We initially utilized two experimental models previously used to study BCR-mediated internalization of surface-associated antigen: F(ab’)₂-anti-mouse IgM+ G (αM, which binds and activates mouse BCRs) immobilized on beads or tethered to planar lipid bilayers (PLB) by biotin-streptavidin interaction. Beads or PLB coated with transferrin (Tf) at similar surface density as αM were used as controls, as Tf does not activate the BCR and interacts with the Tf receptor with similar affinity as the bona fide antigen hen egg lysozyme (HEL) binds to the BCR of transgenic MD4 mouse B-cells (Batista and Neuberger, 1998; Fuchs and Gessner, 2002). Strikingly, live imaging revealed influx of the membrane-impermeable dye propidium iodide (PI) at sites of mouse splenic B-cell contact with αM-beads, indicating that PM permeabilization occurred at bead-binding locations (Figure 1A, Figure 1—figure supplement 1 and Videos 1–3). While similar percentages of B-cells bound αM- or Tf-beads (Figure 1B), a significantly higher fraction of B-cells binding αM-beads became PI-positive (Figure 1C). Flow cytometry analysis confirmed the increased PI entry in B-cells binding αM-beads when compared to Tf-beads (Figure 1D–G and Figure 1—figure supplement 2). Addition of soluble F(ab’)₂-anti-mouse IgM+ G (sαM, also capable of binding and activating the BCR) did not increase the frequency of PI entry in B-cells binding to Tf-beads (Figure 1F). The percentage of cells positive for cleaved caspase-3, an early apoptotic marker, was similar in B-cells interacting or not with αM- or Tf-beads and only increased significantly after treatment with staurosporine (Figure 1—figure supplement 3), suggesting that PM permeabilization is not associated with apoptosis. Similar observations were made using the PLB system that allows lateral movement of the tethered antigen (Dustin et al., 2007). Significantly more B-cells became PI-positive when contacting αM-PLB when compared to Tf-PLB (Figure 1H–J).

PM permeabilization in B-cells binding to αM-PLB was also observed using membrane-impermeable lipophilic FM probes. These fluorescent dyes have been used extensively to assess PM integrity, because they only label the outer PM leaflet of intact cells but rapidly stain intracellular membranes when entering the cytosol (Bansal et al., 2003; McNeil et al., 2003; Demonbreun et al., 2019). After >30 min of interaction with αM-PLB, we observed sudden, massive increases in FM1-43 staining of intracellular membranes, including the nuclear envelope (Figure 1H and I, Figure 1—figure supplements 4 and 5 and Video 4). Consistent with the PI entry results (Figure 1J), significantly more B-cells showed a sudden increase in intracellular FM staining when contacting αM-PLB compared to Tf-PLB (Figure 1K). This characteristic pattern of sudden FM influx with staining of the nuclear envelope was only observed in B-cells that eventually became PI-positive, not in cells that remained PI-negative during interaction with αM-PLB (Figure 1—figure supplement 4). Since FM lipophilic dyes can also be internalized through surface receptor endocytosis, we activated BCR endocytosis by cross-linking surface BCRs using soluble F(ab’)₂ goat-anti-mouse IgM+ G antibodies followed by fluorescent F(ab’)₂ anti-goat-IgG (Song et al., 1995; Hoogeboom and Tolar, 2016). Under these conditions, which did not cause PM permeabilization, we observed FM1-43 uptake appearing as small peripheral puncta that colocalized with BCR cross-linking antibodies. Such endosome-associated FM1-43 staining pattern was markedly different from the sudden, massive FM influx observed shortly before PI entry in permeabilized cells (Figure 1—figure supplement 6 and Video 4). Collectively, these data show that the sudden, massive influx of FM dyes during αM-PLB binding is caused by B-cell permeabilization, and not by a gradual endocytosis of the PM-associated tracer triggered by BCR engagement.

As an independent method to demonstrate antigen-induced permeabilization of B-cells, we took advantage of the ability of membrane-impermeable Ponceau 4R to quench cytosolic fluorophores upon entering cells (Tay et al., 2019). Instead of monitoring nuclear or intracellular membrane staining by membrane-impermeable fluorescent dyes, we determined the percentage of B-cells pre-loaded with carboxyfluorescein succinimidyl ester (CFSE) that lost their cytosolic fluorescence as a consequence of Ponceau 4R entry during PM permeabilization. To validate this method, we first permeabilized B-cells with the pore-forming toxin streptolysin O (SLO). In the presence of Ponceau 4R, the percentage of B-cells with reduced CFSE fluorescence increased significantly after exposure to SLO (Figure 2A and B), mimicking what we previously observed for PI entry in SLO-treated B-cells (Miller et al., 2015). Thus, quenching of cytoplasmic CFSE by the membrane-impermeable Ponceau 4R is a potent indicator of PM permeabilization. Using this method, we compared B-cells incubated with αM- or Tf-PLB by live imaging. A significantly higher fraction of CFSE-labeled B-cells showed fluorescence quenching when interacting with αM-PLB, quantified as the percentage of cells that lost >70% of their initial CFSE fluorescence (Figure 2C and D and Video 5). The average time for detection of αM-PLB-induced B-cell permeabilization measured by this quenching method was similar to what was observed for FM entry, while the average time for intracellular detection of PI showed a ~ 8 min delay (Figure 2E, Figure 1—figure supplement 4 and Video 4). An analysis of the cumulative rate of influx of the three distinct tracers confirmed the small delay in PI detection (Figure 2F). Thus, FM influx and Ponceau 4R-mediated quenching are more sensitive methods for detecting the onset of B-cell PM permeabilization when compared to PI influx, which is only clearly visualized after intercalation into double-stranded DNA inside the nucleus. Based on consistent results obtained with three different methods, we conclude that BCR binding to αM-coated surfaces (but not to soluble αM) causes localized permeabilization of the B-cell PM.

