Lateral mobility of individual integrin nanoclusters orchestrates the onset for leukocyte adhesion

Abstract

Integrins are cell membrane adhesion receptors involved in morphogenesis, immunity, tissue healing, and metastasis. A central, yet unresolved question regarding the function of integrins is how these receptors regulate both their conformation and dynamic nanoscale organization on the membrane to generate adhesion-competent microclusters upon ligand binding. Here we exploit the high spatial (nanometer) accuracy and temporal resolution of single-dye tracking to dissect the relationship between conformational state, lateral mobility, and microclustering of the integrin receptor lymphocyte function-associated antigen 1 (LFA-1) expressed on immune cells. We recently showed that in quiescent monocytes, LFA-1 preorganizes in nanoclusters proximal to nanoscale raft components. We now show that these nanoclusters are primarily mobile on the cell surface with a small (ca. 5%) subset of conformational-active LFA-1 nanoclusters preanchored to the cytoskeleton. Lateral mobility resulted crucial for the formation of microclusters upon ligand binding and for stable adhesion under shear flow. Activation of high-affinity LFA-1 by extracellular Ca²⁺ resulted in an eightfold increase on the percentage of immobile nanoclusters and cytoskeleton anchorage. Although having the ability to bind to their ligands, these active nanoclusters failed to support firm adhesion in static and low shear-flow conditions because mobility and clustering capacity were highly compromised. Altogether, our work demonstrates an intricate coupling between conformation and lateral diffusion of LFA-1 and further underscores the crucial role of mobility for the onset of LFA-1 mediated leukocyte adhesion.

Integrins are transmembrane α/β heterodimeric cell adhesion molecules that play critical roles in cell–cell and cell–extracellular matrix interactions. Most integrins dynamically modulate their adhesiveness through global conformational changes that define their affinity state for ligands (1–3), and by their membrane lateral organization in clusters, also known as avidity (4–6). Although the mechanisms by which conformational changes of integrins leading to enhanced-binding affinity have been well established (1–3, 7, 8), there is no consensus yet on how integrins control their spatial rearrangement on the cell membrane and the role of lateral mobility in cluster formation.

Integrin clustering alone can lead to increased avidity (4) due to cooperative effects on resistance to bond breakage (9). Because bonds between individual integrins and their ligands can be broken with relatively small forces (10), clustering appears as an important mechanism to reinforce adhesion via cooperative binding (10, 11). On leukocytes, rearrangement of integrins into clusters upon activation occurs by diffusion, after release of cytoskeleton constrains on integrin motion (12–14). These observations led to the common belief that inactive integrins are anchored to the cytoskeleton and released when needed to reinforce binding (12, 15, 16), suggesting that integrin activation precedes clustering. However, this model is hard to reconcile with the fact that high-affinity (extend) integrins are more prone to interact with the cytoskeleton via their cytoplasmic tails (2), leading to immobile integrins that will compromise their capacity for lateral diffusion; a process that is imperatively required for clustering (6).

A major integrin involved in the immune system is lymphocyte function-associated antigen 1 (LFA-1). LFA-1 mediates leukocyte migration across the endothelium and within tissues and the formation of the immunological synapse by binding to its major ligand intercellular adhesion molecule 1 (ICAM-1) (17–19). We recently showed that, on resting monocytes, LFA-1 function is associated with its nanoclustering (20) forming nonintermixed nanoscale supramolecular complexes together with lipid raft components (21, 22). Importantly, a subset of these nanoclusters already contained primed—i.e., extended LFA-1 molecules—prior to ligand binding (20, 23). Interestingly, a subset of high-affinity LFA-1 clusters has been also observed upon chemokine stimulation at the immediate contact between crawling T cells and endothelial cells (24). Yet, dynamic insight on the spatial organization of LFA-1 nanoclusters is lacking. Moreover, how these different subsets diffuse on the cell membrane and cooperate with each other to generate adhesion-competent microclusters remains unknown. Here we applied single-molecule approaches together with reporters of LFA-1 conformational state to dissect the relationship between conformation, lateral mobility, and ligand-induced microclustering of LFA-1.

Results

LFA-1 Nanoclusters Are Mobile on Resting Monocytes.

