Protein kinase C (PKC)-ε is required for membrane addition during IgG-mediated phagocytosis, but its role in this process is ill defined. Here, we performed high-resolution imaging, which reveals that PKC-ε exits the Golgi and enters phagosomes on vesicles that then fuse. TNF and PKC-ε colocalize at the Golgi and on vesicles that enter the phagosome. Loss of PKC-ε and TNF delivery upon nocodazole treatment confirmed vesicular transport on microtubules. That TNF+ vesicles were not delivered in macrophages from PKC-ε null mice, or upon dissociation of the Golgi-associated pool of PKC-ε, implies that Golgi-tethered PKC-ε is a driver of Golgi-to-phagosome trafficking. Finally, we established that the regulatory domain of PKC-ε is sufficient for delivery of TNF+ vesicles to the phagosome. These studies reveal a novel role for PKC-ε in focal exocytosis – its regulatory domain drives Golgi-derived vesicles to the phagosome, whereas catalytic activity is required for their fusion. This is one of the first examples of a PKC requirement for vesicular trafficking and describes a novel function for a PKC regulatory domain.

This article has an associated First Person interview with the first author of the paper.

Phagocytosis is the process by which innate immune cells, predominately neutrophils and macrophages, remove particulates from their environment. Phagocytosis is a highly conserved process utilized by lower organisms, such as amoeba and sponges, for nutrition. In higher organisms, phagocytosis has evolved as an immune-mediated mechanism for removal of pathogens, debris and senescent cells (Chen et al., 2007; Desjardins et al., 2005; Uribe-Querol and Rosales, 2020). IgG-mediated phagocytosis is central to resolution of infection (Ben Mkaddem et al., 2019; Uribe-Querol and Rosales, 2020; Zhang et al., 2010). During phagocytosis, IgG-opsonized pathogens engage Fcγ receptors (FcγRs), inducing receptor clustering and activation of downstream signaling networks to coordinate phagosome formation, closure and pathogen clearance. Phagosome formation requires the addition of membrane for pseudopod extension and phagosome maturation (Botelho and Grinstein, 2011; Cannon and Swanson, 1992; Jaumouille and Grinstein, 2016; Lee et al., 2007). The membrane necessary for pseudopod extension is recruited from internal sources such as the endoplasmic reticulum and Golgi (Aderem, 2002; Braun and Niedergang, 2006; Cannon and Swanson, 1992; Gerlach et al., 2020; Holevinsky and Nelson, 1998). Using patch clamping to quantify membrane addition, we previously reported that macrophages add approximately one-third of their surface area from internal sources during spreading on IgG surfaces (i.e. during frustrated phagocytosis; Wood et al., 2013). Work done by the Grinstein laboratory has revealed that vesicle associated membrane protein 3 (VAMP3) concentrates at the early phagosome, suggesting that targeted membrane delivery, or focal exocytosis, is involved in phagosome formation (Bajno et al., 2000; Marion et al., 2012).

Focal exocytosis is essential for various cellular processes, including cell division, migration and neurosecretion. The mechanisms underlying focal exocytosis have been well-studied in polarized cells such as neurons; however, little is known about this process in non-polarized cells. Phagocytosis, with its directed addition of membrane at phagosomes, provides a tractable model for studying focal exocytosis in non-polarized cells. Having previously demonstrated that Golgi-tethered protein kinase C-epsilon (PKC-ε; encoded by PRKCE) is required for membrane addition during FcγR-mediated phagocytosis (Hanes et al., 2017), this current work was undertaken to determine the role of PKC-ε in the delivery and fusion of vesicles into the phagosome.

Protein kinase Cs are serine-threonine kinases grouped into three families (classical, novel and atypical) based on their activators. They share a common domain structure, with a unique regulatory domain tethered to a homologous kinase domain by a flexible hinge. Within the regulatory domain is a pseudosubstrate region (PS) that binds to the active site, maintaining the enzyme in an inactive conformation, a C1 region that binds diacylglycerol (DAG), and, for Ca2+-dependent isoforms, a C2 Ca2+-binding region (Newton, 1997). Upon cell activation, they translocate to membranes, binding their activators, which results in a conformational change that releases the pseudosubstrate, exposing the active site for substrate phosphorylation (Newton, 1997). In resting cells, PKC-ε is predominately cytosolic (Larsen et al., 2000).

We previously identified PKC-ε, one of the novel isoforms, as a critical player in membrane mobilization during phagocytosis (Cheeseman et al., 2006; Newton, 1997). PKC-ε is unique in its substrate specificity as demonstrated in in vitro binding and chimeric studies (Pears et al., 1991; Schaap et al., 1989). We reported that chimeras of the PKC-ε regulatory domain linked to the PKC-δ catalytic domain concentrate at phagosomes but fail to promote phagocytosis, consistent with the unique substrate specificity of PKC-ε (Wood et al., 2013). Additionally, we reported that PKC-ε is tethered to the Golgi through binding of its pseudosubstrate domain to phosphatidylinositol-4-phosphate (PI4P) (Hanes et al., 2017). Dissociation of this Golgi-associated PKC-ε pool using the PI4-kinase inhibitor PIK93, or expression of the Golgi-targeted PI4-phosphatase human (h)Sac1-K2A (Sac1 is also known as SACM1L in mammals), abrogates PKC-ε concentration at the phagosome and slows the rate of phagocytosis, suggesting that Golgi-associated PKC-ε is involved in membrane mobilization. Patch-clamping of wild-type and PKC-ε-null macrophages during frustrated phagocytosis revealed that virtually all of the membrane added in response to FcγR ligation is dependent on PKC-ε (Wood et al., 2013). That dissociation of the Golgi-tethered pool of PKC-ε with PIK93 blocked this membrane addition established that there is a PKC-ε dependent Golgi-to-phagosome vesicular trafficking highway (Hanes et al., 2017). These data allude to a novel mechanism of PKC translocation in which PKC-ε translocates to its site of action from the Golgi rather than the cytosol.

In this study, we investigate the mechanism underlying PKC-ε trafficking from the Golgi for focal exocytosis during FcγR-mediated phagocytosis. Our results demonstrate that PKC-ε traffics on microtubule-associated vesicles from the Golgi to the forming phagosome. Furthermore, PKC-ε, more specifically the regulatory domain of PKC-ε, is required for Golgi-to-phagosome vesicle movement.

PKC-ε enters the phagosome on vesicles

Live imaging of PKC-ε–GFP-expressing macrophages during IgG-mediated phagocytosis revealed PKC-ε puncta exiting the Golgi (Fig. 1A; Movie 1). That the Golgi-associated pool of PKC-ε concentrates at phagosomes (Hanes et al., 2017), coupled with export of PKC-ε from the Golgi as puncta, implies that there is PKC-ε delivery on vesicles. To test this hypothesis, we employed total internal reflection fluorescence microscopy (TIRFM) to image macrophages as they spread on IgG surfaces (frustrated phagocytosis). Imaging the membrane, which is essentially a phagocytic synapse, distinguished vesicle appearance (puncta) at the membrane from cytosolic translocation (a gradual homogenous increase) (Fig. 1B; Movies 24). The pattern of PKC-ε–GFP appearance at the membrane was compared to that of a vesicle marker (VAMP3–GFP) and a cytosolic protein (Akt-PH–GFP). VAMP3–GFP is a transmembrane protein that is delivered to the phagosome on vesicles that fuse; Akt-PH–GFP (the PH domain of Akt) is a lipid reporter that translocates from the cytosol to bind phosphatidylinositol-3,4,5-trisphosphate (PIP3) at the phagosome and dissociates after phagosome closure (Bajno et al., 2000; Marshall et al., 2001; Niedergang and Chavrier, 2004; Yeo et al., 2015). Their patterns of entry were distinct. Akt-PH–GFP fluorescence at the membrane increased homogenously and was lost over time (Fig. 1B, top panel) while VAMP3–GFP appeared as distinct puncta (Fig. 1B, middle panel). PKC-ε–GFP also appeared as puncta (Fig. 1B, bottom panel), indicating that PKC-ε enters the phagosome on vesicles. The average size of PKC-ε punctum is 404±4.3 nm (mean±s.e.m. size quantified from ∼1000 vesicles from eight cells, three independent experiments, using Imaris image analysis software). A result of 400 nm is consistent with these being intracellular vesicles, which range from 100–500 nm (Tubbesing et al., 2020).

Fig. 1.

PKC-ε traffics on and enters the phagosome on vesicles. (A) BMDMs, transduced to express PKC-ε–GFP, were presented with 5 µm IgG-opsonized particles and phagocytosis was followed over time. Images were captured at 3 s intervals over a 10 min period by spinning-disc confocal microscopy. PKC-ε puncta exiting the Golgi are marked by arrows. Temporal-color coding was applied to images post-acquisition using FIJI to track puncta movement towards phagosomes (orange box) and compared to non-involved regions (green box). Shown is a representative event of n=5 (15 cells). (B) BMDMs expressing Akt-PH–GFP, VAMP3–GFP or PKC-ε–GFP were followed with time as they underwent frustrated phagocytosis. Images were captured by TIRFM every 8 s for 10 min. A pseudocolored heatmap was applied to images for optimal visualize of fluorescence intensity (bottom panels). Heatmap ranges from 2000 (blue) to 30,000 arbitrary units (red). Representative frames from time-lapse imaging (Movies 24) are presented; 0 time is set as the first frame in which fluorescence appears. Graph presents the quantification of punctum number for VAMP3 and PKC-ε expressing cells (n=3 independent experiments, at least 8 cells per condition). Each point represents one cell. (C) BMDMs expressing VAMP3–GFP (upper) or Akt-PH–GFP (lower). Cells were subjected to frustrated phagocytosis (5 min, 37°C), fixed, stained for endogenous PKC-ε, and imaged by TIRFM. The colocalization between PKC-ε and VAMP3–GFP or Akt-PH–GFP was quantified by means of a Pearson's correlation coefficient (graph); n=3 (at least 8 cells per condition), each data point represents a cell. Data are presented as mean±s.e.m. ***P<0.005 (two-tailed unpaired Student's t-test). Scale bars: 10 µm.