We next determined whether HEL, a bona fide antigen recognized by the BCR from MD4 mice, also caused B-cell permeabilization when tethered to artificial surfaces or presented as an integral membrane protein (mHEL) on the surface of live cells (Batista et al., 2001). Flow cytometry analysis revealed that similar fractions of MD4 B-cells become PI-positive after binding beads coupled to αM or to HEL (Figure 3A–C). In contrast, WT B-cells binding to HEL- beads showed a low percentage of PI-positive cells, similar to what is observed with Tf-beads (Figures 1F and 3A–C). Importantly, transmembrane mHEL-GFP expressed on the surface of live COS-7 cells co-clustered with the BCR at sites of interaction with MD4 B-cells, followed by PI influx. This dramatic clustering pattern followed by permeabilization was not observed in WT B-cells, whose BCR is incapable of specifically recognizing HEL (Figure 3D and Videos 6 and 7). A significantly higher percentage of MD4 B-cells showed PI influx after interaction with COS-7 cells expressing mHEL-GFP, when compared to WT B-cells (Figure 3E). The percentage of PI-positive MD4 B-cells was also significantly higher after incubation with mHEL-expressing COS-7 cells than with mock-transfected cells (Figure 3F). Collectively, these results show that BCR binding to surface-associated antigen can cause permeabilization of the B-cell PM.

High-affinity binding of the BCR to antigen associated with non-internalizable surfaces induces high levels of BCR signaling, cytoskeleton reorganization, and antigen endocytosis (Batista and Neuberger, 1998; Batista and Neuberger, 2000; Fleire et al., 2006). To examine the impact of the BCR-binding affinity on antigen-induced PM permeabilization, we incubated MD4 B-cells with beads coated with equal densities of HEL or the duck egg lysozyme isoform DEL-I. The MD4 BCR binds DEL-I with >100 fold lower affinity than it binds HEL (Langley et al., 2017). As expected, the percentage of B-cells binding multiple beads was reduced when the BCR-antigen affinity decreased (Figure 4—figure supplement 1A), but B-cells binding one single bead were detected for both HEL and DEL-I and also Tf (Figure 4—figure supplement 1). In these single bead-bound populations, DEL-I-beads caused significantly less PI entry than HEL-beads (Figure 4A). Inhibition of signaling with the Src kinase inhibitor PP2 (Cheng et al., 2001) (iSrc) or the Bruton’s Tyrosine Kinase inhibitor AVL-292 (Aalipour and Advani, 2013) (iBTK) (Figure 4B and C) also reduced PI entry in cells binding HEL-beads (Figure 4D). After contact with αM-PLB or αM-beads but not Tf-PLB or Tf-beads, surface BCRs became polarized toward PLB- or bead-binding sites within ~10 min, a period markedly shorter than what is required for detection of PM permeabilization through FM influx (Figure 4E–H, Figure 4—figure supplement 2, and Video 8). Importantly, the activated form of the actin motor protein NMII, detected through its phosphorylated light chain (pMLC), accumulated along with the BCR at αM-bead-binding sites (Figure 4G, Figure 4—figure supplement 3 and Video 9). The fluorescence intensity ratios (FIR) of surface BCRs and pMLC were significantly higher in B-cells binding αM-beads than in cells binding Tf-beads (Figure 4H and I). Notably, inhibition of NMII motor activity with blebbistatin (Bleb) markedly reduced the number of B-cells that became PI-positive during interaction with αM-beads, without affecting the cells’ ability to bind the beads (Figure 4J and K). Live imaging detected PI entry following a ‘tug-of-war’ between two B-cells simultaneously engaging an αM-bead (Video 3 and Figure 1—figure supplement 1C), further supporting a role for NMII-mediated traction forces in antigen-induced PM permeabilization. Thus, our results indicate that PM permeabilization caused by surface-associated antigen requires strong BCR-antigen interaction and the subsequent activation of signaling and NMII motor activity.

