We have recently reported that disruption of the actin cytoskeleton enhanced N-formylmethionyl-leucyl-phenylalanine (fMLP)-stimulated granule exocytosis in human neutrophils but decreased plasma membrane expression of complement receptor 1 (CR1), a marker of secretory vesicles. The present study was initiated to determine if reduced CR1 expression was due to fMLP-stimulated endocytosis, to determine the mechanism of this endocytosis, and to examine its impact on neutrophil functional responses. Stimulation of neutrophils with fMLP or ionomycin in the presence of latrunculin A resulted in the uptake of Alexa fluor 488-labeled albumin and transferrin and reduced plasma membrane expression of CR1. These effects were prevented by preincubation of the cells with sucrose, chlorpromazine, or monodansylcadaverine (MDC), inhibitors of clathrin-mediated endocytosis. Sucrose, chlorpromazine, and MDC also significantly inhibited fMLP- and ionomycin-stimulated specific and azurophil granule exocytosis. Disruption of microtubules with nocodazole inhibited endocytosis and azurophil granule exocytosis stimulated by fMLP in the presence of latrunculin A. Pharmacological inhibition of phosphatidylinositol 3-kinase, ERK1/2, and PKC significantly reduced fMLP-stimulated transferrin uptake in the presence of latrunculin A. Blockade of clathrin-mediated endocytosis had no significant effect on fMLP-stimulated phosphorylation of ERK1/2 in neutrophils pretreated with latrunculin A. From these data, we conclude that the actin cytoskeleton functions to limit microtubule-dependent, clathrin-mediated endocytosis in stimulated human neutrophils. The limitation of clathrin-mediated endocytosis by actin regulates the extent of both specific and azurophilic granule exocytosis.
- signal transduction
- intracellular granules
endocytosis is a cell-type and cargo-specific mechanism by which the plasma membrane and material from the external environment are internalized (9, 46). There are two broad mechanisms of endocytosis: phagocytosis and pinocytosis. Phagocytosis is used by specialized cells, including macrophages, monocytes, and neutrophils, to clear mircoorganisms and other large opsonized particles. Four mechanisms of pinocytosis have been described: macropinocytosis, clathrin-mediated endocytosis, caveolae-mediated endocytosis, and clathrin- and caveolae-independent endocytosis (9). Macropinocytosis is a mechanism by which relatively large amounts of extracellular fluid and solutes are nonspecifically internalized. Constitutive and ligand-stimulated clathrin-mediated endocytosis occurs in all mammalian cells and is essential for the uptake of key nutrients, such as iron-laden transferrin (9). Ligand-stimulated clathrin-mediated endocytosis results in the localization of transmembrane receptors and their bound ligand in areas of the plasma membrane known as “coated pits,” the main constituent of which is clathrin. These coated pits are internalized followed by scission from the plasma membrane and release of clathrin-coated vesicles into the cytoplasm (9, 23, 53). Caveolae are plasma membrane invaginations formed from cholesterol and sphingolipid-rich microdomains, termed lipid rafts, that are morphologically different from clathrin-coated pits. The mechanisms of clathrin- and caveolin-independent endocytosis are unknown (25, 26). These diverse endocytic pathways function to control a large number of cell processes, including immune surveillance and innate immunity, antigen presentation, nutrient uptake, neurotransmitter release, signal transduction, and protein degradation and trafficking (9, 45, 46, 48).
Actin reorganization has been shown to be required for both phagocytosis and pinocytosis. When particles or opsonized pathogens are engulfed by phagocytosis, pseudopod extensions of the plasma membrane are driven by actin polymerization to encircle the phagocytic target (29, 32). Similarly, actin reorganization is required to form the membrane protrusions leading to macropinocytosis (9). Reorganization of the actin cytoskeleton is postulated to be required at multiple steps of clathrin-mediated endocytosis, including providing the spatial organization of the endocytic machinery, inducing the curvature of the plasma membrane at coated pits, driving the separation of early clathrin-coated endosomes from the plasma membrane, inducing scission of the endosome, and propelling endosome internalization (9, 16, 23, 31). Most, but not all, studies (6, 14, 41) have reported that pharmacological inhibition of actin reorganization prevented endocytosis.