Fig. 1.

PKC-ε traffics on and enters the phagosome on vesicles. (A) BMDMs, transduced to express PKC-ε–GFP, were presented with 5 µm IgG-opsonized particles and phagocytosis was followed over time. Images were captured at 3 s intervals over a 10 min period by spinning-disc confocal microscopy. PKC-ε puncta exiting the Golgi are marked by arrows. Temporal-color coding was applied to images post-acquisition using FIJI to track puncta movement towards phagosomes (orange box) and compared to non-involved regions (green box). Shown is a representative event of n=5 (15 cells). (B) BMDMs expressing Akt-PH–GFP, VAMP3–GFP or PKC-ε–GFP were followed with time as they underwent frustrated phagocytosis. Images were captured by TIRFM every 8 s for 10 min. A pseudocolored heatmap was applied to images for optimal visualize of fluorescence intensity (bottom panels). Heatmap ranges from 2000 (blue) to 30,000 arbitrary units (red). Representative frames from time-lapse imaging (Movies 24) are presented; 0 time is set as the first frame in which fluorescence appears. Graph presents the quantification of punctum number for VAMP3 and PKC-ε expressing cells (n=3 independent experiments, at least 8 cells per condition). Each point represents one cell. (C) BMDMs expressing VAMP3–GFP (upper) or Akt-PH–GFP (lower). Cells were subjected to frustrated phagocytosis (5 min, 37°C), fixed, stained for endogenous PKC-ε, and imaged by TIRFM. The colocalization between PKC-ε and VAMP3–GFP or Akt-PH–GFP was quantified by means of a Pearson's correlation coefficient (graph); n=3 (at least 8 cells per condition), each data point represents a cell. Data are presented as mean±s.e.m. ***P<0.005 (two-tailed unpaired Student's t-test). Scale bars: 10 µm.

Because PKC-ε and VAMP3 present with a similar (punctate) pattern and number of puncta/cell (Fig. 1B), we asked whether they colocalize. Bone marrow-derived macrophages (BMDMs) expressing either VAMP3–GFP or Akt-PH–GFP were subjected to frustrated phagocytosis, fixed, and stained for endogenous PKC-ε (Fig. 1C). The Pearson's correlation coefficient revealed significantly higher colocalization between PKC-ε and VAMP3–GFP compared to PKC-ε and Akt-PH–GFP (Fig. 1C, graph). Colocalization of PKC-ε with a transmembrane protein confirms its presence on vesicles. Additionally, the punctate pattern of endogenous PKC-ε matched that of PKC-ε–GFP, validating that PKC-ε–GFP is a valid readout of its endogenous counterpart. VAMP3 was used here as a positive vesicle marker. It is not required for phagocytosis (Allen et al., 2002) nor is it required for membrane addition or spreading during frustrated phagocytosis (Fig. S1) as these metrics are the same in wild-type and VAMP3−/− BMDMs. Additionally, VAMP3 is not required for trafficking of PKC-ε+ vesicles as PKC-ε puncta were delivered to the phagocytic synapse in VAMP3−/− cells (Fig. S1).

If the PKC-ε that appears at the phagosome originates from the trans-Golgi network (TGN), then depletion of the Golgi pool should reduce PKC-ε puncta at the membrane. PIK93, a PI4K inhibitor that depletes Golgi-tethered PKC-ε (Hanes et al., 2017), reduced the number of PKC-ε–GFP puncta at the phagosome 6-fold compared with untreated controls (Fig. 2). The effect was reversible, as puncta delivery recovered upon PIK93 washout. The PIK93 effect was not due to the disruption of the Golgi integrity, as it did not alter the pattern of GM130 (also known as GOLGA2, a cis-Golgi marker) nor of Golgin-245 (also known as GOLGA4, a TGN marker) staining (Fig. S2). Notably, the GFP signal at the membrane in PIK93-treated cells showed very few puncta and no substantive increase in membrane intensity, supporting the conclusion that the puncta at the membrane are derived from the Golgi and there is little to no contribution from the cytosolic PKC-ε pool.

Fig. 2.

PKC-ε delivery is reversibly inhibited by nocodazole or PIK93. BMDMs expressing PKC-ε–GFP were pre-treated with DMSO (solvent control), PIK93 (100 nM) or nocodazole (10 µM) for 30 min, subjected to frustrated phagocytosis for 30 min, and then fixed. For washouts, the inhibitor-containing medium was replaced with HBSS++ for 5 min prior to fixation. A fire LUT was applied for optimal visualization of puncta. Inserts represent GFP expression. Images were collected by TIRFM and the number of puncta was quantified by FIJI software (see lower DMSO image). Data are presented as number of puncta/cell area to account for differences in cell spreading; n=3 (at least 39 cells per condition), each data point represents a cell. Data are presented as mean±s.e.m. ****P<0.0001, ***P<0.001 (one-way ANOVA with Bonferroni post-test). Scale bars: 10 μm.

Fig. 2.

PKC-ε delivery is reversibly inhibited by nocodazole or PIK93. BMDMs expressing PKC-ε–GFP were pre-treated with DMSO (solvent control), PIK93 (100 nM) or nocodazole (10 µM) for 30 min, subjected to frustrated phagocytosis for 30 min, and then fixed. For washouts, the inhibitor-containing medium was replaced with HBSS++ for 5 min prior to fixation. A fire LUT was applied for optimal visualization of puncta. Inserts represent GFP expression. Images were collected by TIRFM and the number of puncta was quantified by FIJI software (see lower DMSO image). Data are presented as number of puncta/cell area to account for differences in cell spreading; n=3 (at least 39 cells per condition), each data point represents a cell. Data are presented as mean±s.e.m. ****P<0.0001, ***P<0.001 (one-way ANOVA with Bonferroni post-test). Scale bars: 10 μm.

If PKC-ε transits from the TGN to the phagosome on vesicles, microtubules likely serve as the highway for their transport. Consistent with this, we reported that the microtubule-disrupting agent nocodazole prevents membrane addition in response to FcγR ligation (Hanes et al., 2017). But does it block PKC-ε vesicle delivery? Indeed, fewer PKC-ε+ vesicles were delivered in nocodazole-treated BMDMs, an effect that was rescued upon drug washout (Fig. 2, right panel). Together, these data provide the first evidence that a PKC, any PKC, translocates to its site of action on a vesicle. Specifically, the results show that the PKC-ε that concentrates at phagosomes originates in the Golgi and transits on vesicles in a microtubule-dependent manner.

PKC-ε vesicles fuse into the forming phagosome

While the above experiments demonstrate that vesicles dock at the membrane, they do not address their fusion. If the Golgi pool of PKC-ε is required for membrane addition (Hanes et al., 2017), and if PKC-ε traffics to the phagosome on vesicles (above), we would predict that the PKC-ε+ vesicles that approach the membrane actually fuse. When a fluorescent vesicle approaches the membrane, its intensity as measured by TIRFM increases. If the vesicle fuses, the peak intensity will decrease as the fluorescent probe diffuses into the plane of the membrane, resulting in an increase in mean intensity. Thus, vesicle fusion is defined as a rise (approach) and fall (fusion) in maximum fluorescence intensity combined with an increase in overall fluorescence intensity (Jaiswal et al., 2009; Schmoranzer et al., 2000). Thus, we tracked PKC-ε–GFP as BMDMs attached and spread on IgG surfaces and used post-acquisition analytics of puncta regions of interest (ROIs) to determine the maximum (peak) and overall (mean) fluorescence intensity of vesicles with time (Fig. 3; Movie 5). Over time, the peak fluorescent intensity increased. The subsequent decrease in peak intensity was accompanied by an increase in overall fluorescence in the ROI (Fig. 3B,C, left panels; Fig. S3), fulfilling the established criteria for vesicle fusion (Jaiswal et al., 2009; Schmoranzer et al., 2000). Analysis of an equivalent ROI lacking vesicles in the same cell served as an internal control and showed no change in peak or mean fluorescence intensity (Fig. 3B,C, right panels; Fig. S3). These data demonstrate that PKC-ε+ vesicles fuse into the forming phagosome.

Fig. 3.

PKC-ε+ vesicles fuse with the membrane during frustrated phagocytosis. BMDMs were transduced to express PKC-ε–GFP and followed with time as they underwent frustrated phagocytosis. Images were captured by TIRFM every 8 s for 10 min. (A) A single timepoint image from Movie 5 is shown with an ROI containing a vesicle (solid box) and an equivalent area lacking vesicles (dotted box). Left, GFP with dashed lines highlighting edge of cell; right, image after the Fire LUT was applied. (B) A Fire LUT was applied for optimal visualization of fluorescence intensity of fusing and non-fusing (no vesicle) ROIs. Upper panels show fluorescence intensity as a LUT while lower panels use a 3D visualization. (C) Mean and peak fluorescence intensity was measured over time using FIJI software (a.u., arbitrary units) for the ROIs in B. The frame prior to vesicle appearance is marked as 0 s. n=2 (22 events). Scale bars: 10 μm.

Fig. 3.

PKC-ε+ vesicles fuse with the membrane during frustrated phagocytosis. BMDMs were transduced to express PKC-ε–GFP and followed with time as they underwent frustrated phagocytosis. Images were captured by TIRFM every 8 s for 10 min. (A) A single timepoint image from Movie 5 is shown with an ROI containing a vesicle (solid box) and an equivalent area lacking vesicles (dotted box). Left, GFP with dashed lines highlighting edge of cell; right, image after the Fire LUT was applied. (B) A Fire LUT was applied for optimal visualization of fluorescence intensity of fusing and non-fusing (no vesicle) ROIs. Upper panels show fluorescence intensity as a LUT while lower panels use a 3D visualization. (C) Mean and peak fluorescence intensity was measured over time using FIJI software (a.u., arbitrary units) for the ROIs in B. The frame prior to vesicle appearance is marked as 0 s. n=2 (22 events). Scale bars: 10 μm.