Permeabilization with the pore-forming toxin SLO triggers exocytosis of lysosomes in mouse primary B-cells (Miller et al., 2015), a response to Ca²⁺ influx that is observed in several cell types and is required for the resealing of PM wounds (Reddy et al., 2001). To determine if permeabilization by surface-associated αM or HEL triggered exocytosis of lysosomes in B-cells, we first examined whether luminal epitopes of the lysosomal membrane protein LIMP-2 were exposed on the cell surface. Flow cytometry detected surface LIMP-2 in a higher percentage of B-cells binding αM-beads than in B-cells binding Tf-beads (Figure 5A and B). Notably, surface exposure of LIMP-2 was lower in MD4 B-cells binding DEL-I-beads compared to HEL-beads (Figure 5C). These results reveal a close correlation between the extent of PM permeabilization (Figure 4A) and lysosomal exocytosis induced by surface-associated αM, HEL or DEL-I (Figure 5B–C). Surface LIMP-2 was predominantly detected at sites of αM-bead binding (Figure 5D and Figure 5—figure supplement 1) and this polarized pattern, measured by FIR, increased after ~30 min of interaction with αM- but not Tf-beads (Figure 5E). Notably, this timeframe was similar to the average period required for PM permeabilization (Figure 2E). Next, we performed live total internal reflection fluorescence (TIRF) microscopy of B-cells preloaded with the luminal lysosomal probe SiR-Lyso (a membrane-permeable fluorescent peptide that binds to the lysosomal enzyme cathepsin D) while contacting αM-PLB. Exocytosis events were identified by rises in the fluorescence intensity of SiR-Lyso puncta (reflecting lysosome entry into the TIRF evanescent field adjacent to the PM) followed by sharp decreases within ~2 s (reflecting dye dispersion upon fusion of lysosomes with the PM) (Figure 5F and G, Figure 5—figure supplement 2 and Video 10). Exocytosis events were observed in the majority of individual PI-positive cells interacting with αM-PLB (Figure 5H) and occurred predominantly ~30–45 min after αM-PLB contact (Figure 5H and I), a timing similar to PM permeabilization and LIMP-2 exposure. Lysosomal exocytosis events were significantly more frequent in permeabilized B-cells when compared to B-cells that remained intact (Figure 5J). These results show that permeabilization of B-cells by surface-associated antigen triggers exocytosis of lysosomes.

We next determined if B-cells were capable of resealing their PM, by using an assay involving sequential exposure to two different membrane-impermeable fluorescent dyes (Reddy et al., 2001). Resealed cells were quantified by flow cytometry as the percentage of permeabilized cells binding αM-beads (stained intracellularly with FM4-64 kept throughout the assay) that excluded the membrane-impermeable dye SYTOX Blue (added only during the last 10 min of the assay) (Figure 6A). Under these conditions, ~50 % of B-cells permeabilized by surface-associated antigen resealed their PM within the assay period (Figure 6B). Inhibition of lysosomal exocytosis with bromoenol lactone (BEL) (Fensome-Green et al., 2007; Tam et al., 2010) significantly reduced the percentage of resealed cells (Figure 6A and B). We found no evidence that the reduction in resealed cells after BEL treatment was due to toxicity of this inhibitor. B-cell populations with low forward-scatter versus side-scatter values typical of dead cells did not increase after BEL treatment (Figure 6—figure supplement 1). Exposure to BEL also did not increase the small fraction ( < 7%) of Tf-bead-binding B-cells that was permeable to SYTOX Blue (Figure 6—figure supplement 1). These data suggest that lysosomal exocytosis is required for the resealing of B-cells permeabilized by binding to surface-associated antigen. To confirm that individual antigen-permeabilized B-cells resealed, we used live imaging to visualize cells incubated with αM-PLB in the presence of SYTOX Green. PI was then added for the last 10 min of the 4 hr incubation. Time-lapse images showed that B-cells that became permeable to SYTOX Green during interaction with αM-PLB subsequently excluded PI – a direct indication that their PM resealed during the 4 hr assay period (Figure 6C and Video 11). As expected, cells that were already permeable to SYTOX Green at the beginning of the incubation (likely non-viable cells that were damaged prior to the incubation) were also permeable to PI (which causes strong quenching of the SYTOX green fluorescence upon entering cells - Figure 6C and Video 11). Interestingly, primary B-cells permeabilized during interaction with αM-beads (Figure 1A and Video 1) or αM-PLB (Figures 1H and 6C and Video 11) often displayed a shape change visualized as an increase in cell diameter, but after resealing this morphological change was gradually reversed (Figure 6C, Figure 6—figure supplement 2, Videos 11 and 12). Collectively, our results indicate that B-cell PM permeabilization by binding to surface-associated antigen is a reversible event, and that lysosomal exocytosis is required for PM resealing as previously shown for other cell types (Andrews et al., 2014).

