The culturing of control and plcg2kd HL60 cells was as previously reported. Briefly, cells were maintained in RPMI 1640 culture medium [RPMI 1640 medium with 20% (v/v) fetal bovine serum and 25 mM HEPES (Quality Biological, Inc. Gaithersburg, MD)]. HL60 cells were differentiated in RPMI 1640 culture medium containing 1.3% DMSO for 5 days before the experiments. The cells were incubated at 37°C in a humidified 5% CO₂ atmosphere.

The DNA vectors of turboGFP (tGFP)-human CAPRI, active Ras sensor (active Ras binding domain of human Raf1 tagged with mRFP, RBD-RFP), PIP₃ biosensor (PH-GFP), and PM markers (CAAX-mCherry) were from Addgene (Cambridge, MA). The F-actin sensor F-tractin-GFP was obtained from John Hammer (24). The transfection procedure was as previously described (25). Briefly, 2 × 10⁶ cells were centrifuged at 100 × g for 10 min and resuspended in a mixture of 80 µL nucleofection solution V and 20 μL supplement I at room temperature. Six micrograms of plasmid DNA encoding the cDNA of the desired proteins were used for a single transfection reaction using program T-019 on the Amaxa Nucleofector II (Lonza, MD).

Cells were incubated with 100 ng/ml Fluo4 (Invitrogen, Carlsbad, CA) at 37°C for 30 min, washed with RPMI 1640 medium with 25 mM HEPES twice to remove the unstained Fluo-4, and then subjected to the experiments.

Briefly, cells were starved in RPMI medium containing 25 mM HEPES at 37°C for 3 hours. Cells were then collected, resuspended at 2×10⁷ cells/ml, and transfer to a medical cup under constant shaking at 200 rpm for 3 min at room temperature. Cells were stimulated with fMLP at the indicated final concentrations. At the indicated time points before or after stimulation, 100 μl aliquots of the cells were taken from the shaking medical cup to ensure equal number of cells were subject to the subsequent steps. Aliquots were then mixed with immunoprecipitation buffer (IB), including 0.25% NP40, 10 mM Tris (pH7.5) buffer, 150 mM NaCl, 1 mM Na3VO4, 10 mM NaF, and 1X proteinase inhibitor (Rhoche, Basel, Switzerland). The mixtures were incubated on ice for 30 min and then centrifuged at 100,000 × g at 4 °C for 30 min. The supernatants were incubated with agarose beads conjugated with RBD (active Ras binding domain of human Raf1) (Cytoskeleton, Inc. Denver, CO) at 4°C for 2 hours. The agarose beads were washed three times with IB. The protein on the beads was eluted by mixing with 25 μl 2X sample loading buffer (SLD) (Quality Biological Inc, Gaithersburg, MD). The supernatants and eluted proteins were subjected to western blot detection of the indicated proteins.

Cells were plated and allowed to adhere to the cover glass of a 4-well or a 1-well chamber (Nalge Nunc International, Naperville, IL) precoated with Fibronectin (Sigma Aldrich, Saint Louis, MO) for 10 min, and then covered with RPMI 1640 medium with 10% FBS and 25 mM HEPES. For confocal microscopy, cells were imaged using a Carl Zeiss Laser Scanning Microscope Zen 780 (Carl Zeiss, Thornwood, NY) with a Plan-Apochromat 60x/1.4 Oil DIC M27 objective. For the uniform-stimulation experiment of membrane translocation assays, the stimuli were directly delivered to the cells as previously described (25). To visualize the application of the stimuli, Alexa 633 or Alexa 488 was mixed with the fMLP stimuli at a final concentration of 1 μg/ml. For calcium response analysis, identical imaging parameters were used for both CTL and plcg2kd cells. Fluo-4 fluorescence intensity was measured in individual cells before and after stimulation, and the data were extracted and analyzed using GraphPad Prism. A two-tailed unpaired student t-test was used to calculate the p-value for the comparisons of peak responses between CTL and plcg2^kd cells. Statistical significance is indicated as follows: ns (not significant p > 0.1), *(p < 0.1), **(p < 0.01), ***(p < 0.001). The membrane translocation of the indicated protein was measured by the depletion of the interested protein in the cytoplasm as previously described (23). The data obtained were further analyzed with Microsoft Office Excel (Redmond, WA). For quantitative analysis of membrane translocation dynamics of the indicated molecules, the cytosolic depletion of the indicated molecule was measured. Regions of interest (ROIs) in the cytoplasm (avoiding the nucleus area as much as possible) were within the cells throughout the time period of the measurements. The periphery of the cells was marked by the membrane markers. For data analysis, to normalize the effect of photobleaching during data acquisition, the intensity of ROIs in the cytoplasm was divided by the intensity of whole cells at each given time point. To normalize the effect of morphological change during the time period, the above resulting data were divided by the intensity of ROIs in the PM marker channel in the case PM marker was simultaneously monitored. Lastly, the resulting data were divided by that at time 0 s; consequently, the relative intensity of any cells at time 0 s became 1. The graph of mean ± SD is shown.