PKC-ε vesicles align along microtubules

The fact that nocodazole prevents membrane addition (Hanes et al., 2017) and delivery of PKC-ε vesicles to the phagosome (Fig. 2) suggests that PKC-ε+ vesicles travel on microtubules. Thus, we determined the extent to which PKC-ε aligns with microtubules in BMDMs in response to FcγR ligation. BMDMs were subjected to frustrated phagocytosis, fixed, stained for endogenous PKC-ε and α-tubulin, and imaged by Airyscan high resolution microscopy (Fig. 4A; Fig. S4A). Two different methods of alignment analysis were used. First, we traced either along α-tubulin tracks (Fig. 4A,B, solid lines 1–3; Fig. S4A) or across tracks (Fig. 4A,B, dashed lines a–c), to look for colocalization of PKC-ε puncta with α-tubulin. If PKC-ε were associated with MTs, we reasoned that we would see colocalization of (red) PKC-ε with (green) α-tubulin. Line plots of PKC-ε and α-tubulin fluorescence were generated on a single slice from a z-stack image using FIJI software. Regardless of how the lines were drawn, the majority of PKC-ε fluorescence colocalized with α-tubulin (Fig. 4B; Fig. S4B), suggesting that PKC-ε vesicles associate with microtubules. Secondly, whole-cell 3D reconstructed images were generated and, using Imaris software, spots and surfaces were constructed for PKC-ε and α-tubulin, respectively. Analysis revealed the ∼52% of PKC-ε puncta were aligned along microtubules (color-coded pink spots in Fig. 4C; Fig. S4C), consistent with PKC-ε puncta trafficking on microtubules.

Fig. 4.

PKC-ε vesicles align along microtubules. (A) BMDMs underwent frustrated phagocytosis for 30 min and then were fixed and stained for endogenous PKC-ε and microtubules (α-tubulin). Z-stacks were collected with Airyscan high resolution microscopy and 3D projections generated with Imaris software. (B) Line plot analysis of PKC-ε and α-tubulin fluorescent intensity alignment along microtubules (1–3) and across microtubules (a–b). Colored borders correspond to numbered lines in merged image (A) and dashed lines in A correspond to cross-section line plots. (C) 3D reconstruction of PKC-ε puncta and microtubules. Surfaces and spots were generated based on α-tubulin and PKC-ε fluorescence, respectively, using Imaris software. Alignment of PKC-ε spots along the microtubule surface was quantified with Imaris. Aligned PKC-ε puncta (pink spots) were defined as those within 0.2 μm of a microtubule. Non-aligned puncta (blue spots) are defined as those further than 0.2 μm. 52% of the PKC-ε spots were within 0.2 µm of a microtubule; n=3 (18 cells). Scale bars: 10 μm.

Fig. 4.

PKC-ε vesicles align along microtubules. (A) BMDMs underwent frustrated phagocytosis for 30 min and then were fixed and stained for endogenous PKC-ε and microtubules (α-tubulin). Z-stacks were collected with Airyscan high resolution microscopy and 3D projections generated with Imaris software. (B) Line plot analysis of PKC-ε and α-tubulin fluorescent intensity alignment along microtubules (1–3) and across microtubules (a–b). Colored borders correspond to numbered lines in merged image (A) and dashed lines in A correspond to cross-section line plots. (C) 3D reconstruction of PKC-ε puncta and microtubules. Surfaces and spots were generated based on α-tubulin and PKC-ε fluorescence, respectively, using Imaris software. Alignment of PKC-ε spots along the microtubule surface was quantified with Imaris. Aligned PKC-ε puncta (pink spots) were defined as those within 0.2 μm of a microtubule. Non-aligned puncta (blue spots) are defined as those further than 0.2 μm. 52% of the PKC-ε spots were within 0.2 µm of a microtubule; n=3 (18 cells). Scale bars: 10 μm.

While spreading on IgG-coated surfaces allows visualization of events occurring at the ‘phagosome’, IgG surfaces do not replicate the focal activation and cross-linking of FcγRs that occur during target ingestion. If PKC-ε traffics from the Golgi on microtubules, we would predict that PKC-ε+ vesicles preferentially localize on microtubules directed towards the phagosome. To test this, BMDMs were incubated with IgG-opsonized targets, fixed and stained for endogenous PKC-ε and α-tubulin (Fig. 5A; Fig. S5A). We quantified the alignment ratio of PKC-ε vesicles along microtubules directed towards the phagosome versus those in non-involved regions of the cell. Briefly, PKC-ε vesicles within 0.2 µm of microtubules were defined as ‘close to/aligned along’ microtubules while those further than 0.2 µm were designated ‘far/non-aligned’. Significantly more PKC-ε vesicles were aligned along microtubules directed towards the phagosome compared to microtubules in regions of the cell lacking targets (Fig. 5B,C; Fig. S5B). These data reveal that PKC-ε+ vesicles preferentially align along microtubules directed towards phagosomes and support a model in which PKC-ε traffics from the Golgi to sites of phagocytosis on vesicles that travel on microtubules.

Fig. 5.

PKC-ε preferentially aligns along microtubules directed towards phagosomes. (A) BMDMs were incubated with 5 µm IgG-coated targets (dashed circle), fixed, and stained for PKC-ε, microtubules (α-tubulin) and the nucleus (DAPI). Z-stacks were collected with Airyscan high resolution microscopy. Images represent 3D projections of PKC-ε, α-tubulin, and merged channel and were generated by Imaris software. (B) 3D reconstruction of PKC-ε and microtubule staining into spots and surfaces, respectively, by Imaris. (C) Quantification of the ratio of aligned to non-aligned PKC-ε spots on microtubule surfaces directed towards phagosomes (solid box from B) and non-involved regions (dashed box from B). Pink spots represent PKC-ε puncta aligned along microtubules and blue represent those that are not; n=3 (57 events), each data point represents an event. Data are presented as mean±s.e.m. ****P<0.0001 (two-tailed, paired Student's t-test). Scale bars: 10 μm.

Fig. 5.

PKC-ε preferentially aligns along microtubules directed towards phagosomes. (A) BMDMs were incubated with 5 µm IgG-coated targets (dashed circle), fixed, and stained for PKC-ε, microtubules (α-tubulin) and the nucleus (DAPI). Z-stacks were collected with Airyscan high resolution microscopy. Images represent 3D projections of PKC-ε, α-tubulin, and merged channel and were generated by Imaris software. (B) 3D reconstruction of PKC-ε and microtubule staining into spots and surfaces, respectively, by Imaris. (C) Quantification of the ratio of aligned to non-aligned PKC-ε spots on microtubule surfaces directed towards phagosomes (solid box from B) and non-involved regions (dashed box from B). Pink spots represent PKC-ε puncta aligned along microtubules and blue represent those that are not; n=3 (57 events), each data point represents an event. Data are presented as mean±s.e.m. ****P<0.0001 (two-tailed, paired Student's t-test). Scale bars: 10 μm.

PKC-ε vesicles contain TNF

Colocalization of PKC-ε with VAMP3 (Fig. 1) establishes that PKC-ε is on vesicles. If PKC-ε traffics on Golgi-derived vesicles, those vesicles should contain Golgi-associated protein(s). The Stow group reported that tumor necrosis factor (TNF) traffics from the TGN to the phagocytic cup on vesicles (Manderson et al., 2007; Murray et al., 2005). That TNF is synthesized as a transmembrane protein makes it a bona fide marker of Golgi-derived vesicles. Notably, the Golgi-to-phagosome trafficking pattern of TNF during phagocytosis parallels that of PKC-ε. Thus, we tested the hypothesis that PKC-ε and TNF are transported on the same vesicles. First, RAW cells co-expressing PKC-ε–GFP and TNF–mCherry were subjected to synchronized phagocytosis, fixed and imaged (Fig. 6A). A Pearson's correlation coefficient for PKC-ε–GFP and TNF–mCherry was quantified at the phagosome, the Golgi, and a non-involved region of the membrane. PKC-ε and TNF colocalization at the phagosome and Golgi were significantly higher compared to non-involved regions of the membrane (Fig. 6A, graph). Furthermore, BMDMs undergoing synchronized phagocytosis were stained for endogenous PKC-ε and TNF (Fig. 6B). High resolution Airyscan again revealed significantly higher colocalization between PKC-ε and TNF puncta at the phagosome compared to an equivalent ROI in a region of the cell lacking targets (Fig. 6B,C). These results, in a cell line and primary macrophages, using exogenous fluorescent TNF and PKC-ε or their endogenously stained counterparts, produced the same results; there was significantly more colocalization of the two markers at the phagosome compared with other parts of the cells, consistent with their Golgi-to-phagosome trafficking on the same vesicles.

Fig. 6.

PKC-ε colocalizes with TNF in vesicles and at phagosomes. (A) RAW cells, co-expressing TNF–mCherry and PKC-ε–GFP, were subjected to synchronized phagocytosis with 2 μm IgG-opsonized beads and then fixed. Z-stacks were collected with spinning-disc confocal microscopy. Images represent a single slice. Colocalization between TNF and PKC-ε at the Golgi (circle), phagosome (dotted lines), and a non-involved region of the plasma membrane (boxed region) was quantified by means of Pearson's correlation coefficient using Imaris software; n=4 (21-40 cells), each data point represents a cell. Data are presented as mean±s.e.m. ****P<0.0001 (one-way ANOVA with Bonferroni post-test). Scale bar: 10 μm. (B) BMDMs were incubated with 5 μm IgG-opsonized particles and then fixed and stained for endogenous TNF and PKC-ε. Z-stacks were collected with Airyscan high resolution microscopy. Upper panels: images are the 3D projection of z-stack. Scale bar: 10 μm. Lower images, single-slice image of colocalization (enlarged area shown as blue box) between TNF and PKC-ε at phagosomes (dashed cirle line in merge image). Scale bar: 5 μm. (C) Colocalization between TNF and PKC-ε is indicated on a purple (low) to white (high) color scale. The cell periphery is highlighted and the target indicated by the dashed circle. Colocalization at phagosomes (orange box) compared to non-involved regions (green box) was quantified by means of Pearson's correlation coefficient using Imaris software (graph); n=3 (51 cells), each datapoint represents a cell. Data are presented as mean±s.e.m. **P<0.01 (two-tailed, paired Student's t-test). Scale bar: 10 μm.