We investigated the relationship between PM permeabilization by surface-associated antigen and antigen internalization using fluorescent αM covalently bound to beads or tethered to PLB. Live imaging detected αM puncta moving away from bead-binding sites into B-cells, increasing progressively between 30 and 60 min of interaction (Figure 7A and B and Video 13). In contrast, intracellular fluorescent puncta were markedly less abundant during the same time period in cells not binding αM-beads, or binding Tf-beads (Figure 7B). Inhibition of antigen-mediated PM permeabilization with blebbistatin significantly reduced extraction and internalization of αM coupled to beads (Figure 7C). When similar experiments were performed with PLB, the fraction of cells containing internalized αM and the total amount of αM uptake were significantly higher in permeabilized cells with high levels of intracellular FM staining (FM-high), compared to non-permeabilized cells with low FM staining (FM-low) (Figure 7D–F). These data suggest that αM-induced PM permeabilization, rapidly followed by lysosomal exocytosis, promotes extraction and internalization of αM from non-internalizable surfaces.

Next, we investigated whether antigen internalization enhanced by PM permeabilization and lysosomal exocytosis impacts antigen presentation by B-cells. We compared levels of IL-2 secretion by the 3A9 T-cell hybridoma line (Allen and Unanue, 1984) after activation by B-cells exposed to HEL- or DEL-I-beads. B-cells exposed to high concentrations of soluble HEL or DEL-I induced similar levels of IL-2 secretion (Figure 7G), demonstrating that the primary B-cells used in these assays could process and present the conserved peptide present in both HEL and DEL-I for T-cell activation. In contrast, when the B-cells were exposed to lower amounts of surface-associated antigens, B-cells exposed to HEL-beads activated T-cells to produce IL-2 at markedly higher levels than cells exposed to DEL-I-beads (Figure 7H). These results indicate that B-cell permeabilization resulting from high-affinity antigen-BCR interaction, with its corresponding lysosomal exocytosis response, facilitates the presentation of antigen associated with non-internalizable surfaces.

Primary B-cells were isolated from the spleens of wild-type C57BL/6, MD4 transgenic (C57BL/6 background), B10.BR-H2^k2 H2-T18^a/SgSnJJrep (Jackson Laboratories), and F1 of B10.BR-H2^k2 H2-T18^a/SgSnJJrep x MD4 mice using a previously published protocol (Miller et al., 2015). Briefly, mononuclear cells were isolated by Ficoll density-gradient centrifugation (Sigma-Aldrich). T-cells were removed with anti-mouse CD90.2 mAb (BD Biosciences) and guinea pig complement (Innovative Research, Inc) and monocytes and dendritic cells by panning. B-cells were kept at 37 °C and 5 % CO₂ before and during experiments. All procedures involving mice were approved by the Institutional Animal Care and Usage Committee of the University of Maryland.

The A20 B-cell lymphoma line (ATCC #TIB-208) was cultured in DMEM (Lonza) supplemented with 10 % of FBS (Thermo Fisher Scientific), 0.05 mM 2-mercaptoethanol (Sigma-Aldrich), 10 mM MOPS, 100 units/ml penicillin, and 100 µg/ml streptomycin (Gemini) at 37 °C and 5 % CO₂. The 3A9 T-cell hybridoma line (ATCC #CRL-3293) was cultured in DMEM (ATCC) supplemented with 5 % FBS (Thermo Fisher Scientific), 0.05 mM 2-mercaptoethanol (Sigma-Aldrich) at 37 °C and 5 % CO_2. ATCC follows the highest manufacturing standards and uses the most reliable procedures to verify and authenticate every cell line and to ensure there is no mycoplasma contamination.

Latex NH₂-beads (3 μm diameter, 3.5 × 10⁸ beads/preparation, Polysciences) were activated with 8 % glutaraldehyde in 0.5 ml PBS for 120 min under rotation at room temperature, washed with PBS, and incubated overnight with equal molar amounts of F(ab’)₂ goat-anti-mouse IgM+ G (αM, 20 μg/ml, Jackson ImmunoResearch Laboratories), hen egg lysozyme (HEL, 5.8 μg/ml, Sigma-Aldrich), duck egg lysozyme (DEL)-I (Langley et al., 2017), holo-transferrin (Tf, 32 μg/ml, Sigma-Aldrich), Alexa Fluor (AF) 488-conjugated Tf (AF488-Tf, 32 μg/ml, Thermo Fisher Scientific), or AF488-F(ab’)₂ goat-anti-mouse IgM+ G (AF488-αM, 20 μg/ml, Jackson ImmunoResearch Laboratories) in 1 ml PBS. Protein content determination (BCA, Thermo Fisher Scientific) of coupling solutions before and after bead incubation confirmed that similar molar amounts of protein were conjugated in each case. The beads were then blocked with PBS 1 % BSA for 30 min under rotation, washed to remove unconjugated proteins, counted in a Neubauer chamber and stored at 4 °C in PBS containing 1 % BSA and 5 % glycerol. Streptavidin-conjugated Yellow-Green latex beads (2 μm diameter, 5 × 10⁸ beads/preparation, Polysciences) were washed with 1 % BSA in PBS and incubated with Biotin-SP (long spacer)-conjugated Fab fragments of goat-anti-mouse IgG (H + L) (40 μg of biotinylated antibody/mg of beads, Jackson ImmunoResearch Laboratories) for 30 min at 4 °C, washed, counted in a Neubauer chamber and stored at 4 °C in PBS containing 1 % BSA and 5 % glycerol.