We previously showed that LFA-1 forms distinct nanoclusters on resting monocytes prior to ligand binding (20, 21). Although this nanoscale organization resulted crucial for LFA-1 function (20), their dynamic behavior has not been investigated yet. We applied single-dye tracking (SDT) to follow the lateral diffusion of individual sublabeled LFA-1 nanoclusters (Fig. S1 A–D and SI Text) on resting monocytes. At our sublabeling conditions, individual intensity trajectories on fixed cells showed discrete photobleaching steps consistent with single-dye detection (Fig. S1 C and D). Trajectories of LFA-1 nanoclusters labeled with the neutral mAb TS2/4 were generated from multiple movies (Fig. 1A) using a single-molecule detection sensitive epifluorescence microscope (Fig. 1 B and C). At physiological cation conditions (0.4 mM Ca²⁺, Mg²⁺), LFA-1 nanocluster mobility did not exhibit uniform behavior. Whereas some nanoclusters were highly mobile, others showed a more restricted diffusion (Fig. 1D). Individual trajectories were analyzed by generating mean-square displacement (MSD) curves to obtain the diffusion coefficient D at short-time lags. The resultant D distribution varied from 10^-3 to 10^-1 μm²/s (Fig. 1E), evidencing a large heterogeneity in lateral mobility with a small percentage (ca. 5%) of stationary nanoclusters (Fig. S2 and SI Text).

Fig. 1.

LFA-1 nanoclusters diffuse randomly on resting monocytes. (A) Schematic description of the SDT experiments. Individual LFA-1 molecules inside nanoclusters were labeled with TS2/4-ATTO520 or L16-ATTO647N using sublabeling conditions. (B) Selected frame from a movie recorded at 100 ms per frame. Bright spots correspond to individual TS2/4-LFA-1 nanoclusters. White dots indicate the perimeter of the cell. (Scale bar: 5 μm.) (C) Selected LFA-1 nanoclusters at different times to illustrate different mobility behavior: fast (Top), stationary (Middle), and slow (Bottom). (D) Representative LFA-1 nanocluster trajectories displaying different lateral mobility, pseudocolor coded according to their apparent mobility: fast (orange), slow (blue), and stationary (gray). (E) Normalized semilog distribution of D values at short-time lags. The vertical arrow indicates the threshold value of diffusion D_th above which the mobile population for CPD analysis has been selected (370 trajectories from 128 cells in multiple experiments). (F) Square displacement plots of the TS2/4 mobile fractions at different time lags as obtained from CPD analysis (157 trajectories).

To enquire on the type of diffusion exhibited by the mobile LFA-1 nanoclusters over longer time intervals (ca. 1.5 s) we applied cumulative probability distribution (CPD) analysis (25). Using this approach, we could separate the mobile nanoclusters in two groups: (i) slow diffusing with D_s and α_s, and (ii) fast diffusing with D_f and α_f, where D represents the short-time lag diffusion coefficients and α indicates the type of motion (α = 1 for Brownian diffusion; α < 1 for hindered, anomalous diffusion) (SI Text). The analysis also rendered the relative fractions of slow and fast diffusing nanoclusters. At physiological cation conditions, the fast fraction (ca. 64%) of LFA-1 nanoclusters shows D_f = (5.6 ± 0.2) × 10^-2 μm²/s and α_f = 0.94 ± 0.06 (i.e., Brownian) (Fig. 1F and Table S1). The slow fraction (ca. 32%) shows a diffusion coefficient of D_s = (1.4 ± 0.1) × 10^-2 μm²/s and α_s = 0.9 ± 0.2. Thus, in contrast to earlier reports about immobile LFA-1 in resting T cells (12, 15, 16) our data on resting monocytes demonstrate that, at physiological cation conditions, LFA-1 nanoclusters are primarily mobile on the plasma membrane.

To confirm that we are tracking nanoclusters on monocytes, we also determined the diffusion of individual LFA-1 molecules on monocyte-derived immature dendritic cells (imDC), where LFA-1 is organized in a random (monomeric) fashion (20) (Fig. S3 A and B). As expected, the resultant D distribution of LFA-1 on imDCs showed a clear shift toward larger D values compared to that on monocytes (Fig. S3C). To relate the changes in D to the size of the moving particles, we further applied hydrodynamic theory (SI Text). Both the estimated values of D for individual LFA-1 molecules and nanoclusters agree well with those measured on imDCs and monocytes, respectively, supporting our statement that we are tracking individual nanoclusters on monocytes (see SI Text for further details). Additional control experiments at higher labeling conditions (Fig. S3 D and E) and high-density dual-color quantum dot tracking (Fig. S4) further support our assignment (see SI Text).