Neutrophils are the primary cellular components of innate immunity. These cells perform multiple forms of endocytosis, including phagocytosis of pathogens and pinocytosis via clathrin-dependent and -independent mechanisms. We (22) have recently reported that disruption of the actin cytoskeleton enhanced the rate and extent of granule exocytosis after stimulation of neutrophils with N-formylmethionyl-leucyl-phenylalanine (fMLP). On the other hand, plasma membrane expression of a marker of secretory vesicles, complement receptor 1 (CR1), was reduced. The present study was initiated to determine if reduced CR1 expression was due to enhanced fMLP-stimulated endocytosis after disruption of the actin cytoskeleton, to determine the mechanism of this endocytosis, and to examine its impact on neutrophil functional responses. The results show that disruption of the actin cytoskeleton allows microtubule-dependent, clathrin-mediated endocytosis in stimulated human neutrophils and that this endocytosis participates in exocytosis of specific and azurophilic granules.
MATERIALS AND METHODS
Alexa fluor 488-conjugated BSA, Alexa fluor 488-conjugated transferrin, and latrunculin A were from Molecular Probes (Eugene, OR). Cytochalasin D, sucrose, chlorpromazine, monodansylcadaverine (MDC), nystatin, nocodazole, calphostin C, genistein, PP2, ionomycin, and fMLP were from Sigma (St. Louis, MO). The p38 MAPK inhibitor SB-203580 and the MEK1/2 inhibitor PD-098059 were from Calbiochem (La Jolla, CA). The phosphatidylinositol 3-kinase (PI3K) inhibitor LY-294002 was from BioMol (Plymouth Meeting, PA). Phospho-ERK1/2 and total ERK1/2 antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA). FITC-conjugated monoclonal anti-CR1 (CD35) and mouse IgG1 were from Pharmingen (San Diego, CA). FITC-conjugated monoclonal anti-CD66b was from Accurate Chemical (Westbury, NY). FITC-conjugated monoclonal anti-CD63 was from Ancell (Bayport, MN).
Neutrophils were isolated from the blood of healthy donors using plasma-Percoll gradients as previously described (20). After isolation, neutrophils were washed and resuspended in bacterial LPS-free Krebs-Ringer phosphate buffer (pH 7.2) containing 0.2% dextrose (Krebs). Microscopic evaluation of isolated cells showed that >97% of cells were neutrophils. Trypan blue exclusion indicated that >97% cells were viable. The Institutional Review Board of the University of Louisville approved the use of human donors.
To quantify the internalization of BSA or transferrin receptors, neutrophils were incubated with 2 μg/ml Alexa fluor 488-conjugated BSA or 5 μg/ml Alexa fluor 488-conjugated transferrin in the presence and absence of 1 μM latrunculin A or 30 μM cytochalasin D for 30 min at 37°C followed by stimulation with or without 300 nM fMLP for 3 min or 250 nM ionomycin for 15 min. Cells were then washed with 1 mM sodium azide, fixed in paraformaldehyde, and analyzed for internalized Alexa fluor 488 by flow cytometry using mean channel fluorescence intensity as a quantitative measure of uptake. Aliquots of cells were also imaged by a Zeiss Axiovert 100 M confocal microscope with a Zeiss Plan-Neofluar ×100/1.3 oil-immersion lens using LSM510 (version 3.2) software. The excitation wavelength from an argon laser was set at 488 nm and emission was set at 505 nm.
Exocytosis of secretory vesicles, specific granules, and azurophil granules was determined by measuring an increase in plasma membrane expression of CR1 (also known as CD35), CD66b, and CD63, respectively, using flow cytometry as previously described (22, 58). Specific granule exocytosis was also determined by measuring lactoferrin release into neutrophil supernatants using an in-house ELISA. NUNC immunoplates were coated with anti-lactoferrin antibody (1:1,000 dilution of a 4 mg/ml preparation, MP Biomedicals, Aurora, OH) at 37°C. The plate was washed and then blocked with 1% BSA in BBL FTA hemagglutination buffer (BD, Sparks, MD) at 4°C. Lactoferrin standards (Sigma) were made by serial dilution in 1% BSA. Diluted samples and standards were assayed in triplicate. Wells were loaded and incubated at 37°C for 1 h, washed, and then incubated for 1 h at 37°C with peroxidase-conjugated anti-lactoferrin (ICN, Aurora, OH). Wells were washed again and exposed to peroxidase substrate for 3 min. Reactions were stopped with 8 N H2SO4. Plates were read on a Spectramax plate reader at 490 nm using a log-logit standard curve. Azurophil granule exocytosis was also determined by measuring myeloperoxidase (MPO) release into neutrophil supernatants using an ELISA (Assay Designs, Ann Arbor, MI) per the manufacturer's instructions.