Fig. 6.

PKC-ε colocalizes with TNF in vesicles and at phagosomes. (A) RAW cells, co-expressing TNF–mCherry and PKC-ε–GFP, were subjected to synchronized phagocytosis with 2 μm IgG-opsonized beads and then fixed. Z-stacks were collected with spinning-disc confocal microscopy. Images represent a single slice. Colocalization between TNF and PKC-ε at the Golgi (circle), phagosome (dotted lines), and a non-involved region of the plasma membrane (boxed region) was quantified by means of Pearson's correlation coefficient using Imaris software; n=4 (21-40 cells), each data point represents a cell. Data are presented as mean±s.e.m. ****P<0.0001 (one-way ANOVA with Bonferroni post-test). Scale bar: 10 μm. (B) BMDMs were incubated with 5 μm IgG-opsonized particles and then fixed and stained for endogenous TNF and PKC-ε. Z-stacks were collected with Airyscan high resolution microscopy. Upper panels: images are the 3D projection of z-stack. Scale bar: 10 μm. Lower images, single-slice image of colocalization (enlarged area shown as blue box) between TNF and PKC-ε at phagosomes (dashed cirle line in merge image). Scale bar: 5 μm. (C) Colocalization between TNF and PKC-ε is indicated on a purple (low) to white (high) color scale. The cell periphery is highlighted and the target indicated by the dashed circle. Colocalization at phagosomes (orange box) compared to non-involved regions (green box) was quantified by means of Pearson's correlation coefficient using Imaris software (graph); n=3 (51 cells), each datapoint represents a cell. Data are presented as mean±s.e.m. **P<0.01 (two-tailed, paired Student's t-test). Scale bar: 10 μm.

The high degree of PKC-ε and TNF colocalization at the phagosome supports our hypothesis but raises the question of whether PKC-ε associates indiscriminately with intracellular vesicles. To test this, we co-stained BMDMs for PKC-ε and TNF, TGN38 (also known as TGOLN2; a protein that shuttles between the Golgi and plasma membrane) or the early endosome marker, EEA1. Indeed, a calculation of Pearson's correlation coefficients indicate that PKC-ε colocalizes extensively with TNF (0.74, n=5, 38 cells), modestly with TGN38 (0.48, n=5, 36 cells) and poorly with EEA1 (0.29, n=5, 30 cells). The high correlation of PKC-ε with TNF indicates selective association on Golgi-derived vesicles. Its modest association with TGN38 is likely due to the fact that TGN38 cycles from the TGN to the plasma membrane (Reaves et al., 1993). PKC-ε likely associates with the exocytic, but not endocytic, phase of that cycle. The low association with EEA1 would be consistent with this interpretation.

Delivery of vesicles to the phagosome requires PKC-ε

We know that dissociation of PKC-ε from the Golgi (1) inhibits its concentration at the phagosome, (2) blocks membrane addition in response to FcγR ligation, and (3) significantly reduces phagocytosis (Hanes et al., 2017). Using TNF as a surrogate marker for Golgi-derived vesicles, we asked whether PKC-ε is necessary for delivery of TNF to the phagosome. We reasoned that, if PKC-ε and TNF are transported on the same vesicles, and Golgi-associated PKC-ε is necessary for their delivery, then dissociation of PKC-ε from the TGN (by expression of GFP–hSac1-K2A or treatment with PIK93) should inhibit TNF concentration at the phagosome. Indeed, expression of GFP–hSac1-K2A in RAW cells significantly reduced TNF concentration at the phagocytic cup (Fig. 7A). Similarly, wild-type (WT) BMDMs treated with PIK93 delivered significantly fewer TNF puncta to the phagosome compared with controls (Fig. 7B). However, TNF delivery is restored upon PIK93 washout or when PKC-ε is expressed in PKC-ε knockout (εKO) cells (Fig. 7B). Notably, TNF colocalizes with PKC-ε puncta in both WT and εKO re-expressing BMDM (Fig. 7C). Together, these data provide a third line of evidence that PKC-ε is required for TNF vesicle delivery.

Fig. 7.

Golgi-associated PKC-ε is necessary for TNF concentration at phagosomes. (A) RAW cells were transfected with TNF–mCherry with or without GFP–hSac1-K2A and subjected to synchronized phagocytosis. Z-stacks were collected with spinning-disc confocal microscopy. Fluorescence intensity of TNF at phagosomes was normalized to an equivalent membrane ROI in a non-involved region of the cell (indicted with dotted lines, asterisk highlights the phagosome lines) to generate a localization index (graph). A pseudocolored LUT was applied for optimal visualization of TNF fluorescence intensity; n=3 (at least 40 cells per condition), each data point represents a cell. Data are presented as mean±s.e.m. ****P<0.0001 (two-tailed, unpaired Student's t-test). (B) BMDMs from WT and εKO mice were subjected to frustrated phagocytosis, fixed and stained for endogenous TNF. Subset of WT cells were treated with PIK93 (100 nM, 30 min) and either fixed immediately or the drug was washed out (5 min) prior to fixation. PKC-ε–GFP was expressed in a subset of εKO BMDMs (re-expression). Images were collected with TIRFM. The number of TNF puncta was quantified with FIJI software and normalized to cell area (graph); n=3 (at least 34 cells per condition), each data point represents a cell. Data are presented as mean±s.e.m. ****P<0.0001 (one-way ANOVA with Bonferroni post-test). DMSO vs PKC-ε re-expression and PIK93 vs εKO is not significant. (C) PKC-ε–GFP expression in WT and εKO BMDMs. Cells underwent frustrated phagocytosis, and were fixed and stained for endogenous TNF. Images were collected with TIRFM. Colocalization between PKC-ε and TNF was quantified by Pearson's correlation coefficient (graph); n=3 (at least 27 cells per condition), each data point represents a cell. Data are presented as mean±s.e.m. *P<0.05 (two-tailed, unpaired Student's t-test). Scale bars: 10 μm.

Fig. 7.

Golgi-associated PKC-ε is necessary for TNF concentration at phagosomes. (A) RAW cells were transfected with TNF–mCherry with or without GFP–hSac1-K2A and subjected to synchronized phagocytosis. Z-stacks were collected with spinning-disc confocal microscopy. Fluorescence intensity of TNF at phagosomes was normalized to an equivalent membrane ROI in a non-involved region of the cell (indicted with dotted lines, asterisk highlights the phagosome lines) to generate a localization index (graph). A pseudocolored LUT was applied for optimal visualization of TNF fluorescence intensity; n=3 (at least 40 cells per condition), each data point represents a cell. Data are presented as mean±s.e.m. ****P<0.0001 (two-tailed, unpaired Student's t-test). (B) BMDMs from WT and εKO mice were subjected to frustrated phagocytosis, fixed and stained for endogenous TNF. Subset of WT cells were treated with PIK93 (100 nM, 30 min) and either fixed immediately or the drug was washed out (5 min) prior to fixation. PKC-ε–GFP was expressed in a subset of εKO BMDMs (re-expression). Images were collected with TIRFM. The number of TNF puncta was quantified with FIJI software and normalized to cell area (graph); n=3 (at least 34 cells per condition), each data point represents a cell. Data are presented as mean±s.e.m. ****P<0.0001 (one-way ANOVA with Bonferroni post-test). DMSO vs PKC-ε re-expression and PIK93 vs εKO is not significant. (C) PKC-ε–GFP expression in WT and εKO BMDMs. Cells underwent frustrated phagocytosis, and were fixed and stained for endogenous TNF. Images were collected with TIRFM. Colocalization between PKC-ε and TNF was quantified by Pearson's correlation coefficient (graph); n=3 (at least 27 cells per condition), each data point represents a cell. Data are presented as mean±s.e.m. *P<0.05 (two-tailed, unpaired Student's t-test). Scale bars: 10 μm.

The regulatory domain of PKC-ε is sufficient for vesicle delivery

Our earlier work revealed that the isolated regulatory domain of PKC-ε (εRD, aa 1–900) concentrates at the phagocytic cup (Cheeseman et al., 2006; Wood et al., 2013). Given that knowledge, and the current results suggesting that PKC-ε orchestrates TNF delivery, we asked whether εRD per se supports vesicle transport. Using TIRFM and frustrated phagocytosis, we quantified TNF puncta at the membrane in WT and εKO BMDMs expressing εRD–GFP. The number of εRD puncta in WT and εKO were not significantly different from each other or from WT BMDMs expressing full-length PKC-ε–GFP (Fig. 8A). Interestingly, while TNF vesicles are not delivered in εKO BMDMs (Fig. 7B), they are upon expression of εRD in these cells (Fig. 8B,C) and TNF colocalizes with εRD (Fig. 8B,D). These results uncover a novel and intriguing pathway in which the regulatory domain of PKC-ε, independent of its catalytic domain/activity, is necessary for the delivery of Golgi-derived vesicles to the phagosome. As we have previously shown that the catalytic activity of PKC-ε is necessary for membrane fusion (Hanes et al., 2017), these findings are consistent with a model in which PKC-ε serves two functions during phagocytosis – εRD drives vesicle delivery while catalytic activity phosphorylates substrates required for vesicle fusion. These findings provide a foundation for further exploration into the role of εRD at the Golgi and identification of PKC-ε substrates at the phagosome.