Primed LFA-1 Nanoclusters Exhibit Multiple Diffusion Profiles on Resting Monocytes.

We previously showed that, on resting monocytes, approximately 25% of the LFA-1 nanoclusters are in a primed state (20). To investigate the lateral mobility of these primed nanoclusters, we used the conformation-dependent epitope NKI-L16 (L16), which exclusively labels the extended conformation of the α_L subunit (26, 27) (Fig. 1A). Individual trajectories of L16⁺-LFA-1 nanoclusters were generated (Fig. 2A) together with the D histogram (Fig. 2B). A considerable higher fraction (ca. 20%) of L16⁺ nanoclusters was stationary, whereas the remaining approximately 80% exhibited D values similar to those recovered with the TS2/4 reporter consistent with the fact that primed nanoclusters constitute a subset of the overall LFA-1 population. Furthermore, CPD analysis revealed two different diffusion profiles (Fig. 2C): one fraction of L16⁺ nanoclusters (ca. 53%) exhibited fast, nearly Brownian diffusion (α_f = 0.91 ± 0.02), whereas the second fraction (ca. 28%) showed slow and anomalous diffusion (α_s = 0.49 ± 0.06) (Table S1). The overall fractions of stationary, slow, and fast diffusing LFA-1 nanoclusters for the two different reporters used are summarized in Fig. 2D. These results indicate that primed nanoclusters display multiple diffusion behavior with stationary, anomalous, and freely diffusing nanoclusters.

Fig. 2.

Primed LFA-1 nanoclusters exhibit multiple diffusion profiles on resting monocytes. (A) Representative trajectories of primed LFA-1 nanoclusters labeled with L16-ATTO647N. (B) Normalized semilog distribution of D values at short-time lags of L16⁺-LFA-1 nanoclusters (bars) compared to that of the total (TS2/4) LFA-1 population (dashes) (669 trajectories from 49 cells on multiple experiments). (C) Square displacement plots by fitting the CPD of the L16 mobile fractions at different time lags (380 trajectories). (D) Normalized fractions of stationary (gray), slow (blue), and fast (orange) mobile subpopulations for the total and primed nanoclusters. Error bars represent the standard deviation.

Reduction of Extracellular Ca²⁺ Restricts the Mobility of LFA-1 Nanoclusters.

Low extracellular Ca²⁺ or a combination of Mg²⁺/EGTA has been extensively used to promote high-affinity LFA-1 (14, 28). We thus performed SDT experiments as a function of Ca²⁺ reduction to selectively induce extended nanoclusters and monitor their lateral mobility on monocytes. Reduction of extracellular Ca²⁺ from 0.4 to 0.1 mM modestly reduced the TS2/4-LFA-1 mobile fraction (Fig. 3 A and G). In contrast, the diffusion of L16⁺ nanoclusters remained unaltered (Fig. 3 B and H). Strikingly, further reduction of Ca²⁺ down to 0.04 mM increased the number of stationary LFA-1 nanoclusters, regardless of the reporter used (Fig. 3 C, D, G, and H). Hence, the fraction of stationary LFA-1 nanoclusters increased from 5% at 0.4 mM Ca²⁺ to 43% at 0.04 mM Ca²⁺ for the TS2/4 reporter (Fig. 3G and Table S1) and from approximately 20% to 36% for the primed (L16⁺) nanoclusters (Fig. 3H and Table S1). Consistent with this increase of stationary nanoclusters, the binned histograms of the diffusion coefficients of the total and primed nanoclusters revealed an approximately twofold increase in the occurrence of lowest diffusion values when extracellular Ca²⁺ was reduced to 0.04 mM (red arrows in Fig. 3 E and F). Moreover, the D curve distributions of the total and primed nanoclusters decayed sharper at 0.04 mM, implying a severe slowing down of mobility (Fig. 3 E and F). CPD analysis further showed a significant decrease on the fast mobile fractions and an anomalous behavior at 0.04 mM Ca²⁺ regardless of the reporter used (Fig. 3 G and H, and Table S1). To rule out any effect of Mg²⁺, we repeated experiments at low extracellular Ca²⁺ levels in the presence of Mg²⁺ and observed no differences on LFA-1 mobility (Fig. S5), consistent with the fact that the effect of Mg²⁺ is locally confined to the headpiece of the integrin (2, 28). Collectively, these results reveal that reduction of extracellular Ca²⁺ dramatically decreases the mobility of LFA-1 on monocytes by immobilizing a large number of nanoclusters and severely restricting the diffusion of the remaining mobile ones.

