EM structure of the ectodomain of integrin CD11b/CD18 and localization of its ligand-binding site relative to the plasma membrane

Abstract

One-half of the integrin α-subunit Propeller domains contain and extra vWFA domain (αA domain), which mediates integrin binding to extracellular physiologic ligands via its metal-ion-dependent adhesion site (MIDAS). We used electron microscopy to determine the 3D structure of the αA-containing ectodomain of the leukocyte integrin CD11b/CD18 (αMβ2) in its inactive state. A well defined density for αA was observed within a bent ectodomain conformation, while the structure of the ectodomain in complex with the Fab fragment of mAb107, which binds at the MIDAS face of CD11b and stabilizes the inactive state, further revealed that αA is restricted to a relatively small range of orientations relative to the Propeller domain. Using Fab 107 as probe in fluorescent lifetime imaging microscopy (FLIM) revealed that αA is positioned relatively far from the membrane surface in the inactive state, and a systematic orientation search revealed that the MIDAS face would be accessible to extracellular ligand in the inactive state of the full-length cellular integrin. These studies are the first to define the 3D EM structure of an αA-containing integrin ectodomain and to position the ligand-binding face of αA domain in relation to the plasma membrane, providing new insights into current models of integrin activation.

Figure 1. EM refinement of the CD11b/CD18 ectodomain.

(A) Example images from the final round of refinement. On the left is an average image generated from reference-free alignment and averaging which preceded the refinement itself. The averages have been displayed with the final map projection that they most closely resemble. Results from the final round of refinement are displayed adjacent to the reference-free average, showing from left to right of each panel the average from reference free alignment (number 1), an example raw particle from the final projection class (number 2), the average generated from the final projection class (number 3), and the projection from the final map for this class (number 4). The letters on the left indicate the class in the distribution histogram (B). The side of each box is 232Å. (B) Projection histogram showing the number of particles identified as belonging to a particular Euler angle orientation for the final round of refinement. Each circle indicates a particular class while the number of particles is indicated as a gray scale. The scale bar is indicated to the right. The letters refer to the classes shown in (A). (C) Fourier shell correlation analysis to determine the resolution of the final map. The total dataset was separated randomly into two equal groups. The particle classes for each group were independently aligned with the final map projection and averaged to generate two independent maps. The graph displays the correlation in Fourier space between the maps. The resolution of 1/(26 Å) is determined by a value of 50% correlation.

Figure 2. Surface shaded density of the final 3D map of the CD11b/CD18 ectodomain.

A, B) are two views of the map rotated by 45°. The isosurface has been chosen to enclose 100% of the expected protein volume. C and D) the same views of the map as in A and B, shown as a transparent isosurface, but also displaying the αA-lacking αVβ3 ectodomain (pdb 3IJE) model as a ribbon diagram. A clear density corresponding to the CD11bA domain is seen, fitted here with a ribbon diagram of the crystal structure of the isolated CD11bA (pdb 1jlm), with the MIDAS ion in cyan. The αV subunit and CD11bA are shown in blue and β3-subunit in red. The spherical metal ions in the Propeller and α-genu are in green and that in the βA ADMIDAS is in magenta. E, F) a ribbon diagram of the crystal structure of CD11c/CD18 ectodomain fitted into the CD11b/CD18 EM map (shown as a transparent isosurface). Most of Calf-1/Calf-2 and β TD domains do not fit the 3D map. CD11cA fits better within the map density but extends somewhat outside it (compare CD11cA with that of CD11bA [shown in yellow, oriented as in C, with its MIDAS metal ion in orange]). The three metal ions in the CD11c Propeller are in green, the CD11cA MIDAS ion in cyan, and the βA ADMIDAS ion in magenta. F) a 45° clockwise rotation of the Figure in (E).

Figure 3. EM analysis of the CD11b/CD18-Fab107 complex and orientation of the CD11bA MIDAS face.

(A) Reference-free alignment and averaging of CD11b/CD18-Fab107 particles. The top row shows a k-means classification of 10,661 raw particles classified into 10 groups and averaged. The number of particles for each class is indicated at the top of each average. Several of the groups are similar, suggesting that there are fewer than 10 unique preferred orientations for the particles on the grid. The bottom row shows the model projection that best fits each of the particles shown in the top row. Projections in the bottom row were generated from the model shown in (B) which was generated by sequentially fitting a projection for the integrin followed by the projection for the CD11bA/Fab107 complex. For comparison, the model generated by fitting the CD11c/CD18 ectodomain X-ray structure is shown in (C). In this model, the CD11bA domain has the same relative orientation to the rest of the integrin as found in the X-ray structure. (D) Shows a direct comparison of the two models for the average, which most easily allows identification of individual integrin domains within the projection. Displayed alongside the average (center image) is the projection from the model in (B) which best fits the average (left image), compared to the projection from the model in (C) (right image). E, ribbon diagram of the Propeller/αA of CD11c/CD18 ectodomain crystal structure oriented as in Figure 2E, showing crystal contacts from two symmetry-related molecules (Calf-2 in gray and Thigh in magenta) that rotate MIDAS face of CD11cA domain by ∼35° (red arrows) relative to its position in the unrestricted EM model (colored as in Figure 2E). F, same Figure in (E) rotated clockwise by 135°.

Figure 4. FLIM lifetimes of the Alexa488 donor fluorescence in the presence and absence of the acceptor FM.

A) Representative Alexa488 fluorescence intensity and lifetime images. The panel shows examples of data collected in the absence (left, WT 488) and presence of the acceptor FM (right, WT 488+FM). The pseudocolor scale is shown at the bottom. B, histogram of measured Alexa488 lifetimes generated by integrating lifetime measurements for individual cells. The histogram bars show the number of cells in 50 ps bins for Alexa488 in the absence (blue) and presence (red) of the acceptor FM. The average and standard deviation were generated by fitting the histogram to a single Gaussian function (shown as a solid line).

Figure 5. Modeling of the relative membrane orientation of CD11b/CD18 by combining the FLIM and EM data.

A) A display of the allowed orientations of the integrin ectodomain. To determine the allowed orientations, the protein is first rotated in the membrane plane and subsequently rotated down. For each rotation pair, the model is evaluated for clashes with the membrane. Shown in magenta are the centroids for Fab107 for all of the resulting angles. For illustration purposes, the integrin α- (blue), β- (red) subunits and Fab107 (green) chains are displayed as a ribbon diagram for one of the resulting orientations. A section of a model DMPC membrane is shown as a wire diagram. The model of the CD11b/CD18 ectodomain is taken from the fit to the EM map. The distance of the centroid for the displayed model is compatible with the measured FLIM distance. B) The same view as in (A) but rotated in the y-axis by 45°. The α/β TM domains (modeled after the NMR structure of αIIbβ3 TM domains [46]) are displayed for illustrative purposes only, and were not used in the orientation search.