Fig. 8.

The regulatory domain of PKC-ε is sufficient for TNF vesicle delivery. (A,B) WT and εKO BMDMs expressing PKC-ε–GFP, the regulatory domain of PKC-ε (εRD–GFP) or GFP only (control) underwent frustrated phagocytosis and were then fixed and stained for endogenous TNF. Images were collected by TIRFM. (A) PKC-ε–GFP and εRD–GFP expression in WT and εKO BMDMs. Images are shown with puncta visualized by a Fire LUT with corresponding GFP images in inserts. The number of puncta/cell area was calculated (graph); n=3 (at least 30 cells per condition), each data point represents a cell. Data are presented as mean±s.e.m. n.s., not significant (P>0.05) (one-way ANOVA with Bonferroni post-test). (B) Endogenous TNF staining in PKC-ε–GFP- and εRD–GFP-expressing BMDMs. A Fire LUT was applied for optimal visualization of TNF puncta, with corresponding TNF fluorescence staining shown in inserts. (C) The number of TNF puncta/cell area was calculated (graph); n=3 (at least 30 cells PKC-ε/εRD-expressing cells and 11 cells εKO), each data point represents a cell. Data are presented as mean±s.e.m. ****P<0.0001, **P<0.01, *P<0.05, n.s., not significant (P>0.05) (one-way ANOVA with Bonferroni post-test). (D) Colocalization between TNF and εRD–GFP was quantified by means of Pearson's correlation coefficient (graph); n=3 (30 cells per condition), each data point represents a cell. Data are presented as mean±s.e.m. n.s., not significant (P>0.05) (two-tailed, paired Student's t-test). Scale bars: 10 μm.

Fig. 8.

The regulatory domain of PKC-ε is sufficient for TNF vesicle delivery. (A,B) WT and εKO BMDMs expressing PKC-ε–GFP, the regulatory domain of PKC-ε (εRD–GFP) or GFP only (control) underwent frustrated phagocytosis and were then fixed and stained for endogenous TNF. Images were collected by TIRFM. (A) PKC-ε–GFP and εRD–GFP expression in WT and εKO BMDMs. Images are shown with puncta visualized by a Fire LUT with corresponding GFP images in inserts. The number of puncta/cell area was calculated (graph); n=3 (at least 30 cells per condition), each data point represents a cell. Data are presented as mean±s.e.m. n.s., not significant (P>0.05) (one-way ANOVA with Bonferroni post-test). (B) Endogenous TNF staining in PKC-ε–GFP- and εRD–GFP-expressing BMDMs. A Fire LUT was applied for optimal visualization of TNF puncta, with corresponding TNF fluorescence staining shown in inserts. (C) The number of TNF puncta/cell area was calculated (graph); n=3 (at least 30 cells PKC-ε/εRD-expressing cells and 11 cells εKO), each data point represents a cell. Data are presented as mean±s.e.m. ****P<0.0001, **P<0.01, *P<0.05, n.s., not significant (P>0.05) (one-way ANOVA with Bonferroni post-test). (D) Colocalization between TNF and εRD–GFP was quantified by means of Pearson's correlation coefficient (graph); n=3 (30 cells per condition), each data point represents a cell. Data are presented as mean±s.e.m. n.s., not significant (P>0.05) (two-tailed, paired Student's t-test). Scale bars: 10 μm.

In summary, these studies reveal a novel paradigm for focal exocytosis in non-polarized cells. Specifically, PKC-ε translocates on vesicles that travel on microtubules. The vesicles carry TNF from the Golgi to the phagosome and PKC-ε is required for their delivery. The vesicles fuse, expanding the membrane for pseudopod extension and rapid phagocytosis. The most intriguing finding from this work is that regulatory domain of PKC-ε is required for the transit of vesicles, defining a PKC function independent of its catalytic activity. While many questions remain to be answered, these findings open the door to investigations of PKC-ε orchestration of focal exocytosis in non-polarized cells.

Efficient IgG-mediated phagocytosis requires membrane addition for pseudopod extension (Braun and Niedergang, 2006; Cannon and Swanson, 1992; Gerlach et al., 2020; Holevinsky and Nelson, 1998) but the source of that membrane and the mechanism for its focal delivery is unclear. Work by the Grinstein group demonstrated the targeted delivery of VAMP3+ vesicles to forming phagosomes (Bajno et al., 2000). Their findings unequivocally demonstrated focal exocytosis of vesicles at forming phagosomes. Our studies identified PKC-ε as a critical player for pseudopod extension, and whole-cell patch clamping established that macrophages expand their membrane by about one-third as they spread on IgG surfaces (Hanes et al., 2017; Wood et al., 2013). PKC-ε-null macrophages are unable to add membrane in response to FcγR ligation, a deficit that was reversed upon expression of PKC-ε in PKC-ε-null BMDMs (Hanes et al., 2017).

As the majority of macrophage PKC-ε is cytosolic (Larsen et al., 2000), and PKCs translocate from the cytosol to their sites of activity, we predicted that cytosolic PKC-ε concentrated at phagosomes to facilitate membrane addition. Several findings led us to rethink this assumption. First, using capacitance to quantify membrane expansion, we found no addition of membrane in cells treated with the microtubule-disrupting drug nocodazole, or with PIK93, which selectively dissociates PKC-ε from the Golgi (Hanes et al., 2017). If there was significant cytosolic translocation, we would have expected at least some membrane addition in the presence of these inhibitors. Secondly, our previous studies revealed a Golgi-associated pool of PKC-ε. We discovered that the pseudosubstrate region of PKC-ε interacts with PI4P, tethering PKC-ε to the TGN. Blocking this interaction through selective removal of PI4P with hSac1-K2A, or inhibition of PI4 kinase with PIK93, dissociates PKC-ε from the TGN, prevents PKC-ε concentration at phagosomes, and significantly reduces phagocytosis (Hanes et al., 2017). These findings support a model in which the Golgi-tethered, rather than cytosolic, PKC-ε is critical for phagocytosis. Notably, hSac1-K2A and PIK93 provided us with tools to selectively deplete the Golgi-tethered pool of PKC-ε, allowing us to study its role in phagocytosis. Finally, during live imaging of phagocytosis in PKC-ε–GFP expressing BMDMs, we observed PKC-ε puncta exiting the Golgi (Movie 1), providing real-time evidence that PKC-ε translocates from the Golgi on a vesicle, an unexpected discovery.

This raised the question of how Golgi-tethered PKC-ε is transported to the phagosome. The most likely mechanism would be vesicular trafficking on microtubules. Indeed, others have reported that delivery of lipids and proteins from the Golgi to the plasma membrane on vesicles (Stalder and Gershlick, 2020). We had previously reported that nocodazole prevented membrane addition in response to FcγR ligation (Hanes et al., 2017), implicating microtubules and, by extension, vesicular trafficking in phagosome formation. TIRFM allowed us to visualize the ‘phagosome’, revealing that PKC-ε approaches the membrane on a vesicle, a vesicle that then fuses (Figs 13). To our knowledge, this is the first evidence that any PKC translocates on a vesicle to the plasma membrane and represents the first novel finding of this work.

If Golgi-tethered PKC-ε traffics on vesicles, but its dissociation from the Golgi blocks phagocytosis and PKC-ε concentration (Hanes et al., 2017), how could we study the role of PKC-ε at the Golgi? Work from the Stow laboratory documenting that TNF (a transmembrane protein that travels through the Golgi) concentrates at phagosomes (Manderson et al., 2007; Murray et al., 2005) provided us with a candidate marker for Golgi-derived vesicles trafficking to phagosomes. Once we established that PKC-ε and TNF colocalized at the Golgi and at the phagosome (Fig. 6), we could use TNF to ask whether PKC-ε was simply a passenger on vesicles or played a more active role. Quantifying TNF appearance at the membrane in PKC-ε-null BMDMs, and upon re-expression of PKC-ε in null cells, revealed that PKC-ε is required for TNF+ vesicle trafficking to the phagosome, representing a second novel finding of this work – that PKC-ε is required for Golgi-derived vesicle delivery to the sites of FcγR ligation membrane (Fig. 7).

Given that PKC-ε transits from the Golgi to phagosomes on vesicles along microtubules and drives vesicle delivery, we are left with the question of how does PKC-ε facilitate Golgi-to-phagosome vesicular trafficking. The catalytic activity of PKC-ε is required for membrane fusion, but not for concentration of PKC-ε at the phagosome (Hanes et al., 2017; Wood et al., 2013). The regulatory domain of PKC-ε concentrates at the phagosome, but is it sufficient for delivery of vesicles? εRD–GFP expressed in εKO macrophages appeared as puncta by TIRFM that were indistinguishable from the pattern produced by full-length PKC-ε–GFP (Fig. 8A). That expression of εRD–GFP in εKO cells restored delivery of TNF puncta demonstrates that εRD is sufficient for vesicular trafficking from the Golgi (Fig. 8B), the third novel finding from this work.

Finally, what, if any, is the role of VAMP3 during this process? VAMP3 is a marker of recycling endosomes (McMahon et al., 1993). While it colocalizes with TNF and PKC-ε at the phagosome (Fig. 2B), it is not required for FcγR-mediated phagocytosis (Allen et al., 2002 and Fig. S1). A model consistent with all the data is that PKC-ε (actually εRD) orchestrates vesicle formation at the Golgi (D'Amico and Lennartz, 2018; Miralles and Lennartz, 2018). The vesicles then ‘mature’ as they move along microtubules to the phagosome, fusing with VAMP3+ endosomes en route. Upon docking, the catalytic activity of PKC-ε phosphorylates as yet unknown substrate(s) for vesicle fusion, pseudopod extension and phagocytosis. In this model, the two domains of PKC-ε serve different functions – εRD at the Golgi for vesicle formation and catalytic activity to promote vesicle fusion. As we have previously reported that chimeras containing εRD and the catalytic domain of (the closely related) PKC-δ concentrate at the phagosome but do not support phagocytosis (Wood et al., 2013), we would argue that the proteins phosphorylated at the phagosome are uniquely PKC-ε substrates.