Fig. 3.

The lateral mobility of LFA-1 nanoclusters is affected by extracellular Ca²⁺. (A, C, and E) Diffusion coefficients for the total (TS2/4) LFA-1 population at (A) 0.1 mM Ca²⁺ (bars) compared to 0.4 mM Ca²⁺, Mg²⁺ (dashes); (C) 0.04 mM Ca²⁺ (bars) compared to 0.4 mM Ca²⁺, Mg²⁺ (dashes); (E) binned distribution of D values for different extracellular Ca²⁺ conditions. The vertical arrow points to the lowest binned D value, and the curved arrow highlights the slopes of the D values. (B, D, and F) D values for L16⁺ nanoclusters at (B) 0.1 mM Ca²⁺ (bars) compared to 0.4 mM Ca²⁺, Mg²⁺ (dashes); (D) 0.04 mM Ca²⁺ (bars) compared to 0.4 mM Ca²⁺, Mg²⁺ (dashes); (F) binned distribution of D values for different extracellular Ca²⁺ levels. (G and H) Normalized fractions of stationary, slow, and fast mobile subpopulations for (G) the total (TS2/4), and (H) primed (L16⁺) nanoclusters at different Ca²⁺ levels. Error bars represent the standard deviation (p values are compared to the respective populations at 0.4 mM Ca²⁺, Mg²⁺: ∗  = p < 0.001 compared to the stationary fraction; # = p < 0.00003 compared to the fast fraction; †  = p < 0.03 compared to the stationary fraction; ‡  = p < 0.001 compared to the fast fraction). (I) Binned distribution of D values for TS2/4 (black squares) and L16 (green circles) at (Left) 0.4 mM Ca²⁺, Mg²⁺; (Center) 0.1 mM Ca²⁺; (Right) 0.04 mM Ca²⁺. TS2/4 trajectories: 369 (0.4 mM Ca²⁺, Mg²⁺), 318 (0.1 mM Ca²⁺), and 129 (0.04 mM Ca²⁺). L16 trajectories: 669 (0.4 mM Ca²⁺, Mg²⁺), 656 (0.1 mM Ca²⁺), and 102 (0.04 mM Ca²⁺).

Remarkably, the slopes of the binned D distribution for the total LFA-1 nanoclusters (Fig. 3E) became progressively steeper, indicating a gradual slowing down of mobility as a function of Ca²⁺ reduction. Because Ca²⁺ depletion favors the extended form of the integrin (2, 28), the changes in diffusion of the whole LFA-1 population might reflect a shift in the conformation equilibrium from mainly inactive to the primed form. Indeed, the binned diffusion histograms of the total and primed nanoclusters (Fig. 3I) showed an evolution profile in which both distributions became more like each other as the concentration of Ca²⁺ was reduced, until they remarkably matched each other at 0.04 mM Ca²⁺. Moreover, at 0.04 mM Ca²⁺, the diffusion behavior of the total and primed nanoclusters became quite similar to each other implying full conversion from inactive to primed LFA-1 nanoclusters (Fig. 3 G–I and Table S1). To further confirm that the immobilization of LFA-1 is due to priming, we also performed SDT experiments using Mn²⁺, a strong integrin activator (3, 28), and the activating antibody KIM185 (29). In both cases, the stationary population increased considerably, similarly to the results obtained at low-Ca²⁺ conditions (Fig. S6).

The Actin Cytoskeleton Regulates the Diffusion of LFA-1 Nanoclusters at Low Extracellular Ca²⁺ Conditions.

The immobilization of LFA-1 nanoclusters on the cell membrane at low-Ca²⁺ levels suggests their anchoring to the cytoskeleton. We thus used cytochalasin D (CytoD) to test the effect of disrupting cytoskeletal interactions on LFA-1 diffusion. Treatment with CytoD released a significant fraction of stationary L16⁺ nanoclusters at 0.4 mM Ca²⁺, Mg²⁺ (Fig. S7), and importantly released most of the L16⁺ nanoclusters at 0.04 mM Ca²⁺ (Fig. 4 A and C). This effect was extended to the total LFA-1 nanoclusters at 0.04 mM Ca²⁺ (Fig. 4 B and C). These data demonstrate that, at low extracellular Ca²⁺ levels, LFA-1 nanoclusters anchor to the cytoskeleton and suggest that extracellular activators such as cations have a potent role on global integrin conformation by coupling extracellular stimuli to intracellular signals that affect integrin lateral mobility.