Colocalization of neutrophil granule markers and transferrin.
Neutrophils (5 × 106 cells/ml) were incubated with 5 μg/ml Alexa fluor 488-conjugated transferrin in the presence of 1 μM latrunculin A for 30 min at 37°C followed by stimulation with 300 nM fMLP for 3 min. Cells were washed with Krebs solution, fixed in 3.7% paraformaldehyde for 15 min at room temperature, permeabilized with 2% saponin for 15 min at room temperature, and then incubated with phycoerythrin-conjugated CR1, CD66b, or CD63 (final concentration: 8 μg/ml). Cells were imaged by a Zeiss Axiovert 100 M confocal microscope with a Zeiss Plan-Neofluar ×100/1.3 oil-immersion lens using LSM510 (version 3.2) software.
Western blot analysis.
Neutrophils (1 × 107 cells/ml) were incubated for 30 min at 37°C in the presence or absence of 225 mM sucrose, 40 μM chlorpromazine, or 350 μM MDC and 1 μM latrunculin A, followed by stimulation with 300 nM fMLP for 1 min. Stimulation was ended by centrifugation at 2,500 g for 20 s, and pelleted cells were lysed in 100 μl of ice-cold lysis buffer [20 mM Tris·HCl (pH 7.5), 150 mM NaCl, 1% (vol/vol) Triton X-100, 0.5% (vol/vol) Nonidet P-40, 20 mM NaF, 20 mM NaVO3, 1 mM EDTA, 1 mM EGTA, 5 mM PMSF, 21 μg/ml aprotinin, and 5 μg/ml leupeptin]. Samples were centrifuged at 15,000 g for 15 min at 4°C. Equal amounts of protein (50 μg) were separated by 10% SDS-PAGE and subjected to standard immunoblot procedures. Blots were probed with phospho-ERK1/2 (1:1,000) or total ERK1/2 (1:1,000) in 1% milk and Tris-buffered saline-Tween 20. The appropriate secondary antibodies were used at 1:2,000. Protein signals were visualized by enhanced chemiluminescence (Amersham Pharmacia Biotech, Piscataway, NJ) as directed by the manufacturer. Densitometry analysis of the Western blot bands was performed with ImageQuant software (Molecular Dynamics, Sunnyvale, CA).
All data are expressed as means ± SE. Statistical analysis was performed using ANOVA with the Tukey-Kramer multiple-comparison test. Differences were considered statistically significant when P ≤ 0.05.
Disruption of the actin cytoskeleton enhances clathrin-mediated endocytosis.
To determine if disruption of the actin cytoskeleton resulted in enhanced endocytosis in fMLP-stimulated neutrophils, Alexa fluor-488-conjugated BSA uptake was measured by flow cytometry. As shown in Fig. 1A, pretreatment of neutrophils with 1 μM latrunculin A, a concentration that disrupts basal and fMLP-stimulated F-actin (22), resulted in a significant increase in Alexa fluor-488-conjugated BSA uptake after stimulation with 300 nM fMLP compared with untreated cells. Neither fMLP nor latrunculin A alone induced significant BSA uptake, indicating that both disruption of the actin cytoskeleton and a subsequent stimulus were necessary to observe endocytosis of BSA by this method.
Transferrin binding to its receptor induces receptor internalization through a clathrin-mediated mechanism (9, 54), and endocytosis of ligand-bound chemoattractant receptors in human neutrophils is clathrin dependent (30, 50, 59). Clathrin-mediated endocytosis has been reported to be inhibited by hypertonicity (10, 21), MDC (1, 7, 42, 47), and chlorpromazine (1, 56). To determine if the combination of disruption of the actin cytoskeleton and fMLP stimulation induced clathrin-mediated endocytosis, transferrin uptake was measured in neutrophils incubated with or without hypertonic sucrose, chlorpromazine, or MDC before latrunculin A treatment and fMLP stimulation. As shown in Fig. 1B, latrunculin A pretreatment, followed by stimulation with fMLP, resulted in a significant increase in Alexa fluor-488-conjugated transferrin uptake, which was blocked by incubation with hypertonic sucrose, chlorpromazine, or MDC for 30 min. These data demonstrate that clathrin-mediated endocytosis was enhanced by disruption of the actin cytoskeleton and that all three inhibitors (hypertonic sucrose, chlorpromazine, and MDC) effectively inhibited clathrin-mediated endocytosis in human neutrophils. To determine if clathrin-mediated endocytosis was also responsible for the previously observed albumin uptake, neutrophils were incubated with hypertonic sucrose, chlorpromazine, or MDC before latrunculin A treatment and fMLP stimulation. As shown in Fig. 1A, incubation with hypertonic sucrose, chlorpromazine, or MDC for 30 min completely inhibited fMLP-stimulated internalization of albumin in latrunculin A-treated cells. Additionally, albumin and transferrin uptake were observed by confocal microscopy. These observations showed that labeled albumin and transferrin were internalized, not associated with the plasma membrane, when neutrophils pretreated with latrunculin A were stimulated with fMLP (see Fig. 3, for example). Moreover, incubation of neutrophils with hypertonic sucrose, chlorpromazine, or MDC significantly prevented internalization of labeled albumin and transferrin, consistent with the flow cytometric data (data not shown).