Summary

The data presented include three novel findings: (1) PKC-ε is delivered to the phagosome on vesicles, (2) these vesicles originate in the Golgi, carry TNF, and their delivery requires PKC-ε, and (3) the regulatory domain of PKC-ε is sufficient for vesicle delivery (Fig. S6). These findings challenge the current paradigm of PKC activation by translocation from the cytosol and raise numerous questions, the answers to which will provide insight into novel PKC functions. There are several unanswered questions. What is the role of PKC-ε at the Golgi? How does the regulatory domain, independent of the catalytic domain (and activity) mediate vesicular trafficking? How are vesicles targeted to the phagosome? And what is/are the PKC-ε substrates at the phagosome that are required for vesicle fusion?

Perhaps the most exciting avenue of investigation will be the relay between FcγR ligation and the Golgi. Pilot studies indicate that Syk signaling is required for PKC-ε concentration at the phagosome. If so, then there must be some FcγR-to-Golgi-to-phagosome relay system that rapidly recruits Golgi vesicles specifically to phagosomes for focal exocytosis. Elucidation of this signaling network may provide insight into diseases such as cancer, where elevated PKC-ε promotes focal exocytosis for tumor cell metastasis (Gorin and Pan, 2009; Pan et al., 2005; Tachado et al., 2002).

Buffers and reagents

ACK lysis buffer is 0.5 M NH4Cl, 1 mM KHCO3, and 0.1 mM EDTA (pH 7.4). HBSS++ is Hanks’ balanced salt solution (HBSS) containing 4 mM sodium bicarbonate, 10 mM HEPES, 1.5 mM each CaCl2, and MgCl2 (pH 7.4). Paraformaldehyde (PFA) was from Polysciences, Inc (18814-10). DMSO was from Sigma (D2438). PIK93 was from Echelon Biosciences (Salt Lake City, UT) and was used at 100 nM. Nocodazole was from Cayman Chemicals (Ann Arbor, MI) and was used at 10 µM. Blocking buffer was PBS containing 0.5% fish gelatin, 2% BSA and thimerosal. Nucleofections were performed with a Mouse Macrophage Nucleofector Kit (Lonza, VPA-1009).

Mice and cells

PKC-ε+/− heterozygotes on the C57/B16 background were purchased from the Jackson Laboratory (stock# 004189; Bar Harbor, ME) and bred in the Albany Medical Center Animal Resources Facility. Heterozygotes were crossed to produce the wild-type and PKC-ε−/− mice. VAMP-3-null mice were a generous gift from Dr Sidney Whiteheart (University of Kentucky, Lexington, Kentucky, USA) and generated by Dr Jeffery Pessin (Albert Einstein College of Medicine, New York City, New York, USA) (Allen et al., 2002). Cells from both male and female were used; there were no effects of sex on experimental outcome (average age 3–6 months; age was not found to be a relevant factor). All animal procedures were approved by the Albany Medical Center Institutional Animal Care and Use Committee. Cells obtained from one mouse represent one independent experiment.

Bone marrow-derived macrophages

Mice were euthanized and their femurs and pelvis were removed. Bone marrow was collected, and red blood cells were lysed with ACK lysis buffer. Bone marrow stem cells were differentiated by incubation in bone marrow medium (BMM), which contains phenol-red free (high glucose) Dulbecco's modified Eagle's medium (DMEM), 10% PBS, 20% L-cell conditioned medium (Hanes et al., 2017), sodium bicarbonate, and gentamicin (RPI, G38000). BMDM were used 7–14 days after harvesting.

RAW 264.7 mouse macrophages

The RAW subclone, LacR/FMLPR.2 (RAW cells) were a generous gift from Dr Steven Greenberg (Colombia University, New York, NY) (Cox et al., 1997). Cells were maintained in media containing DMEM, 10% FBS, and sodium bicarbonate. As cell contamination prevents transfection of this cell line, the extent of validation was that the cells expressed plasmids and phagocytosis of controls/concentration of PKC-ε at phagosomes occurred.

IgG-coating of targets and surfaces

Acid-washed 2 or 5 µm borosilicate glass beads (Duke Standards, Thermo Fisher Scientific, 9005) or 15 mm glass coverslips or Mattek dishes (Cellvis, #D29-14-1.5-N) were sequentially coated with poly-L-lysine, dimethylpimelimidate and 1% IgG-free BSA. Free active groups were blocked with 0.5 M Tris-HCl (pH 8.0) (1 h, room temperature) before opsonizing with rabbit anti-BSA IgG (Sigma, B1520). IgG-free BSA treatment and blocking prior to opsonization was omitted for beads and surfaces opsonized with human IgG (Sigma, I4506). Free active groups were blocked overnight with 0.5 M Tris-HCl (pH 8.0) (4°C, overnight). Before use, beads and surfaces were washed with PBS and diluted in HBSS++.

Phagocytosis

Phagocytosis of IgG-opsonized targets

Time courses with 5 µm human IgG-opsonized beads were performed as previously described (Wood et al., 2013). 5 µm opsonized beads were added (∼3 beads per cell) and then incubated at 37°C. Cells were fixed (4% PFA in PBS) and prepared for immunofluorescent staining. Images of phagocytosing cells were collected with Airyscan high resolution microscopy. Cells that did not internalize beads were excluded from analysis.

Frustrated phagocytosis

Synchronized time courses on IgG-coated coverslips were performed. Cells were cooled on ice for 30 min and transferred to a 37°C heating block. Cells were fixed (4% PFA in PBS) after 30 min and prepared for immunofluorescence staining. For live imaging, ice steps were eliminated. Cells were imaged via TIRFM as they sat and spread onto rabbit or human IgG-coated surfaces. Cells that did not internalize beads were excluded from imaging and analysis.

Synchronized phagocytosis

Cells were pre-incubated on ice for 30 min and then an additional 15 min with IgG-opsonized beads. After, cells were transferred to a 37°C heating block and then fixed (4% PFA in PBS). Images were collected using spinning-disc confocal microscopy. Cells that did not internalize beads were excluded from analysis.

Spreading assays

Assays were performed as previously described (Wood et al., 2013). Briefly, ice-cold cells were allowed to attach to (rabbit) IgG-coated coverslips (30 min on ice). Plates were transferred to a 37°C water bath, cells allowed to spread for 15 min, then fixed (4% PFA PBS, 10 min). The exposed IgG was detected using Alexa Fluor 568 anti-rabbit-IgG antibody. Coverslips were mounted and imaged by epifluorescence. The tight apposition of cells with the IgG surface prevents antibody from getting under cells, generating ‘black holes’, the area of which was quantified. Spreading assays parallel the results of capacitance and thus can be used as a readout of membrane addition (Wood et al., 2013).

Expression of exogenous proteins

Nucleofection

Plasmids Akt-PH–GFP, VAMP-3–GFP, and PKC-ε–GFP were delivered to BMDMs following manufacturer's instructions with the Mouse macrophage kit (VPA-1009). 3×106 BMDMs were suspended briefly in 82 µl macrophage solution and 18 µl supplement; DNA (5 µg) was added and cells were transferred to a sterile electroporation cuvette. The nucleofector was set to the Y-001 program and after cells were transferred to a six-well dish with 2 ml of BMM.

Transfection

RAW 264.7 cells were cultured as described above. Cells (5×104) were seeded onto 12-mm Matteks or coverslips and transfected with Lipofectamine 2000 at a 3:1 ratio (Lipofectamine 2000:DNA). A detailed protocol has been previously published (Wood et al., 2013). Akt-PH-GFP, GFP-hSac-1-K2A and TNF-mCherry plasmids were generous gifts from Tamas Balla (NIH, NICHD, Bethesda, MD, USA; Varnai and Balla, 1998), Peter Mayinger (Oregon Health and Science University, Portland, OR, USA; Mayinger, 2009) and Jennifer Stow (The University of Queensland, St Lucia, QLD, Australia; Murray et al., 2005), respectively. The construction of the PKC-ε–GFP plasmid has been previously described (Shirai et al., 1998)

Viral transduction

Retroviral construction of PKC-ε–GFP and BMDM transduction have been previously published (Hanes et al., 2017; Wood et al., 2013).

Immunofluorescence staining

Antibodies

Antibodies used were as follows. Mouse anti-PKC-ε (Santa Cruz Biotechnology, sc-1681, 1:50, used only in Fig. 1); mouse anti-TNF (abcam, ab1793, 1:100); rabbit anti-PKC-ε (Millipore, 06-991, 1:200); and mouse anti-GM130 (BDBiosciences, bd610822, 1:500). Primary antibodies were diluted in blocking buffer. Anti-α-tubulin-FITC (Sigma-Aldrich, F2168, 1:250 diluted in 1% BSA); anti-Golgin-245 conjugated to Alexa Fluor 555 (Bioss, bs13487R-A555, 1:100 diluted in 1% BSA); goat anti-rabbit-IgG conjugated to Alexa Fluor 488 (Life Technologies, A21069, 1:500); goat anti-mouse-IgG conjugated to Alexa Fluor 488 (Life Technologies, A11017, 1:500); goat anti-mouse-IgG conjugated to Alexa Fluor 568 (Life Technologies, A11019, 1:500). Secondary antibodies were incubated in blocking buffer containing 10% serum. Antibody specificity was confirmed using εKO macrophages (little/no staining).

Cells were fixed with 4% PFA in PBS and permeabilized with 0.04% Triton X-100 in PBS. Blocking was undertaken for 1 h at room temperature with blocking buffer. Primary antibodies were incubated with cells overnight at 4°C. Secondary antibodies were incubated with cells for 1 h at room temperature. Cells were stained with DAPI (5 min) and post fixed (1% PFA PBS, 10 min). Coverslips were mounted with Prolong Glass Diamond Mount Medium (Invitrogen, P36980) and Matteks were stored in thimerosal in HBSS++ (4°C).