Fig. 4.

The actin cytoskeleton regulates the diffusion of LFA-1 nanoclusters at low Ca²⁺ levels. (A and B) Normalized D distribution for (A) primed (L16⁺) and (B) total (TS2/4) nanoclusters at 0.04 mM Ca²⁺, with cells treated with CytoD (1 μg/mL) (bars), compared to the control (DMSO, 1%) (dashes). (C) Normalized fractions of stationary (gray), slow (blue), and fast (orange) mobile subpopulations for L16 (Left) at 0.04 mM Ca²⁺ in DMSO and after CytoD treatment. For this particular dataset, D_th = 0.002 μm²/s. Due to limited statistics, only the stationary fractions of TS2/4 (Right) in DMSO and after CytoD treatment were estimated. Error bars represent the standard deviation (∗  = p < 0.02 compared to stationary in DMSO; # = p < 0.006 compared to stationary in DMSO). TS2/4 trajectories: 55 (DMSO) and 35 (CytoD). L16 trajectories: 94 (DMSO) and 72 (CytoD).

The Decreased Mobility of LFA-1 Nanoclusters at Low-Ca²⁺ Levels Correlates with Restricted Microclustering Compromising Cell Adhesion and Spreading.

Because several studies suggest that LFA-1 mobility is associated with its ability to form microclusters (12, 30), we investigated the effect of extracellular Ca²⁺ on ligand-induced LFA-1 clustering. For this purpose, dense ICAM-1 patterned surfaces (5-μm squares) were prepared using microcontact printing (23). Clear accumulation of LFA-1 to the ICAM-1 regions was observed at physiological conditions compared to that at low-Ca²⁺ levels (Fig. 5 A and B). To quantify the degree of clustering we separated the data in two categories: (i) patches larger than the diffraction limit of our microscope (ca. 350-nm diameter) and defined them as microclusters and, (ii) smaller spots with a size limited by diffraction and defined them as nanoclusters (SI Text).

Fig. 5.

LFA-1 microclustering and cell adhesion under static conditions depends on extracellular Ca²⁺. (A and B) Representative frames of movies recorded in TIRF mode of L16⁺-LFA-1 nanoclusters on monocytes seeded on ICAM-1/BSA micropatterns at (A) 0.4 mM Ca²⁺, Mg²⁺ and (B) 0.04 mM Ca²⁺. Microclusters are surrounded in red, and nanoclusters are highlighted by white arrows. (Scale bars: 5 μm.) (C) Percentage of cells displaying microclusters at different cation conditions. Error bars are the standard deviation from three independent experiments. (D) Intensity distribution of nanoclusters on ICAM-1 regions at 0.04 mM Ca²⁺ (bars) and 0.4 mM Ca²⁺, Mg²⁺ (dashes). (E and F) Interference reflection microscopy of two representative monocytes seeded for 20 min on ICAM-1 substrates at (E) 0.4 mM Ca²⁺, Mg²⁺ and (F) 0.04 mM Ca²⁺. Attachment areas were (153 ± 100) μm² in E and (115 ± 44) μm² in F, thus a reduction of 75% in firm contact area (55 cells inspected in each condition, over 2 separate experiments). (Scale bars: 5 μm.) (G) Histogram of firm cell attachment ratio for monocytes seeded for 20 min on ICAM-1 substrates at 0.04 mM Ca²⁺ (bars) and 0.4 mM Ca²⁺, Mg²⁺ (dashes). The attachment ratio was estimated by dividing the main contact area as measured by interference reflection microscopy by the corresponding bright field image. Mean attachment ratio is 0.70 (σ = 0.19) at 0.4 mM Ca²⁺, Mg²⁺ and 0.55 (σ = 0.16) at 0.04 mM Ca²⁺, thus a reduction to 79% compared to the total cell size at 0.4 mM Ca²⁺. (H) Relative cell adhesion at 0.4 mM Ca²⁺, Mg²⁺ (white) and 0.04 mM Ca²⁺ (black) at different seeding times on ICAM-1 substrates. Results from two independent experiments from 15 different bright field images at each condition.