PLB were prepared as previously described (Dustin et al., 2007; Liu et al., 2012; Spillane and Tolar, 2017). Briefly, liposomes were generated from 5 mM 1,2-dioleoyl-sn-glycero-3-phosphocholine plus 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-cap-biotin (Avanti Polar Lipids) at a 100:1 molar ratio by sonication. Eight-well coverslip chambers (Lab-Tek) were incubated with liposomes for 20 min at room temperature and washed with PBS. The chambers were then incubated with 1 μg/ml streptavidin (Jackson ImmunoResearch Laboratories) for 10 min, washed, and incubated with 10 μg/ml mono-biotinylated Fab’ goat-anti-IgM+ G (αM-PLB) (Liu et al., 2012) or the same molar amount of biotinylated Tf (16 μg/ml, Sigma-Aldrich) (Tf-PLB) for 10 min at room temperature.

COS-7 cells were transiently transfected with mHEL-GFP (Batista et al., 2001) (plasmid kindly provided by Dr. Michael Gold, University of British Columbia) using Lipofectamine 3000 (Thermo Fisher Scientific) and a published protocol (Wang et al., 2018a), and used for experiments 24 hr post-transfection.

Mouse splenic B-cells were incubated with beads coated with αM, HEL, DEL-I or Tf in DMEM containing 6 mg/ml BSA (DMEM-BSA) at a cell:bead ratio of 1:2 (or as indicated), or with soluble F(ab’)₂ goat-anti-mouse IgM+ G (sαM, 0.5 μg/ml) for 30 min at 37 °C with 5 % CO₂. Propidium iodide (PI, Sigma-Aldrich) was present during the 37 °C incubation as an indicator of PM permeabilization. Cells were then analyzed by flow cytometry (BD FACSCanto II) at 10,000 cell counts/sample. Bead-bound cells were identified based on their forward- (FSC) and side-scatter (SSC) properties and on fluorescence intensity (FI) when using fluorescent beads (Figure 1—figure supplement 2). The percentages of PI-positive (PI+) cells among the bead-bound cell populations were quantified using FlowJo 10.1 software.

To assess PM permeabilization by protein-coated beads, mouse splenic B-cells or a B-cell line (A20) were incubated for 30 min at 4 °C in 35 mm glass-bottom dishes (MatTek) coated with poly-lysine and then with protein-coated beads at a cell:bead ratio of 1:2 for another 30 min at 4 °C. Cells were washed with DMEM-BSA and imaged in a Live Cell System chamber (Pathology Devices) at 37 °C with 5 % CO₂ in the presence of 50 μg/ml PI (Sigma-Aldrich) with or without 50 µM blebbistatin (Sigma-Aldrich). Images were acquired for 60 min at one frame/15–30 s using a spinning disk confocal microscope (UltraVIEW VoX, PerkinElmer with a 63 × 1.4 N.A. oil objective). Images were analyzed using Volocity Suite (PerkinElmer) and NIH ImageJ. More than 200 cells from three independent experiments were analyzed for each condition.

To assess PM permeabilization after binding to ligand-coated PLB, splenic B-cells were incubated with FM1-43FX or FM4-64FX (Thermo Fisher Scientific) in DMEM-BSA for 5 min at 4 °C, added to coverslip chambers containing mono-biotinylated Fab’ goat-anti-IgM+ G or biotinylated Tf tethered to PLB, and imaged immediately at 37 °C with 5 % CO₂ using a spinning disk confocal microscope (UltraVIEW VoX, PerkinElmer with a 63 × 1.4 N.A. oil objective) with or without 50 μg/ml PI and/or 10 µg/ml FM1-43FX or FM4-64FX (Thermo Fisher Scientific). Images were acquired at one frame/6–10 s and analyzed using Volocity (PerkinElmer) and NIH ImageJ. For quantitative analysis, the mean fluorescence intensity (MFI) of FM1-43FX or FM4-64FX in a defined area was measured using Volocity (PerkinElmer). More than 270 cells from three independent experiments were analyzed for each condition. For 4 hr videos, DMEM without phenol red containing 2 % FBS was used, and images were acquired at one frame/30 s in the presence of PI (50 μg/ml).

PM permeabilization was also assessed using Ponceau 4R-mediated quenching of a cytosolic fluorescent dye. B-cells were pre-stained with 1 µM CFSE (Thermo Fisher Scientific) for 10 min at 37 °C, washed with DMEM, incubated with αM- or Tf -PLB and analyzed in a spinning disk confocal microscope (UltraVIEW VoX, PerkinElmer with a 40 × 1.4 N.A. oil objective) in the presence or absence of 1 mM Ponceau 4R (Sigma-Aldrich). More than 480 cells from four independent experiments were analyzed for each condition. To validate this method, cells pre-stained with CFSE were incubated with or without 800 ng/ml SLO in the presence or absence of 1 mM Ponceau 4R (Sigma-Aldrich) for 10 min and analyzed by flow cytometry (BD FACSCanto II) at 10,000 cell counts/sample.