Patch clamping

Whole-cell patch clamping was performed as described previously (Wood et al., 2013). Briefly, cells were lifted from plates with 5 mM EDTA, resuspended in HBSS++, plated onto coverslips, and incubated on ice for 30 min (2×105 cells per coverslip). Cells were then transferred to 37°C for 5 min prior to being bathed in a balanced salt solution (10 mM HEPES, pH 7.4 containing 145 mM NaCl, 5 mM KCl, 1 mM CaCl2, 2 mM MgCl2, and 5 mM glucose) for capacitance measurements. Cells were maintained at a 0 mV holding potential.

Imaging

Total internal reflection fluorescence microscopy

TIRFM was performed on a Zeiss laser TIRF 3 system using 100×/1.25 NA oil objective. Live images were collected every 8 s over a 10-min period.

High-resolution airyscan microscopy

High-resolution airyscan microscopy was performed on a Zeiss LSM880 confocal microscope system with an Airyscan detector running under Zeiss ZEN2.3 software. A 63×/1.4 NA oil objective was used to collect z-stacks and live images. Raw images were processed using airyscan processing in ZEN Black 2.3 software.

Spinning-disc confocal microscopy

Spinning-disc confocal microscopy was performed on a Olympus IX81-DSU with 100×/1.4 NA oil objective with a Hamamatsu electron-multiplying charge-coupled device camera, driven by MetaMorph software (Molecular Devices, Sunnyvale, CA). Live images were collected every 5 s over a 10-min period.

Analysis

Vesicle fusion

Videos were analyzed using FIJI software. Fusing vesicles were isolated by cropping, and the maximum and minimum fluorescent intensities were measured and plotted. Measurements were taken each frame, starting before the vesicle appears to after the fusion event. Intensities for each event were plotted onto separate graphs. Pseudo-colored look-up table (LUT) was applied to images to better visualize fusion events. 3D surface plots and frame montage were also generated in FIJI.

Microtubule alignment

FIJI software was used to generate color line plots of DM1-α and PKC-ε fluorescent intensities. For 3D reconstructions, surfaces were generated based on DM1-α staining and spots for PKC-ε puncta in Imaris x64 version 9.5.0. Xtension ‘find spots close to surface’ (https://imaris.oxinst.com/learning/view/article/spots-close-to-surface-xtension) was performed with a threshold set to 0.2 µm. The percentage of spots within 0.2 µm or closer (‘close spots’) was quantified in cells undergoing frustrated phagocytosis. ROIs for cells phagocytosing targets (ROIs of 6 µm by 6 µm around the phagosome or a non-involved region of the cell) were generated. The number of ‘close’ PKC-ε spots to DM1-α surface was divided by the number of PKC-ε puncta further than 0.2 µm (‘far spots’) to generate a ratio. The ratio of close versus far spots was analyzed in Prism.

Phagosome colocalization

Imarisx64 version 9.5.0. 3D renderings of phagosome were created based on TNF or PKC-ε fluorescence expression. Mask of 3D rendering was applied to fluorescent channels. Pearson's correlation coefficient was calculated. ROIs for colocalization were made from masked channels. The same steps were applied to a non-involved region of the membrane for comparison.

Vesicle colocalization

Imarisx64 version 9.5.0. was used. For airyscan images, 3D renderings were created in phagosome and non-involved ROIs based on fluorescent stains and used to mask each channel. Pearson's correlation coefficient was calculated between masked PKC-ε and TNF channels. For TIRF images, two methods were utilized to determine Pearson's correlation coefficient: (1) FIJI plugin ‘Just Another Colocalisation’ (Fig. 1C) and (2) Imarisx64 software. Background fluorescence was subtracted, and an ROI mask was applied based on PKC-ε staining.

TNF concentration at phagosomes

Analysis of TNF at phagosomes was performed using FIJI software (Wood et al., 2013). Mean fluorescence intensity at phagosomes and non-involved regions of similar size were measured. The localization index was calculated by normalizing TNF fluorescence intensity at phagosome to non-involved regions. Values higher than 1.0 signify concentration.

Puncta quantitation

Two methods were used to quantify the number of puncta: (1) FIJI plugin ‘Mosaic Particle Tracker 2D/3D’, where parameters were set to a radius of 3 pixels (48 µm), cutoff t 1, and a percentile of 0.5% (Cardinale et al., 2012) (Fig. 1B), and (2) using the ‘Find Maxima’ function in FIJI software was utilized to quantify the number of puncta at the cell surface in TIRF images. Number of puncta was normalized to cell area. Background subtraction was applied to images for optimal visualization of puncta.

Statistical analysis

All data are represented as mean±s.e.m. Significance was calculated by two-tailed paired and unpaired Student's t-test or one-way ANOVA with a Bonferroni post-test. P≤0.05 was considered significant. Each experiment sample represents one animal.

We thank Dr Joseph Mazurkiewicz and the Albany Medical College Imaging Facility for use of the TIRF and high-resolution Airyscan confocal microscopes, Drs Kate Tubbesing and Ling Wang for assistance with Imaris imaging analysis. Furthermore, we thank Drs Margarida Barroso, Jeremy Logue and James Drake for constructive feedback and reading the manuscript. Finally, many thanks to Deborah Moran and Rosemary Prestipino for administrative help.

Author contributions

Conceptualization: M.T., M.R.L.; Methodology: A.E.D., A.C.W., C.M.Z., X.Z., A.M., M.T.; Validation: A.E.D., M.R.L.; Formal analysis: A.E.D., A.C.W., C.M.Z.; Investigation: A.E.D., C.M.Z., A.M., M.R.L.; Resources: A.E.D., M.R.L.; Data curation: A.E.D., A.C.W., C.M.Z., X.Z.; Writing - original draft: A.E.D., X.Z., M.R.L.; Writing - review & editing: A.E.D., C.M.Z., M.R.L.; Visualization: A.E.D., X.Z.; Supervision: M.T., M.R.L.; Funding acquisition: M.T.

Funding

This work was supported by the National Institutes of Health (9R01 GM090325 to M.R.L.; R01HL097111 and R01HL123364 to M.T.) and the Johnathan R. Vasiliou Foundation (to M.R.L.). Deposited in PMC for release after 12 months.

The peer review history is available online at https://journals.biologists.com/jcs/article-lookup/doi/10.1242/jcs.258886.