From 20 to 30 cells studied at each condition, the number of L16⁺ microclusters on ICAM-1 regions was significantly different, with approximately 45% of the cells displaying L16⁺ microclusters at 0.4 mM Ca²⁺, Mg²⁺, and approximately 20% at 0.04 mM Ca²⁺ (Fig. 5C). Interference reflection microscopy further showed extensive areas of close cell-ligand substrate contacts at 0.4 mM Ca²⁺, Mg²⁺, whereas reduced contact areas and finger-like extensions (filopodia) were observed at low-Ca²⁺ levels (Fig. 5 E–G). In addition, the number of cells that firmly adhered at low-Ca²⁺ levels reduced substantially compared to physiological Ca²⁺ levels (Fig. 5H).

In the case of L16⁺ nanoclusters, we did not observe significant differences in the intensity distributions (Fig. 5D), indicating that low extracellular Ca²⁺ has no affect on LFA-1 nanoclustering. These results were further confirmed by transmission electron microscopy of whole-mount monocytes (Fig. S8). Thus, although LFA-1 nanoclustering remains unperturbed at low-Ca²⁺ levels, ligand-induced microclustering is greatly restricted and directly correlated with its decreased mobility on the cell membrane. Furthermore, these results demonstrate that extracellular Ca²⁺ is not only an important regulator of ligand-binding affinity but it is also required to couple ligand binding to global conformational changes and downstream signals that trigger cell spreading and adhesion on ligand bearing substrates under static conditions.

LFA-1 Mediated Adhesion of Monocytes Under Shear-Flow.

Because extracellular Ca²⁺ affected conformational state, lateral mobility and microclustering of LFA-1 on ICAM-1 bearing patterns, we studied the effect of extracellular Ca²⁺ on cell adhesion to ICAM-1 substrates under shear flow. At low shear-flow conditions (0.2 dyn/cm²) highest adhesion was found at 1 mM (Mn²⁺, Ca²⁺, Mg²⁺), followed by 0.4 mM (Ca²⁺, Mg²⁺) (Fig. 6A), conditions where LFA-1 is laterally mobile and able to increase its avidity state. Remarkably, adhesion was lowest at 0.04 mM Ca²⁺ and slightly higher at 0.04 mM Ca²⁺, 1 mM Mg²⁺ (Fig. 6A), a situation where high-affinity nanoclusters are generated but lateral mobility is compromised. Thus, despite the fact that low extracellular Ca²⁺ levels promote primed LFA-1, the integrin is not able to support firm adhesion.

Fig. 6.

Extracellular Ca²⁺ plays a differential role on LFA-1 adhesiveness under shear-flow conditions. Cell binding to ICAM-Fc coated surfaces quantified as a percentage of maximum binding observed at 1 mM Ca²⁺, Mg²⁺, Mn²⁺ at (A) 0.2 dyn/cm² and (B) 0.5 dyn/cm². Error bars are the standard deviation over average binding calculated over five different areas. One out of two experiments are presented (∗  = p < 0.007 compared to 1 mM Ca²⁺, Mg²⁺, Mn²⁺; # = p < 0.005 compared to 1 mM Ca²⁺, Mg²⁺, Mn²⁺; †  = p < 0.01 compared to 0.04 mM Ca²⁺; ‡  = p < 0.007 compared to 0.4 mM Ca²⁺.

At higher shear-flow conditions (0.5 dyn/cm²), firm cell adhesion was equally maintained with 1 mM (Mn²⁺, Ca²⁺, Mg²⁺) (Fig. 6B). However, cell adhesion was highly dependent on the presence of Mg²⁺, regardless of the Ca²⁺ levels used (Fig. 6B). Because low extracellular Ca²⁺ favors the extended form of the integrin and Mg²⁺ increases the binding strength of LFA-1 to ICAM-1 without affecting its lateral mobility (Fig. S5), these results indicate that at higher shear forces affinity of LFA-1 to its ligand plays a more prominent role than its lateral mobility.