To assess the ability of antigen exposed on the surface of mammalian cells to permeabilize B-cells, COS-7 cells mock-transfected or transfected with mHEL-GFP were seeded on fibronectin-coated coverslips and cultured for 24 hr. WT or MD4 B-cells pre-stained with AF674-conjugated Fab fragments of donkey-anti–mouse IgM+ G (Jackson ImmunoResearch Laboratories) were then added to the COS-7 cells in the presence of 50 μg/ml PI and imaged immediately at 37 °C with 5 % CO₂ using a spinning disk confocal microscope (UltraVIEW VoX, PerkinElmer with a 40 × 1.3 N.A. oil objective). Images were acquired at one frame/20 s and analyzed using NIH ImageJ software. More than 240 cells from three independent experiments were analyzed for each condition.

Splenic B-cells were pretreated or not with 1 µM staurosporine (Abcam) for 24 hr at 37 °C in DMEM-BSA to induce apoptosis (Diaz et al., 2004), exposed to αM- or Tf-beads for 30 min at 37 °C, washed, fixed with 4 % paraformaldehyde (PFA), blocked with 1 % BSA, and permeabilized with 0.05 % saponin. Cells were then incubated with antibodies specific for cleaved caspase-3 (Asp175) (Cell Signaling Technology) followed by AF488 donkey-anti-rabbit IgG (Life Technologies) and analyzed by flow cytometry (BD FACSCanto II) at 10,000 cell counts/sample. The percentages of cells with cleaved caspase-3 staining were determined using FlowJo 10.1 software.

BCR signaling was analyzed using both flow cytometry and western blotting. For flow cytometry assays, splenic B-cells from MD4 mice were pretreated or not with 5 μM of the Src kinase inhibitor PP2 (Millipore) (Cheng et al., 2001) for 30 min at 37 °C (conditions selected not to cause B-cell toxicity) and then incubated with HEL-beads in the presence or not of the inhibitor at 37 °C for 30 min. Cells were fixed with 4 % PFA, permeabilized with 0.05 % saponin, incubated with mouse anti-phosphotyrosine mAb (4G10, Millipore) followed by AF488-goat-anti-mouse IgG_2b (Thermo Fisher Scientific) secondary antibodies, and analyzed by flow cytometry (BD FACSCanto II) at 10,000 cell counts/sample. The data were analyzed using FlowJo 10.1 software.

For western blot assays, splenic B-cells from MD4 mice were pretreated or not with 10 nM of the BTK inhibitor AVL-292 (Selleckchem) (Aalipour and Advani, 2013) for 30 min at 37 °C (conditions selected not to cause B-cell toxicity) and incubated with HEL-beads in the presence or not of the inhibitor at 37 °C for 30 min. Cells were then lysed using RIPA buffer (150 mM NaCl₂, 1 % of NP40, 0.5 % Sodium deoxycholate, 0.1 % SDS, 50 mM Tris, pH 8.0) containing protease and phosphatase inhibitors (50 mM NaF, 1 mM Na₃VO₄ and 10 mM Na₄P₂O₇) at 4 °C. Cell lysates were run in 4–20% gradient SDS-PAGE gels (Bio-Rad) (5 × 10⁶ cells/ lane) and transferred (Bio-Rad Trans-Blot transfer system) to PVDF membranes (Millipore). The membranes were blotted with rabbit anti-phospho-BTK (pBTK; Abcam) or anti-BTK (Cell Signaling Technology) antibodies followed by HRP-conjugated anti-rabbit antibodies (Jackson ImmunoResearch Laboratories) and visualization using ECL substrate (Bio-Rad) and imaging (iBright FL-1500), (Thermo Fisher Scientific).

To check if signaling affected PM permeabilization, splenic B-cells from MD4 mice were pretreated or not with 5 μM PP2 (Cheng et al., 2001) or 10 nM AVL-292 (Aalipour and Advani, 2013) for 30 min at 37 °C and then incubated with HEL-beads in the presence or not of the inhibitor and 50 μg/ml PI (Sigma-Aldrich) at 37 °C for 30 min. The percentage of PI+ cells was expressed relative to the untreated condition.

BCRs on the surface of mouse splenic B-cells were stained with Cy3-Fab donkey-anti–mouse IgM+ G (Jackson ImmunoResearch Laboratories) for 30 min at 4 °C. Cells were then incubated with αM- or Tf-beads at 4 °C for 30 min and 37 °C for different lengths of time. Cells were fixed with 4 % PFA, permeabilized with 0.05 % saponin, and incubated with rabbit anti-phosphorylated myosin light chain 2 (pMLC2) antibodies (Cell Signaling Technology) to label activated NMII (Bresnick, 1999), followed by AF633-goat-anti-rabbit IgG (Invitrogen). Cells were analyzed by confocal fluorescence microscopy (Zeiss LSM710 with a 63 × 1.4 N.A. oil objective). The percentages of cells with polarization of surface labeled BCRs and activated NMII towards bead-binding sites were quantified by visual inspection. More than 300 cells from three independent experiments were analyzed for each condition.