The procedure was as previously reported (26). Briefly, differentiated cells were loaded onto fibronectin-coated 4-µm EZ-TAXIScan chambers. The chemoattractants at the indicated concentrations were added to the other side of the well across the terrace that the cells chemotax through. The cells migrated for 30 min at 37°C. Images were taken at 30-s intervals. For chemotaxis parameter measurements, 20 cells in each group were analyzed with DIAS software (27). The bar graphs of chemotaxis parameters in mean and SD were plotted with Microsoft Office Excel (Redmond, WA).

Calcium oscillations are a ubiquitous signaling phenomenon that occurs spontaneously or are triggered by receptor-ligand binding. In this study, we distinguish these two mechanisms by referring to the non-ligand-induced [Ca²⁺] increase as spontaneous calcium oscillation (or calcium oscillation), and the ligand-induced one as calcium response. We previously reported fMLP-induced calcium responses in control (CTL) HL60 cells and plcg2 stably knocked-down (plcg2^kd ) HL60 cells (20). Upon saturating fMLP stimulation, both CTL and plcg2^kd cells exhibited comparable peak calcium. However, plcg2^kd cells displayed a significantly shorter duration of calcium elevation and a marked reduction in secondary sporadic calcium spikes, suggesting that PLCγ2 plays a role in sustaining calcium signaling. To further dissect the function of PLCγ2 in calcium response and spontaneous calcium oscillation, we monitored calcium dynamics in both CTL and plcg2^kd cells upon stimulation with three concentrations of fMLP (10 nM, 1 nM, and 0.1 nM). plcg2^kd cells exhibit approximately 90% knockdown efficiency of PLCγ2, as previously shown (20). To visualize stimulus application, especially at subsensitive concentrations, we co-applied fMLP with Alexa 633 (red) and monitored calcium signaling using the Fluo-4 indicator (green) ( Figure 1 ). At 10 nM and 1 nM fMLP, both CTL and plcg2kd cells exhibited calcium responses ( Figures 1A, C , Supplementary Video S1, S2 ). CTL cells typically show the initial synchronized calcium rise—likely corresponding to the chemoattractant-induced response—and the subsequent sporadic, asynchronous calcium oscillation observed in individual cells. However, the amplitude and duration of calcium signals were significantly reduced in plcg2^kd cells ( Figures 1B, D ). Calcium response in each individual CTL and plcg2^kd cell further confirms the above observation ( Supplementary Figure S1 ). indicating that PLCγ2 contributes to both the amplitude and persistence of GPCR-mediated calcium responses upon stimuli at a moderate ligand concentration. Upon 0.1 nM fMLP stimulation (a subsensitive concentration for CTL cells in the previous report) (23), neither cell type show synchronized calcium response ( Figure 1E , Supplementary Video S3 ) (23). Importantly, CTL cells display sporadic calcium responses and spontaneous calcium oscillation as previously described (1). In contrast, plcg2^kd cells rarely showed sporadic calcium activity (a representative responses of multiple plcg2kd cells shown in Figure 1E , lower panel), indicating a critical role of PLCγ2 in maintaining spontaneous calcium oscillation under resting conditions. Together, these findings demonstrate that PLCγ2 not only constitutes the GPCR-mediated calcium response but also mediates the spontaneous calcium oscillation in the resting neutrophils.

