-1 binding to HA cells with a potency greater than MHM24. It also inhibited ICAM-1 binding to PBMCs, blocked adhesion of HA cells to keratinocytes, and inhibited PHA-induced lymphocyte proliferation with potencies comparable with MHM24. These results indicate that specifically targeting the HA I domain is sufficient to inhibit LFA-1-mediated, adhesive functions. AL-57 represents a therapeutic candidate for treatment of inflammatory and autoimmune diseases.

Fig. 1

AL-57 Fab-phage specifically binds to HA cells in a Mg²⁺-dependent manner. Whole cell ELISAs were carried out with unfixed HA and LA cells in the presence or absence of Mg²⁺. ELISA signals of AL-57 phage, a negative control phage, and no phage control are shown. Absorbance at 450 nm (A450 nm) of each sample condition is plotted in the histogram.

Fig. 2

AL-57 IgG1 specifically binds to HA cells but not LA cells. Binding of AL-57 IgG1 and MHM24 to HA and LA cells was determined by flow cytometric analysis. (A) Representative histograms of samples in the absence or presence of AL-57 IgG1 or MHM24 at 1.7 nM. The x-axis depicts the relative fluorescence intensity of individual cells, and the y-axis represents the cell number. Neg. Control indicates negative control with just the secondary PE-labeled antibody used for staining. The numbers shown are relative values of mean fluorescence intensity (MFI) with relative percentages of positive cells in parenthesis. (B) Concentration-dependent binding of AL-57 IgG1, hIgG1 control, and MHM24 to HA and LA cells. Cells were incubated with serially diluted AL-57 IgG1, hIgG1 control, or MHM24, followed by staining with secondary anti-hIgG- or anti-mouse IgG-PE. Plotted in each diagram is MFI as a function of the IgG concentration.

Fig. 3

AL-57 IgG4 and ICAM-1 bind to the activated HA form of LFA-1 on PBMCs. Binding of AL-57 IgG4 (A) and ICAM-1 (B) to PBMCs with inactivating buffer (CaCl₂/MgCl₂) or with activating buffer (MgCl₂/EGTA) in the presence or absence of PMA (10 ng/ml) or DTT (500 μM) was determined by flow cytometric analysis as described in Materials and Methods. The concentration of each IgG or ICAM-1 used was 10 μg/ml. For AL-57 staining, Neg. Control indicates negative control with just the secondary PE-labeled antibody used for staining; MHM24 was used as a control that did not distinguish between the HA and LA forms of LFA-1. For ICAM-1 staining, a multimeric complex [ICAM-1/streptavidin (SV)-PE] of biotinylated human ICAM-1-Fc and PE-labeled streptavidin was used; SV-PE only indicates negative control with just the PE-labeled streptavidin used for staining. Shown here are representative histograms of samples under indicated conditions. The x-axis depicts the fluorescence intensity of individual cells, and the y-axis represents the cell number. The numbers shown are relative values of MFI with relative percentages of positive cells in parenthesis.

Fig. 4

PMA-induced enhancement in AL-57 and ICAM-1 binding to PBMCs is blocked by PKC inhibitors. Binding of AL-57 IgG4 (A) and ICAM-1 (B) to PBMCs with activating buffer (MgCl₂/EGTA) in the presence or absence of PMA (10 ng/ml) and PKC inhibitor GF109203X (10 μM) or staurosporine (1 μM) was determined by flow cytometric analysis. The concentration of AL-57 IgG4 or ICAM-1 used was 10 μg/ml. For AL-57 IgG4 staining, Neg. Control indicates negative control with just the secondary PE-labeled antibody used for staining; MHM24, a nonselective binder, was also used as a control. For ICAM-1 staining, a multimeric complex (ICAM-1/SV-PE) was used; SV-PE only indicates negative control with just the PE-labeled streptavidin used for staining. Shown here are representative histograms of samples under indicated conditions. The x-axis depicts the fluorescence intensity of individual cells, and the y-axis represents the cell number. The numbers shown are relative values of MFI with relative percentages of positive cells in parenthesis.

Fig. 5

AL-57 IgG1 inhibits ICAM-1 binding to HA cells. ICAM-1 binding in the presence or absence of each IgG was measured by staining HA cells with a soluble, multimeric ICAM-1 complex with ICAM-1-Fc concentration at 2 μg/ml and flow cytometric analysis. (A) Representative histograms of samples in the absence or presence of AL-57 IgG1, MHM24, or hIgG1 control at 4 μg/ml (26.8 nM). The x-axis depicts the fluorescence intensity of individual cells, and the y-axis represents the cell number. SV-PE only indicates negative control with PE-labeled streptavidin alone for staining. The numbers shown are relative values of MFI with relative percentages of positive cells in parenthesis. (B) IC₅₀ determination of AL-57 IgG1 and MHM24. The percentage of positive cells from each sample was plotted as a function of the IgG concentration. From the representative plots shown here, IC₅₀ values were calculated to be 0.3 nM for AL-57 IgG1 and 2 nM for MHM24 using SigmaPlot 8.0 software. (C) Summary of IC₅₀ values from three independent experiments.

Fig. 6

AL-57 IgG1 inhibits HA cell adhesion to keratinocytes. HA cells were labeled with calcein AM, preincubated with AL-57 IgG1, MHM24, or hIgG1 control, and added to a keratinocyte monolayer. After washing to remove unattached cells, remaining cells were photographed using fluorescence microscopy, and total fluorescence was measured using a fluorescence plate reader. (A) Representative pictures from samples treated without or with AL-57 IgG1, MHM24, or hIgG1 control at concentrations of 6.7 or 67 nM. (B) IC₅₀ determination of AL-57 IgG1 and MHM24. Relative fluorescence units (RFU) from each sample were plotted as a function of the IgG concentration. From the representative plots shown here, IC₅₀ values were calculated to be 1.2 nM for AL-57 IgG1 and 0.7 nM for MHM24 using SigmaPlot 8.0 software. (C) Summary of IC₅₀ values from four independent experiments.

Fig. 7

AL-57 IgG1 inhibits PHA-induced cell proliferation. PBMCs were treated with PHA at 1 μg/ml in the presence of Mg²⁺, serially diluted IgG for 3 days, and then analyzed for proliferation using a BrdU chemiluminescence assay. (A) IC₅₀ determination of AL-57 IgG1 and MHM24. Relative luminescence units per second (rlu/s) from each sample were plotted as a function of the IgG concentration. From the representative plots shown here, IC₅₀ values were calculated to be 0.2 nM for AL-57 IgG1 and 0.8 nM for MHM24, using SigmaPlot 8.0 software. (B) Summary of IC₅₀ values from three independent experiments.