Nystatin, a cholesterol-aggregating drug that blocks clathrin-independent endocytosis (44, 50), failed to alter fMLP-stimulated albumin or transferrin uptake (data not shown). Neutrophils pretreated with 30 μM cytochalasin D, which disrupts the actin cytoskeleton by a mechanism different from that of latrunculin A, also induced a significant increase in Alexa fluor-488-conjugated transferrin uptake after stimulation with fMLP (data not shown).
Clathrin-mediated endocytosis regulates neutrophil granule exocytosis.
To determine if endocytosis plays a role in fMLP-stimulated exocytosis, neutrophils were incubated with or without latrunculin A before stimulation with fMLP in the presence or absence of hypertonic sucrose, chlorpromazine, or MDC. As reported previously (22), fMLP or latrunculin A alone stimulated a significant increase in plasma membrane expression of CR1, a marker of secretory vesicles, whereas fMLP stimulation of neutrophils pretreated with latrunculin A resulted in decreased CR1 expression (Fig. 2A). To determine if that reduction in CR1 expression was due to clathrin-mediated endocytosis, we examined CR1 expression in neutrophils stimulated for 10 min by fMLP with or without latrunculin A pretreatment in the presence or absence of hypertonic sucrose, chlorpromazine, or MDC. Pretreatment with hypertonic sucrose, chlorpromazine, or MDC significantly reduced CR1 expression stimulated by fMLP alone (P < 0.05), and chlorpromazine and MDC significantly reduced latrunculin A-stimulated CR1 expression (P < 0.001 and P < 0.05, respectively). The decrease in CR1 expression stimulated by fMLP plus latrunculin A was prevented by pretreatment with hypertonic sucrose (P < 0.001), chlorpromazine (P < 0.01), and MDC (P < 0.05). Nystatin had no effect on CR1 expression under those same conditions (data not shown). These data suggest that fMLP normally stimulates secretory vesicle exocytosis, resulting in increased CR1 expression, whereas reuptake of CR1 by clathrin-mediated endocytosis in stimulated neutrophils is inhibited by an intact actin cytoskeleton.
Stimulation of neutrophils with fMLP, latrunculin A, or the combination of the two induced a significant increase in the expression of CD66b, a marker of specific granules (Fig. 2B). Both chlorpromazine and MDC significantly reduced CD66b expression stimulated by fMLP, latruculin A, or the two together, whereas hypertonic sucrose pretreatment had no effect on CD66b expression under any of those conditions (Fig. 2B). To further evaluate specific granule exocytosis, lactoferrin release was measured in neutrophils stimulated with fMLP, latrunculin A, or the combination of the two. As shown in Fig. 2C, lactoferrin release was significantly increased in the presence of latrunculin A followed by fMLP stimulation, and both chlorpromazine and MDC significantly reduced lactoferrin release. The partial inhibition by chlorpromazine and MDC suggests that endocytosis contributes to optimal specific granule exocytosis, but is not a necessary component for this process.
As previously reported (22), neither fMLP nor latrunculin A alone increased the expression of CD63, a marker of azurophil granules (Fig. 2D). Pretreatment with latrunculin A followed by stimulation with fMLP resulted in a significant increase in CD63 expression (P < 0.001). Pretreatment with hypertonic sucrose or MDC significantly inhibited azurophil granule exocytosis under those conditions. No differences in CD63 expression were observed with chlorpromazine pretreatment. MPO release from azurophil granules exhibited a pattern similar to that of CD63 expression (Fig. 2E). A significant increase in MPO release occurred when neutrophils were pretreated with latrunculin A before fMLP stimulation. MPO release was significantly reduced by pretreatment with either chlorpromazine or MDC (Fig. 2E). These results suggest that clathrin-mediated endocytosis is required for optimal azurophil granule exocytosis.