The direct connection between calcium oscillation and cell sensitivity to chemoattractant stimulation is unclear. We have previously demonstrated that CAPRI mediates the deactivation of the GPCR-mediated Ras signaling to facilitate Ras adaptation in human neutrophils (23). In resting neutrophils, CAPRI is predominantly cytosolic; however, a fraction of CAPRI localizes in the plasma membrane (PM), where it regulates basal Ras activity and thereby modulate cell sensitivity. The PM localization of CAPRI depends on its C2-domain and a proper increase in intracellular calcium ([Ca²⁺]). Nonetheless, whether PLCγ2 contributes to the calcium increase required for CAPRI membrane recruitment in resting cells has not been fully elucidated. As previously reported (23), migrating CTL cells actively recruit CAPRI-GFP to the leading fronts, whileas plcg2^kd cells rarely show CAPRI-GFP enrichment at the protrusion sites ( Figure 2A ). Upon 10 nM fMLP stimulation, CTL cells display a robust membrane translocation of CAPRI-GFP ( Figure 2A , upper panel, and Supplementary Video S4 , upper panel). In contrast, plcg2^kd cells display significantly reduced PM translocation of CAPRI-GFP ( Supplementary Video S4 , lower panel), consistent with the previous report. Upon 0.1 nM fMLP stimulation, neither CTL nor plcg2^kd cells show detectable CAPRI-GFP membrane translocation ( Figure 2C , Supplementary Video S5 ). While migrating CTL cells consistently localize CAPRI-GFP to the leading fronts, while plcg2^kd cells do not. We confirmed the above observation in many cells and proceeded to quantify CAPRI plasma membrane (PM) translocation in both CTL and plcg2^kd cells. To assess membrane translocation quantitatively, we measured cytosolic depletion of CAPRI-GFP as previously reported. Regions of interest (ROIs) were selected in the cytoplasm, avoiding the nucleus area whenever possible, and were tracked throughout imaging period. Due to the migratory behavior of the cells, most of the cells were not suitable for quantitative measurement over time. Therefore, we selected 4 to 5 cells that displayed typical cell response with minimal movement for quantitative analysis. The results shown in Figure 2B and 2D support our observations. Collectively, these results indicate that membrane targeting of CAPRI is impaired in plcg2^kd cells under both resting and stimulated conditions.

To investigate the consequence of impaired PM targeting of CAPRI in plcg2^kd cells, we biochemically assessed Ras activation using a pull-down assay in a large population of both CTL and plcg2^kd cells upon fMLP stimulation at different concentrations as previously reported (23). Same amount of CTL and plcg2kd cells were collected and basolated on ice for 10 min. Stimulation at a final concentration of 10 nM fMLP triggered a clear Ras activation in CTL cells, with a significantly stronger activation in plcg2^kd cells ( Figure 3A ). In contrast, 0.1 nM fMLP did not trigger notable Ras activation in CTL cells but induced a clear Ras activation in plcg2^kd cells, indicating increased sensitivity in the cells lacking PLCγ2. Densitometric analysis from three independent experiments ( Figure 3B ) supports this observation. The intensity of active Ras in CTL cells at time 0 s was normalized to 1, and values at other time points are presented as the ratio of intensity at the given time point (I_t) to that at time 0 (I₀). To further confirm the increased sensitivity of plcg2^kd cells, we monitored the temporospatial Ras activation using a live-cell confocal imaging with the active Ras probe (RBD-RFP, red) ( Figure 3C ). fMLP was co-applied with Alexa488 (green) to visualize the stimulus application. RBD-RFP was found to localize at the protrusion site of both resting CTL and plcg2^kd cells ( Supplementary Video S6 ). Upon 10 nM fMLP stimulation, CTL cells exhibited a robust PM translocation, plcg2^kd cells showed a significantly stronger and more sustained PM translocation of RBD-RFP, consistent with the previous report ( Figure 3C , upper panel, Supplementary Video S6 ) (20). When stimulated with 0.1 nM fMLP, CTL cells did not display a marked overall PM translocation of RBD-RFP; instead, RBD-RFP remained localized to the protrusion site or the leading front of a migrating cell ( Figure 3C , lower panel, Supplementary Video S7 , upper panel). In contrast, plcg2^kd cells showed a clear PM translocation of RBD-RFP followed by sustained localization at expanding protrusion sites upon the same low-dose fMLP stimulation ( Supplementary Video S7 , lower panel). Quantitative measurement of RBD-RFP membrane translocation across multiple cells further confirms the above observation ( Figure 3D ). In summary, plcg2kd cells exhibit enhanced Ras activation upon chemoattractant stimulation and respond to concentrations of fMLP that are subsensitive for CTL cells. These findings demonstrate that PLCγ2 is not only required for chemoattractant-induced Ras adaptation but also plays a critical role in regulating neutrophil sensitivity through CAPRI membrane recruitment.