Aderem
,
A.
(
2002
).
How to eat something bigger than your head
.
Cell
110
,
5
-
8
.
Allen
,
L. A.
,
Yang
,
C.
and
Pessin
,
J. E.
(
2002
).
Rate and extent of phagocytosis in macrophages lacking vamp3
.
J. Leukoc. Biol.
72
,
217
-
221
.
Bajno
,
L.
,
Peng
,
X.-R.
,
Schreiber
,
A. D.
,
Moore
,
H.-P.
,
Trimble
,
W. S.
and
Grinstein
,
S.
(
2000
).
Focal exocytosis of VAMP3-containing vesicles at sites of phagosome formation
.
J. Cell Biol.
149
,
697
-
706
.
Ben Mkaddem
,
S.
,
Benhamou
,
M.
and
Monteiro
,
R. C.
(
2019
).
Understanding Fc receptor involvement in inflammatory diseases: from mechanisms to new therapeutic tools
.
Front. Immunol.
10
,
811
.
Botelho
,
R. J.
and
Grinstein
,
S.
(
2011
).
Phagocytosis
.
Curr. Biol.
21
,
R533
-
R538
.
Braun
,
V.
and
Niedergang
,
F.
(
2006
).
Linking exocytosis and endocytosis during phagocytosis
.
Biol. Cell
98
,
195
-
201
.
Cannon
,
G. J.
and
Swanson
,
J. A.
(
1992
).
The macrophage capacity for phagocytosis
.
J. Cell Sci.
101
,
907
-
913
.
Cardinale
,
J.
,
Paul
,
G.
and
Sbalzarini
,
I. F.
(
2012
).
Discrete region competition for unknown numbers of connected regions
.
IEEE Trans. Image Process.
21
,
3531
-
3545
.
Cheeseman
,
K. L.
,
Ueyama
,
T.
,
Michaud
,
T. M.
,
Kashiwagi
,
K.
,
Wang
,
D.
,
Flax
,
L. A.
,
Shirai
,
Y.
,
Loegering
,
D. J.
,
Saito
,
N.
and
Lennartz
,
M. R.
(
2006
).
Targeting of protein kinase C-epsilon during Fcgamma receptor-dependent phagocytosis requires the epsilonC1B domain and phospholipase C-γ1
.
Mol. Biol. Cell
17
,
799
-
813
.
Chen
,
G.
,
Zhuchenko
,
O.
and
Kuspa
,
A.
(
2007
).
Immune-like phagocyte activity in the social amoeba
.
Science
317
,
678
-
681
.
Cox
,
D.
,
Chang
,
P.
,
Zhang
,
Q.
,
Reddy
,
P. G.
,
Bokoch
,
G. M.
and
Greenberg
,
S.
(
1997
).
Requirements for both Rac1 and Cdc42 in membrane ruffling and phagocytosis in leukocytes
.
J. Exp. Med.
186
,
1487
-
1494
.
D'Amico
,
A. E.
and
Lennartz
,
M. R.
(
2018
).
Protein kinase C-epsilon in membrane delivery during phagocytosis
.
J. Immunol. Sci.
2
,
26
-
32
.
Desjardins
,
M.
,
Houde
,
M.
and
Gagnon
,
E.
(
2005
).
Phagocytosis: the convoluted way from nutrition to adaptive immunity
.
Immunol. Rev.
207
,
158
-
165
.
Gerlach
,
B. D.
,
Marinello
,
M.
,
Heinz
,
J.
,
Rymut
,
N.
,
Sansbury
,
B. E.
,
Riley
,
C. O.
,
Sadhu
,
S.
,
Hosseini
,
Z.
,
Kojima
,
Y.
,
Tang
,
D. D.
et al. 
(
2020
).
Resolvin D1 promotes the targeting and clearance of necroptotic cells
.
Cell Death Differ.
27
,
525
-
539
.
Gorin
,
M. A.
and
Pan
,
Q.
(
2009
).
Protein kinase C epsilon: an oncogene and emerging tumor biomarker
.
Mol. Cancer
8
,
9
.
Hanes
,
C. M.
,
D'Amico
,
A. E.
,
Ueyama
,
T.
,
Wong
,
A. C.
,
Zhang
,
X.
,
Hynes
,
W. F.
,
Barroso
,
M. M.
,
Cady
,
N. C.
,
Trebak
,
M.
,
Saito
,
N.
et al. 
(
2017
).
Golgi-associated protein kinase C-epsilon is delivered to phagocytic cups: role of phosphatidylinositol 4-phosphate
.
J. Immunol.
199
,
271
-
277
.
Holevinsky
,
K. O.
and
Nelson
,
D. J.
(
1998
).
Membrane capacitance changes associated with particle uptake during phagocytosis in macrophages
.
Biophys. J.
75
,
2577
-
2586
.
Jaiswal
,
J. K.
,
Rivera
,
V. M.
and
Simon
,
S. M.
(
2009
).
Exocytosis of post-Golgi vesicles is regulated by components of the endocytic machinery
.
Cell
137
,
1308
-
1319
.
Jaumouille
,
V.
and
Grinstein
,
S.
(
2016
).
Molecular mechanisms of phagosome formation
.
Microbiol. Spectr.
4
.
Larsen
,
E. C.
,
DiGennaro
,
J. A.
,
Saito
,
N.
,
Mehta
,
S.
,
Loegering
,
D. J.
,
Mazurkiewicz
,
J. E.
and
Lennartz
,
M. R.
(
2000
).
Differential requirement for classic and novel PKC isoforms in respiratory burst and phagocytosis in RAW 264.7 cells
.
J. Immunol.
165
,
2809
-
2817
.
Lee
,
W. L.
,
Mason
,
D.
,
Schreiber
,
A. D.
and
Grinstein
,
S.
(
2007
).
Quantitative analysis of membrane remodeling at the phagocytic cup
.
Mol. Biol. Cell
18
,
2883
-
2892
.
Manderson
,
A. P.
,
Kay
,
J. G.
,
Hammond
,
L. A.
,
Brown
,
D. L.
and
Stow
,
J. L.
(
2007
).
Subcompartments of the macrophage recycling endosome direct the differential secretion of IL-6 and TNFalpha
.
J. Cell Biol.
178
,
57
-
69
.
Marion
,
S.
,
Mazzolini
,
J.
,
Herit
,
F.
,
Bourdoncle
,
P.
,
Kambou-Pene
,
N.
,
Hailfinger
,
S.
,
Sachse
,
M.
,
Ruland
,
J.
,
Benmerah
,
A.
,
Echard
,
A.
et al. 
(
2012
).
The NF-kappaB signaling protein Bcl10 regulates actin dynamics by controlling AP1 and OCRL-bearing vesicles
.
Dev. Cell
23
,
954
-
967
.
Marshall
,
J. G.
,
Booth
,
J. W.
,
Stambolic
,
V.
,
Mak
,
T.
,
Balla
,
T.
,
Schreiber
,
A. D.
,
Meyer
,
T.
and
Grinstein
,
S.
(
2001
).
Restricted accumulation of phosphatidylinositol 3-kinase products in a plasmalemmal subdomain during Fcγ receptor-mediated phagocytosis
.
J. Cell Biol.
153
,
1369
-
1380
.
Mayinger
,
P.
(
2009
).
Regulation of Golgi function via phosphoinositide lipids
.
Semin. Cell Dev. Biol.
20
,
793
-
800
.
McMahon
,
H. T.
,
Ushkaryov
,
Y. A.
,
Edelmann
,
L.
,
Link
,
E.
,
Binz
,
T.
,
Niemann
,
H.
,
Jahn
,
R.
and
Südhof
,
T. C.
(
1993
).
Cellubrevin is a ubiquitous tetanus-toxin substrate homologous to a putative synaptic vesicle fusion protein
.
Nature
364
,
346
-
349
.
Miralles
,
G. M.
and
Lennartz
,
M. R.
(
2018
). Vesicular trafficking: from the Golgi to the plasma membrane. In
Phagocytosis: Overview, History, and Role in Human Health and Disease
(ed.
C.
Rosales
), pp.
135
-
162
.
New York
:
Nova Science Publishers, Inc
.
Murray
,
R. Z.
,
Kay
,
J. G.
,
Sangermani
,
D. G.
and
Stow
,
J. L.
(
2005
).
A role for the phagosome in cytokine secretion
.
Science
310
,
1492
-
1495
.
Newton
,
A. C.
(
1997
).
Regulation of protein kinase C
.
Curr. Opin. Cell Biol.
9
,
161
-
167
.
Niedergang
,
F.
and
Chavrier
,
P.
(
2004
).
Signaling and membrane dynamics during phagocytosis: many roads lead to the phagos(R)ome
.
Curr. Opin. Cell Biol.
16
,
422
-
428
.
Pan
,
Q.
,
Bao
,
L. W.
,
Kleer
,
C. G.
,
Sabel
,
M. S.
,
Griffith
,
K. A.
,
Teknos
,
T. N.
and
Merajver
,
S. D.
(
2005
).
Protein kinase Cε is a predictive biomarker of aggressive breast cancer and a validated target for RNA interference anticancer therapy
.
Cancer Res.
65
,
8366
-
8371
.
Pears
,
C.
,
Schaap
,
D.
and
Parker
,
P. J.
(
1991
).
The regulatory domain of protein kinase C-epsilon restricts the catalytic-domain-specificity
.
Biochem. J.
276
,
257
-
260
.
Reaves
,
B.
,
Horn
,
M.
and
Banting
,
G.
(
1993
).
TGN38/41 recycles between the cell surface and the TGN: brefeldin A affects its rate of return to the TGN
.
Mol. Biol. Cell
4
,
93
-
105
.
Schaap
,
D.
,
Parker
,
P. J.
,
Bristol
,
A.
,
Kriz
,
R.
and
Knopf
,
J.
(
1989
).
Unique substrate specificity and regulatory properties of PKC-epsilon: a rationale for diversity
.
FEBS Lett.
243
,
351
-
357
.
Schmoranzer
,
J.
,
Goulian
,
M.
,
Axelrod
,
D.
and
Simon
,
S. M.
(
2000
).
Imaging constitutive exocytosis with total internal reflection fluorescence microscopy
.
J. Cell Biol.
149
,
23
-
32
.
Shirai
,
Y.
,
Kashiwagi
,
K.
,
Yagi
,
K.
,
Sakai
,
N.
and
Saito
,
N.
(
1998
).
Distinct effects of fatty acids on translocation of gamma- and epsilon-subspecies of protein kinase C
.
J. Cell Biol.
143
,
511
-
521
.
Stalder
,
D.
and
Gershlick
,
D. C.
(
2020
).
Direct trafficking pathways from the Golgi apparatus to the plasma membrane
.
Semin. Cell Dev. Biol.
107
,
112
-
125
.
Tachado
,
S. D.
,
Mayhew
,
M. W.
,
Wescott
,
G. G.
,
Foreman
,
T. L.
,
Goodwin
,
C. D.
,
McJilton
,
M. A.
and
Terrian
,
D. M.
(
2002
).
Regulation of tumor invasion and metastasis in protein kinase C epsilon-transformed NIH3T3 fibroblasts
.
J. Cell. Biochem.
85
,
785
-
797
.
Tubbesing
,
K.
,
Ward
,
J.
,
Abini-Agbomson
,
R.
,
Malhotra
,
A.
,
Rudkouskaya
,
A.
,
Warren
,
J.
,
Lamar
,
J.
,
Martino
,
N.
,
Adam
,
A. P.
and
Barroso
,
M.
(
2020
).
Complex Rab4-mediated regulation of endosomal size and EGFR activation
.
Mol. Cancer Res.
18
,
757
-
773
.
Uribe-Querol
,
E.
and
Rosales
,
C.
(
2020
).
Phagocytosis: our current understanding of a universal biological process
.
Front. Immunol.
11
,
1066
.
Varnai
,
P.
and
Balla
,
T.
(
1998
).
Visualization of phosphoinositides that bind pleckstrin homology domains: calcium- and agonist-induced dynamic changes and relationship to myo-[3H]inositol-labeled phosphoinositide pools
.
J. Cell Biol.
143
,
501
-
510
.
Wood
,
T. R.
,
Chow
,
R. Y.
,
Hanes
,
C. M.
,
Zhang
,
X.
,
Kashiwagi
,
K.
,
Shirai
,
Y.
,
Trebak
,
M.
,
Loegering
,
D. J.
,
Saito
,
N.
and
Lennartz
,
M. R.
(
2013
).
PKC-epsilon pseudosubstrate and catalytic activity are necessary for membrane delivery during IgG-mediated phagocytosis
.
J. Leukoc. Biol.
94
,
109
-
122
.
Yeo
,
J. C.
,
Wall
,
A. A.
,
Luo
,
L.
and
Stow
,
J. L.
(
2015
).
Rab31 and APPL2 enhance FcgammaR-mediated phagocytosis through PI3K/Akt signaling in macrophages
.
Mol. Biol. Cell
26
,
952
-
965
.
Zhang
,
Y.
,
Hoppe
,
A. D.
and
Swanson
,
J. A.
(
2010
).
Coordination of Fc receptor signaling regulates cellular commitment to phagocytosis
.
Proc. Natl. Acad. Sci. USA
107
,
19332
-
19337
.

Competing interests

The authors declare no competing or financial interests.

Supplementary information