Discussion

Earlier models of LFA-1 regulation proposed that the molecule is inactive and confined to the cytoskeleton in resting cells and released from its cytoskeletal constrains upon cell activation, resulting in increased mobility and clustering (12, 13, 15, 16). However, recent work on resting T cells demonstrated a more complex mobility pattern of LFA-1 with distinct diffusion profiles and different subsets of conformational states (31). Here we showed that at physiological cation conditions, LFA-1 nanoclusters are predominantly mobile, with a subset of nanoclusters being in the extended, primed form. These primed nanoclusters showed stationary, anomalous, and free diffusion behavior. As the L16 epitope recognizes full extension of the α_L subunit (26, 27), regardless of whether the ligand-binding headpiece is closed (intermediate form) or opened (high-affinity form), the diverse mobility profiles recovered in our studies might correspond to the different binding states of LFA-1. More comprehensive models describing different conformational states of LFA-1 already hypothesized the existence of multiple intermediate states depending on their transient interaction with the cytoskeleton (6). Our results now demonstrate this complexity. We propose that this repertoire of multiple possibilities offered by LFA-1 is beneficial for its diverse adhesive processes dynamically regulating migration, rolling, and firm arrest.

Our results also provide evidence for the strong effects of extracellular Ca²⁺ on conformational state, lateral mobility, and ligand-induced microclustering of LFA-1. Although previous works demonstrated the importance of extracellular Ca²⁺ regulating integrin activation (2, 28, 32), our data prove that extracellular Ca²⁺ also couples ligand binding to global conformational changes and downstream signaling that trigger cell spreading and firm adhesion in the presence of its ligand. The mechanism by which extracellular Ca²⁺ couples conformational changes to lateral mobility on the cell membrane must be related to the profound changes that Ca²⁺ transmits throughout the global structure of the integrin toward the cytoplasmic tails, by acting on its hybrid domain (33). Swinging out of the hybrid domain upon lowering of extracellular Ca²⁺ results in integrin extension and transmembrane domain separation, so that both α_L and β₂ cytoplasmic tails become available for interaction with other proteins on the inner side of the membrane that will regulate the lateral mobility of the receptor and its interaction with the cytoskeleton (6).

Although our SDT experiments have been performed in the absence of mechanical stimuli, recent simulations predicted that force accelerates the swinging out of the hybrid domain, facilitating the transition into a high-affinity open headpiece state (33). Low shear forces were also seen to induce conformational changes on the α-I domain of LFA-1, increasing adhesive interactions with its ligand by stabilizing the open conformation of this domain (34). Interestingly, it has been recently suggested that preformed integrin anchorage to the cytoskeleton could be key for the integrin to load low forces and undergo instantaneous activation by surface-bound ligands (35). Our data demonstrate that, in the resting state, a significant fraction of LFA-1 is in primed form and preanchored to the cytoskeleton. Activation of the integrin (by Ca²⁺ removal, Mn²⁺, or activating antibodies) leads to an increase of cytoskeleton anchorage forming nascent sites for adhesion after ligand engagement. It will be then expected that this preformed cytoskeleton anchorage create the perfect scenario for tensile forces to facilitate/accelerate the process of leukocyte adhesion on blood vessels. Low-affinity, mobile LFA-1 nanoclusters, as identified here, might contribute by diffusing to the adhesion sites initiated by the anchored extended LFA-1 subsets.

Materials and Methods

Cell Culture and Labeling for Single-Dye Tracking.

THP-1 monocytes were cultured in RPMI medium 1640 Dutch modification medium supplemented with 10% fetal calf serum and antibioticantimycotic from Gibco. Cells were labeled with 2 μg/mL TS2/4-ATTO520, or 2 μg/mL L16-ATTO647N to allow for single-dye tracking experiments (SI Text and Fig. S1). Further details on substrate preparation, micropatterns, labeling conditions under different cation conditions, and CytoD treatment are found in SI Text.

Shear-Flow Chamber Assay.

Details of the experiments are described in SI Text.

Single-Dye Tracking.

Experiments were performed using a homemade single-molecule Epi/total internal reflection fluorescence (TIRF) microscope. Samples were illuminated either in Epi or TIRF modes. Excitation was provided by a He:Ne laser (4 ms at 633 nm, 1 kW/cm²) or an Ar⁺-Kr⁺ laser (2 ms at 514.5 nm, ca. 2 kW/cm²). Fluorescence was collected with a 1.45 N.A. oil immersion objective and guided into an intensified CCD camera. Individual frames were retrieved at a frame rate of 10–20 Hz. Experiments were performed at 37 °C. MSD analysis was used to analyze trajectory data and to derive the short-range D values. CPD analysis was performed according to Schütz et al. (25) to derive the long-term D values of mobile trajectories (see SI Text for further details).