Colocalization of markers of clathrin-mediated endocytosis and neutrophil granules.
Our flow cytometric data showed that pretreatment of human neutrophils with latrunculin A followed by fMLP stimulation enhanced transferrin uptake and reduced CR1 expression. To determine if these events were connected, neutrophils were incubated with latrunculin A before stimulation with fMLP in the presence of Alexa fluor-488-conjugated transferrin, stained for CR1, and then examined by confocal microscopy. Although CR1 was present in the cytosol independent of transferrin, transferrin always colocalized with CR1 (Fig. 3A), supporting the conclusion that CR1 is internalized by clathrin-mediated endocytosis. To determine if endosomes and specific or azurophil granules interact during exocytosis, colocalization of transferrin with CD66b or CD63 was determined by confocal microscopy after fMLP stimulation in the presence of latrunculin A. Both CD66b (a marker of specific granules) and CD63 (a marker of azurophil granules) were detected in the cytosol and at the plasma membrane (Fig. 3, B and C, in red); however, neither granule marker colocalized with transferrin (Fig. 3, B and C). These findings do not support endosome-granule fusion as the mechanism by which endocytosis contributes to exocytosis.
Ionomycin-stimulated exocytosis and endocytosis.
Previous reports have shown that formyl peptide receptors undergo clathrin-mediated endocytosis after ligand binding (50), and clathrin-mediated endocytosis of G protein-coupled receptors results in the activation of some signal transduction pathways (27, 28, 51, 59). To determine if the effects of hypertonic sucrose, chlorpromazine, and MDC were due to inhibition of endocytosis of ligand-bound formyl peptide receptors, we examined the effect of disruption of the actin cytoskeleton on endocytosis and exocytosis stimulated by a receptor-independent mechanism. Similar to the results with fMLP, pretreatment of neutrophils with latrunculin A significantly enhanced endocytosis of Alexa fluor-488-conjugated BSA (Fig. 4A) and Alexa fluor-488-conjugated transferrin (Fig. 4B) stimulated by the calcium ionophore ionomycin. That enhanced endocytosis was blocked by pretreatment with hypertonic sucrose, chlorpromazine, and MDC.
The ability of ionomycin to stimulate exocytosis of secretory vesicles, specific granules, and azurophil granules in the presence and absence of latrunculin A and clathrin-mediated endocytosis inhibitors is shown in Fig. 5, A–C. Incubation for 15 min with ionomycin stimulated increased plasma membrane expression of CR1 and CD66b but not CD63. Pretreatment with latrunculin A, followed by stimulation with ionomycin, resulted in a significant decrease in CR1 expression and a significant increase in CD63 expression. CD66b expression did not differ between treatments with ionomycin alone and ionomycin plus latrunculin A. Similar to the results with fMLP, blockade of clathrin-mediated endocytosis with hypertonic sucrose, chlorpromazine, or MDC prevented the reduction in CR1 expression and the increase in CD63 expression under these conditions (Fig. 5, A and C). Inhibition of clathrin-mediated endocytosis by chlorpromazine or MDC significantly reduced CD66b expression stimulated by ionomycin, latrunculin A, or the two together (Fig. 5B). Whereas hypertonic sucrose significantly reduced ionomycin-induced CD66b expression, it had no effect after the disruption of the actin cytoskeleton with latrunculin A (Fig. 5B). These data suggest that the relationship between clathrin-mediated endocytosis and specific and azurophil granule exocytosis is independent of endocytosis of ligand-bound receptors.
Clathrin-mediated endocytosis after disruption of the actin cytoskeleton is dependent on microtubules.
Previous studies (23, 41) concluded that the actin cytoskeleton participated in endocytosis by providing the force necessary for the deformation of the plasma membrane, for the internalization and scission of clathrin-coated vesicles, and/or the intracellular mobility of endocytic vesicles. As our results showed enhanced clathrin-mediated endocytosis with disruption of the actin cytoskeleton, we evaluated the role of the microtubule cytoskeleton in neutrophil clathrin-mediated endocytosis. Neutrophils were pretreated with latrunculin A, in the presence or absence of nocodazole, before stimulation with fMLP in the presence of Alexa fluor-488-conjugated transferrin. As shown in Fig. 6, nocodazole significantly inhibited transferrin uptake, suggesting that microtubules participate in clathrin-mediated endocytosis in human neutrophils.
