WASP family proteins regulate the mobility of the B cell receptor during signaling activation

Abstract

Regulation of membrane receptor mobility tunes cellular response to external signals, such as in binding of B cell receptors (BCR) to antigen, which initiates signaling. However, whether BCR signaling is regulated by BCR mobility, and what factors mediate this regulation, are not well understood. Here we use single molecule imaging to examine BCR movement during signaling activation and a novel machine learning method to classify BCR trajectories into distinct diffusive states. Inhibition of actin dynamics downstream of the actin nucleating factors, Arp2/3 and formin, decreases BCR mobility. Constitutive loss or acute inhibition of the Arp2/3 regulator, N-WASP, which is associated with enhanced signaling, increases the proportion of BCR trajectories with lower diffusivity. Furthermore, loss of N-WASP reduces the diffusivity of CD19, a stimulatory co-receptor, but not that of FcγRIIB, an inhibitory co-receptor. Our results implicate a dynamic actin network in fine-tuning receptor mobility and receptor-ligand interactions for modulating B cell signaling.

Fig. 1. Single-particle tracking reveals wide range of BCR mobility.

a Panel showing primary murine B cell spreading (IRM, above) and BCR clustering (TIRF, below). Scale bar is 3 μm. b Experimental schematic, indicating activated murine primary B cells, placed on supported lipid bilayers coated with mono-biotinylated fragments of antibody (mbFab). Cells are imaged in TIRF mode and the concentration of AF546 labeled mbFab is kept low enough to image single-molecule events. c Representative TIRF image with the bright dots representing single BCR molecules. The cell contour is obtained from an IRM image taken after TIRF imaging. Scale bar is 1 μm. d The collection of tracks obtained for a control cell during a 10-min period imaged at 33 Hz for 1000 s every minute. The tracks are color coded for diffusivity. Scale bar is 1 μm. e Cumulative distribution function (CDF) for the diffusivities measured at 1, 3, 5, 7, and 9 min after activation for BCR in B cells from control mice. f Boxplot showing BCR diffusivities at the indicated time points (N = 15 cells). The mean is marked with red diamonds, the bottom line represents the lower quartile, the upper line the upper quartile, the whiskers show the extent of the rest of the data, and red crosses are the outliers. Significance of differences was tested using the Kruskal–Wallis test (***p < 0.001; 1 min vs 3 min, p = 0.0008; min 1 vs min 5, p = 0.000038; min 3 vs min 5, p = 0.1767; min 5 vs min 7, p = 0.8614).

Fig. 2. Perturbation expectation maximization analysis identifies eight distinct diffusive states for BCR in control cells.

a Characteristic tracks belonging to each of the BCR diffusive states identified by pEM. Diffusivity increases from State 1 (slowest) to State 8 (fastest). Scale bar is 1 μm. b Ensemble mean-square displacement (eMSD) plots for each of the states. Colors corresponding to different states are as shown in the legend. (N = 15 cells). c Plot showing the mean diffusivity for the trajectories belonging to each state at every time point. Error bars represent the standard error of the mean. d Plots showing the fraction of BCR tracks that are sorted in each state at every time point. Error bars represent a confidence interval of 95% on the population fraction calculation. e Plot of pair correlation as a function of distance for all states. Source data are provided as a Source Data file.

Fig. 3. Inhibition of actin nucleation decreases BCR diffusivity.

a Plots of BCR diffusivity distributions for cells treated with CK666 (inhibitor of Arp2/3 complex) or SMIFH2 (inhibitor of formins). (p < 0.001, Kruskal–Wallis test for comparison between DMSO and CK666, or DMSO and SMIFH2). b Population fraction over time for cells treated with CK666. c Population fraction over time for cells treated with SMIFH2. The colors corresponding to the different states are as shown in f. d BCR diffusivity distribution for cells treated with wiskostatin (Wisko) compared with DMSO control. (p < 0.01, Kruskal–Wallis test for comparison between DMSO and Wisko) e Population fraction over time for cells treated with wiskostatin. Error bars in b, c, and e represent a confidence interval of 95% on the population fraction calculation. f Overall distribution of population fractions for cells treated with wiskostatin, CK666 and SMIFH2 (Number of cells: DMSO, N = 14; Wisko, N = 11; CK666, N = 10; SMIFH2, N = 16). Source data are provided as a Source Data file.

Fig. 4. N-WASP knockout leads to predominance of BCR molecules in lower mobility diffusive states.

a Collection of tracks obtained from pEM analysis of BCR molecules in a cNKO cell during a 10-min period. The tracks are color coded for diffusivity. Scale bar is 1 μm. b Cumulative distribution function for diffusivities measured at 1, 3, and 5 min after BCR stimulation in B cells from cNKO mice. c Plots of population fractions of eight distinct diffusive states as a function of time for BCR in cNKO cells. Error bars represent a confidence interval of 95% on the population fraction calculation. The colors corresponding to the different states are as shown in f. d Pair correlation function plots of the trajectories in different diffusive states for cNKO cells. e The distribution of diffusivities from the 5–10 min time period after activation, for BCR in control, WKO, and cNKO cells. The distributions for control are significantly different from WKO and cNKO cells (control cells N = 15, WKO cells N = 21 cells, p < 0.001 and cNKO cells N = 17, p < 0.0001 Kruskal–Wallis test). f Comparative population fractions for BCR in different states over the entire time period in control, WKO, and cNKO cells. Significance levels for the differences are in Supplementary Table 1. Source data are provided as a Source Data file.