Mouse splenic B-cells were pretreated or not with 12 μM bromoenol lactone (BEL, Sigma-Aldrich) in DMEM-BSA for 30 min at 37 °C before and during assays, to inhibit lysosomal exocytosis and PM repair (Fensome-Green et al., 2007). Cells were then incubated with αM-beads (1:2 cell:bead ratio) with or without inhibitors at 4 °C for 5 min and 37 °C for 30 min in the presence of FM4-64FX (Thermo Fisher Scientific) to stain wounded cells. Cells were then incubated with SYTOX Blue nucleic acid stain (300 nM, Invitrogen) at 4 °C for 10 min to stain cells that failed to repair PM wounds during the 30 min incubation. Cells were analyzed by flow cytometry (BD FACSCanto II) at 10,000 cell counts/sample. Cells that were FM4-64FX positive but SYTOX Blue negative were identified as permeabilized cells that resealed. The percentages of resealed cells among all bead-bound permeabilized cells were quantified using FlowJo 10.1 software.

To assess the resealing capacity of B-cells permeabilized by ligand-coated PLB using live cell imaging, splenic B-cells were incubated with SYTOX Green (Thermo Fisher Scientific) in DMEM-BSA for 5 min at 4 °C and added to coverslip chambers containing mono-biotinylated Fab’ goat-anti-IgM+ G or biotinylated Tf tethered to PLB. Cells were imaged at one frame/30 s for 4 hr at 37 °C with 5 % CO₂ using a spinning disk confocal microscope (UltraVIEW VoX, PerkinElmer with a 63 × 1.4 N.A. oil objective), followed by addition of 50 μg/ml PI (Thermo Fisher Scientific) at the end of the assay and final image acquisition.

Surface BCRs of splenic B-cells were labeled with Cy5-Fab donkey-anti mouse IgG (Jackson ImmunoResearch) at 4 °C for 30 min. Cells were incubated with αM-PLB in the presence of FM 4–64 FX (Thermo Fisher Scientific) and imaged immediately at 37 °C with 5 % CO₂ using a spinning disk confocal microscope (UltraVIEW VoX, PerkinElmer with a 60 × 1.4 N.A. oil objective). Images were acquired at one frame/20 s for 60 min and analyzed using a custom-made MATLAB script (MathWorks) and NIH ImageJ software. BCR polarization was analyzed using maximal projection of XZ images and quantified by the MFI ratio between defined regions within the bottom half (closer to PLB) and the top half (away from PLB) of individual cells. Cells with bottom to top ratios ≥ 2 were considered polarized. More than 20 cells from three independent experiments were analyzed.

To detect LIMP-2 exposed on the cell surface, splenic B-cells (C57BL/6 or MD4) were incubated with αM-, HEL-, DEL-I or Tf-beads for 30 min at 37 °C, cooled to 4 °C, and incubated with rabbit-anti-LIMP-2 antibodies (Sigma-Aldrich) for 60 min at 4 °C. Cells were then washed and fixed with 4 % PFA, washed, blocked with 1 % BSA in PBS and incubated with AF488 donkey-anti-rabbit IgG (Life Technologies) secondary antibodies. For intracellular LIMP-2 staining, B-cells were fixed with 4 % PFA, washed, permeabilized with 0.05 % saponin for 20 min, and incubated with rabbit anti-LIMP-2 antibodies followed by AF488 donkey-anti-rabbit IgG. Flow cytometry (BD FACSCanto II) was performed at 10,000 cell counts/sample. Cells were also analyzed by confocal fluorescence microscopy (Leica SPX5 with a 63 × 1.4 N.A. oil objective). Polarization of LIMP-2 toward bound beads was quantified by calculating the fluorescence intensity ratio (FIR) of anti-LIMP-2 at the B-cell-bead contact site relative to the opposite side of the cell PM, using NIH ImageJ and a custom-made MATLAB script (MathWorks).

Individual events of lysosome exocytosis were captured using total internal reflection fluorescence (TIRF). Splenic B-cells were preloaded with SiR-Lysosome (1 µM, Cytoskeleton) in the presence of verapamil (10 µM, Cytoskeleton) for 30 min at 37 °C. Cells were added to coverslip chambers containing mono-biotinylated Fab’ goat anti-IgM+ G tethered to PLB and imaged at 37 °C with 5 % CO₂ in the presence of PI (50 µg/ml, Sigma-Aldrich) using a TIRF microscope (NIKON Eclipse Ti-E TIRF, 63 × 1.49 NA oil objective). Images were acquired at eight frames/s during 15–20 min intervals of the 45 min incubation and analyzed using NIH ImageJ and Nikon NIS Elements software. Increases in the FI of individual SiR-Lysosome puncta (reflecting lysosome movement within the TIRF evanescent field toward the PM in contact with PLB) followed by sharp decreases within a period of 1–2 s (corresponding to a loss of the SiR-Lysosome signal upon PM fusion) were scored as exocytosis events (Jaiswal et al., 2002). More than 20 cells were analyzed in four independent experiments.