PI₃Kγ, a direct effector of Ras, catalyzes the conversion of phosphatidylinositol (4,5)-bisphosphate (PI(4,5)P2, PIP₂) to phosphatidylinositol (3,4,5)-trisphosphate (PtdIns(3,4,5)P3, PIP₃) (2, 28, 29). PI₃Kγ activation recruits and activates the PIP₃-binding serine/threonine kinase Akt on the plasma membrane, which plays a critical role in neutrophil chemotaxis (3). To examine the consequences of the increased Ras activation in plcg2^kd cells, we next investigated PI₃K activation in both CTL and plcg2^kd cells. PI₃K activation was monitored by visualizing PIP₃ production using a PIP₃ biosensor, PH-GFP, through confocal microscopy. In the resting cells, PH-GFP (green) localizes predominantly in the cytosol ( Figure 4A ). It also colocalizes with a plasma membrane (PM) marker (red) at the protrusion sites as previous report (17), which appeared more pronounced in plcg2^kd cells. Upon 10 nM fMLP stimulation, CTL cells displayed a transient translocation of PH-GFP to the cell periphery, followed by sustained colocalization with the PM marker at protruding sites ( Figure 4A , upper panel, Supplementary Video S8 ). Upon the same stimulation, plcg2^kd cell shows a significantly stronger and more prolonged PH-GFP translocation to PM ( Figure 4A , lower panel; Supplementary Video S8 ). When stimulated with 0.1 nM fMLP, CTL cells did not show a clear global PM translocation of PH-GFP; instead, PH-GFP remained localized at the protrusion site of migrating cells ( Figure 4C , upper panel, Supplementary Video S9 ). In contrast, plcg2^kd cells exhibited a clear PM translocation of PH-GFP, followed by continuous localization in the expanding protrusion sites upon the same 0.1 nM fMLP stimulation, ( Figure 4C , lower panel, Supplementary Video S9 , Figure 4D ). We confirmed the above observation in many cells and proceeded to quantify PH-GFP PM translocation in both CTL and plcg2^kd cells. Cytosolic depletion of PH-GFP was measured using the same approach. ROIs in the cytoplasm were selected and tracked over time. To correct for photobleaching, cytoplasmic ROI intensity was normalized to the whole-cell intensity at each time point. To account for morphological changes, the values were further normalized to the intensity of the PM marker channel. Finally, all data were normalized to the intensity at time 0 s, so that the relative intensity at time 0 was set to 1. This quantitative analysis supports our observations ( Figures 4B, D ). plcg2^kd cells display a significantly stronger PM translocation of PH-GFP ( Supplementary Figure S2 ). In conclusion, plcg2kd cells exhibit elevated PI3K activation and are capable of responding to chemoattractant stimulation at concentrations that are subsensitive for CTL cells.