Fig. 5. N-WASP expression modulates CD19 diffusivity.

a Instant SIM images of activated B cell, showing that AF546 labeled BCR (red) and AF488 labeled CD19 (green) reside in clusters that colocalize to within the ~150 nm resolution limit. Scale bar is 5 μm. b Intensity profiles for BCR (red) and CD19 (green) fluorescence along the yellow lines as drawn in a. c Compilation of CD19 tracks over a 10-min period in an activated control B cell. Scale bar is 1 μm. d Plot showing the mean diffusivity of each of the eight diffusive states obtained from pEM analysis as a function of time. The colors corresponding to the different states are as shown in h. Error bars represent the standard error of the mean. e, f Pair correlation function plot for all states for control and N-WASP-KO cells respectively. g Cumulative Probability distribution of diffusivities showing that mobility of CD19 in cNKO cells is significantly lower than in control cells (control cells N = 10, cNKO cells N = 11, p = 0.0013 Kruskal–Wallis test). h, i Comparison of population fractions of BCR and CD19 in different states for control and cNKO cells, respectively. Source data are provided as a Source Data file.

Fig. 6. FcγRIIB mobility is mildly affected by the lack of N-WASP.

a A set of single-molecule tracks of FcγRIIB from an activated B cell over a 10-min period. Scale bar is 1 μm. b Cumulative distribution plots for diffusivity of FcγRIIB molecules in control and cNKO cells (control cells N = 12, cNKO cells N = 12). c pEM analysis of single FcγRIIB molecule trajectories uncovered seven states. Plot shows the mean diffusivity of each state at every time point. Error bars represent the standard error of the mean. The colors corresponding to the different states are as shown in f. d, e Pair correlation function plot for all states in control and cNKO states respectively. f Comparison of the population fraction of different diffusive states of FcγRIIB in control and cNKO cells. Source data are provided as a Source Data file.

Fig. 7. Effect of actin regulators on actin dynamics in activated B cells.

a iSIM images of activated Lifeact-EGFP B cells at consecutive time points for two conditions: DMSO carrier-control and wiskostatin (10 μM concentration). The initial time corresponds to 5 min after spreading initiation. The blue arrows in the images indicate the emergence of actin foci and the yellow arrows point to spreading and contraction of the lamellipodial region of the cells. Scale bars are 2 μm. b STICS (Spatio-temporal image correlation spectroscopy) vector map showing actin flows represented by velocity vectors indicating flow direction and color coded for flow speed. In the zoomed region, the velocity vectors show the flow direction and flow speed. The vector map is overlaid on top of a grayscale image of Lifeact-EGFP. c Pseudocolor map of actin flow speeds corresponding to 2 min after cell spreading for representative DMSO-control and wiskostatin-treated cells. d Cumulative probability distribution of actin flow speeds for DMSO-control cells (blue, N = 11 cells) and cells treated with wiskostatin (red, N = 12 cells). (p = 0.00074, Kruskal–Wallis test). e Directional coherence maps indicating the flow directions, which ranged from inward (1) to outward (−1). f, g Probability density function plots showing directional coherence values of actin flow in cells during the early stage of activation (f) or cells in the late stage of activation (g), with subplots highlighting the flow fraction defined as inward flow (see Methods). During early stages, the fraction of inward flow is 0.143 for DMSO and 0.1425 for wiskostatin-treated cells (p = 0.5188—not significant); during late stages, the fraction of inward flow is 0.113 for DMSO and 0.1518 for wiskostatin-treated cells (p < 0.001).

Fig. 8. The actin cytoskeleton regulates B cell receptor mobility and signaling in different stages.

Representative cartoon showing receptor distributions on a section of the B cell membrane: a Resting B cell membrane: actin networks restrict receptor lateral movement and interactions. b B cell membrane at the early signaling activation stage. Actin remodeling enhances receptor mobility allowing for interactions between receptors, specifically BCR and CD19, enhancing signaling. Actin flows towards the center and edges of the immune synapse in similar proportions. c B cell membrane at later activation stages. Top: actin flows stir the cytoplasm at the membrane vicinity, increasing the mixing of receptors in the membrane and thereby allowing signal inhibitory molecules to downregulate BCR signaling. Bottom: N-WASP knockout reduces actin dynamics and changes the balance of actin flow directionality at later stages (5–10 min) of activation, leading to enhanced signaling.