Mouse splenic B-cells were incubated with F(ab’)₂ goat-anti-mouse IgM+ G (10 μg/ml, Jackson ImmunoResearch Laboratories) for 10 min, followed by AF674-conjugated donkey-anti-goat (10 μg/ml, Invitrogen) for 30 min at 4 °C in coverslip chambers, to label and crosslink surface BCRs. FM1-43FX (10 µg/ml, Thermo Fisher Scientific) was added at the last 5 min of the 30 min incubation at 4 °C. Cells were washed and imaged at 37 °C with 5 % CO₂ in the presence of 50 μg/ml PI and 10 µg/ml FM1-43FX using a spinning disk confocal microscope (UltraVIEW VoX, PerkinElmer with a 63 × 1.4 N.A. oil objective). Images were acquired at one frame/30 s for 60 min and analyzed using Volocity (PerkinElmer).

Mouse splenic B-cells were pre-treated or not with 12 μM bromoenol lactone (BEL, Sigma-Aldrich) in DMEM-BSA for 30 min at 37 °C and then incubated with Tf-beads (1:2 cell-bead ratio) with or without the inhibitors at 37 °C for 30 min in the presence of SYTOX Blue (300 nM, Invitrogen). Cells were analyzed by flow cytometry (BD FACSCanto II) at 10,000 cell counts/sample. Bead-bound cells and SYTOX-Blue-positive cells were gated. The percentages of SYTOX Blue positive cells among all bead-bound permeabilized cells were quantified using FlowJo 10.1 software.

For live imaging of antigen internalization, splenic B-cells were incubated with AF488-αM-beads (1:4 cell:bead ratio) in the presence of 1 μM SiR-Lysosome and 10 μM verapamil for 30 min at 4 °C, washed with DMEM-BSA and imaged by confocal fluorescence microscopy (Leica SPX5 with a 63 × 1.4 N.A. oil objective) for 60 min at one frame/min at 37 °C. Live time-lapse images were analyzed using NIH ImageJ.

For fixed cell imaging, splenic B-cells were pretreated or not with 50 µM blebbistatin on poly-lysine coated slides for 30 min at 4 °C and incubated with AF488-αM beads or AF488-Tf-beads at 37 °C for varying lengths of time in the presence or not of 50 μM blebbistatin. After fixation with 4 % PFA, cells were imaged by confocal fluorescence microscopy (Zeiss LSM710 with a 63 × 1.4 N.A. oil objective). Percentages of cells with intracellularly located AF488-αM puncta were determined by visual inspection of images. More than 200 cells from three independent experiments were analyzed for each condition.

For live imaging of B-cells interacting with PLB, mouse splenic B-cells were added to coverslip chambers containing PLB coated with AF488-conjugated mono-biotinylated Fab’ goat-anti-mouse IgM+ G and incubated at 37 °C with 5 % CO₂ in the presence of 10 µg/ml FM 4–64 FX (Thermo Fisher Scientific) for varying lengths of time. Samples were then moved to 4 °C for 5 min and immediately imaged using a confocal microscope (Leica SPX5 with a 63 × 1.4 N.A. oil objective). Internalization of antigen was quantified by determining the percentages of cells with intracellularly-located AF488-Fab’ goat-anti-mouse IgM+ G puncta in each field and by measuring the AF488 FI associated with intracellular puncta in individual cells, using a custom-made MATLAB (MathWorks) script. Cells with high FM staining were identified as wounded and those with low FM staining as unwounded. More than 15 fields or ~90 cells from three independent experiments (high or low FM staining) were analyzed for each condition.

To detect antigen presentation to T-cells, splenic B-cells from F1 mice of a crossing between B10.BR-H2^k2 H2-T18^a/SgSnJJrep and MD4 mice were co-cultured with 3A9 T-cell hybridoma cells (ATCC CRL-3293) at equal concentrations (3.75 × 10⁶ cells/ml). Cells were incubated in DMEM supplemented with 5 % FBS and 0.05 mM 2-mercaptoethanol for 72 hr in the presence or not of soluble HEL or DEL-I (10 µg/ml), or of beads coated with HEL, DEL-I or Tf (1:4 cell: bead ratio). After incubation, the concentration of IL-2 in the supernatant was measured using an IL-2 ELISA kit (Biolegend).

Statistical significance was assessed using unpaired, two-tailed Student’s t-tests (Prism - GraphPad software) when only two groups were compared, and one-way ANOVA (parametric) or Kruskal-Wallis (non-parametric) when three or more groups were compared. All data were presented as the mean ± SD (standard deviation).