Neutrophils utilize GPCR/G protein complexes to regulate multiple signaling pathways that coordinate actin cytoskeleton dynamics and drive cell migration. To evaluate the role of PLCγ2 in chemoattractant GPCR-mediated actin assembly in neutrophils, we monitored actin polymerization using a fluorescent F-actin probe, F-tractin–GFP (green), in live cells via fluorescence microscopy ( Figure 5 ). In resting cells, F-tractin-GFP (green) localizes primarily in the cytosol and cortical regions, where it colocalizes with a plasma membrane (PM) marker (red) on the membrane and protrusion sites ( Figure 5A ). Upon 10 nM fMLP stimulation at 2 s, more F-tractin-GFP translocated to the cell cortex at around 10 to 40 s, then mostly returned to the cytosol at about 60 s, and then translocated to the leading front again at around 80 s in CTL cells ( Figure 5A , upper panel, Supplementary Video S10 ). In response to the same 10 nM fMLP stimulation, plcg2^kd cells displayed a continuous, persistent translocation of F-tractin-GFP and colocalized with the PM marker on the plasma membrane ( Figure 5A , lower panel, Supplementary Video S10 ). Using the same quantification method as in Figure 4B , we further quantified the actin polarization of CTL and plcg2^kd cells by the membrane translocation of F-tractin-GFP and confirmed that plcg2^kd cells display elevated and prolonged actin polymerization compared to CTL cells ( Figure 5B ). We further determined the actin polymerization of both CTL and plcg2^kd cells to 0.1 nM fMLP stimulation. In response to 0.1-nM fMLP stimulation, most CTL cells (∼90%) did not show the clear membrane translocation of F-tractin–GFP to the PM, while they showed cortex localization of F-tractin and PM marker ( Figure 5C , upper panel, Supplementary Video S11 ). In contrast, more than 80% of plcg2^kd cells showed the clear membrane translocation of F-tractin-GFP upon 0.1 nM fMLP stimulation ( Figure 5C , lower panel, Supplementary Video S11 ). Quantitative measurement of membrane translocation of F-tractin in CTL and plcg2^kd cells shows a normal oscillation of actin polymerization in CTL cells, while a clear actin polymerization in plcg2^kd cells upon 0.1 nM fMLP stimulation ( Figure 5D ). Together, these results demonstrate that plcg2kd cells are capable of initiating actin polymerization in response to chemoattractant stimulation at subsensitive concentrations and exhibit prolonged, elevated actin assembly in response to low-dose chemoattractant stimulation.

We found that capri^kd neutrophils, which lack Ras inhibitor CAPRI, display an increased sensitivity and elevated activation of Ras and its downstream effectors (23). More importantly, capri^kd neutrophils display an altered chemotaxis behavior: an improved chemotaxis in the gradients at subsensitive concentrations, a normal chemotaxis in the gradients at medium concentrations, and an impaired chemotaxis in the gradients at saturating concentrations. That is, neutrophils lacking CAPRI display an upshift in concentration range for chemotaxis (30). We have previously shown an impaired chemotaxis of plcg2^kd cells upon chemoattractant gradients at a saturating concentration (20). Next, we examined the chemotaxis behavior of CTL and plcg2^kd cells in the gradients of three chemoattractants at either medium (100 nM) or subsensitive (0.1 nM) concentrations ( Figure 6 , Supplementary Video S12 ). In the absence of a gradient, plcg2kd cells appeared to exhibit a broader random walk compared to CTL cells, although the difference was not statistically significant ( Figures 6A, B ). When exposed to gradients generated from a source at medium concentrations (100 nM), CTL and plcg2^kd cells displayed overall similar chemotaxis capability, although plcg2^kd cells displayed slightly decreased speed and total path length ( Figure 6B ). In the gradients generated from the sources of 0.1 nM, most CTL cells displayed random migration, while most plcg2^kd cells displayed a clear directed cell migration along the direction of the gradient. In conclusion, plcg2^kd cells display a concentration-dependent, altered chemotaxis behavior: an improved chemotaxis in the gradients at a subsensitive concentration, while a normal chemotaxis in the gradients at medium concentrations. Combined with the previous report (20), our results demonstrate that neutrophils lacking PLCγ2 display an upshift of the concentration ranges of diverse chemoattractants for an efficient chemotaxis.