The effect of disruption of microtubules with nocodazole on exocytosis of the different granule subsets is shown in Fig. 7. Similar to the results with inhibition of endocytosis, pretreatment with nocodazole significantly restored the increase in CR1expression and blunted the decrease in CD63 expression after stimulation by fMLP in the presence of latrunculin A. However, disruption of microtubules had no effect on fMLP-stimulated CD66b expression. These data suggest that microtubule reorganization is necessary for clathrin-mediated endocytosis, CR1 internalization, and azurophil granule exocytosis in human neutrophils.
Signal transduction pathways mediating endocytosis.
A number of signaling pathways, including those containing p38 MAPK, PI3K, and nonreceptor tyrosine kinases, regulate exocytosis in human neutrophils (17, 37, 55, 58). To evaluate the role of these and other kinases in clathrin-mediated endocytosis, neutrophils were pretreated with pharmacological inhibitors of various kinases before measurements of fMLP-stimulated transferrin uptake. As shown in Table 1, pretreatment with PD-098059 (an ERK inhibitor), LY-294002 (a PI3K inhibitor), genistein (a nonspecific tyrosine kinase inhibitor), or calphostin C (a PKC inhibitor) significantly reduced transferrin uptake in human neutrophils treated with a combination of latrunculin A and fMLP. Inhibition of p38 MAPK with SB-203580 and Src tyrosine kinases with PP2 had no effect. The concentrations of PP2 and SB-203580 used in these experiments have been shown to inhibit substrate phosphorylation by Src tyrosine kinases and p38 MAPK (data not shown). Thus, despite the relationship between endocytosis and exocytosis, distinctly different patterns of signal transduction pathways mediate fMLP-stimulated endocytosis and exocytosis in human neutrophils.
The relationship between ERK1/2 activity and endocytosis was of particular interest as the activation of ERK through β-arrestins has been reported to be dependent on clathrin-mediated endocytosis of G protein-coupled receptors (27, 28, 51). Thus, we evaluated ERK1/2 phosphorylation stimulated by fMLP under conditions of enhanced and inhibited endocytosis. As shown in Fig. 8, fMLP-stimulated ERK1/2 phosphorylation was modestly enhanced by pretreatment with latrunculin A. Treatment with hypertonic sucrose, chlorpromazine, or MDC had no effect on fMLP-stimulated ERK1/2 phosphorylation in either the presence or absence of latrunculin A. These data suggest that receptor endocytosis plays no role in fMLP-stimulated ERK1/2 activation, and disruption of the actin cytoskeleton enhances ERK1/2 activation independently of the effect on endocytosis.
We (22) have previously reported that the actin cytoskeleton acts as a barrier to stimulated exocytosis of neutrophil granules. The results of the present study demonstrate that the actin cytoskeleton also functions to limit clathrin-mediated endocytosis after stimulation of human neutrophils. This conclusion is supported by data showing that disruption of the actin cytoskeleton by latrunculin A or cytochalasin D, followed by stimulation with fMLP or ionomycin, induced a significant increase in the uptake of albumin and transferrin. Endocytosis of transferrin is clathrin mediated (9, 54), and endocytosis of both albumin and transferrin was inhibited by hypertonic sucrose, chlorpromazine, and MDC, agents that block clathrin-mediated endocytosis (1, 7, 10, 21, 42, 47, 56). Although neutrophils lack caveolin (49), they still form lipid raft domains. Additionally, endocytosis of IL-2 receptors in lymphocytes lacking caveolin 1 occurred through a clathrin-independent, cholesterol-sensitive pathway (25, 26). Our data showed that nystatin, a cholesterol-aggregating drug, did not alter albumin or transferrin uptake, ruling out a caveolin-like mechanism of endocytosis. Previous studies (6, 9, 14, 23, 31, 41) have indicated that actin reorganization was required for, or in some circumstances played no role in, clathrin-mediated endocytosis. However, Puthenveedu and von Zastrow (40) reported that some endocytic cargo interacted with the actin cytoskeleton, and they postulated that this interaction delayed endocytosis. Our data indicate that the actin cytoskeleton can impair endocytosis in some cells, although the mechanism for that inhibition was not determined. Thus, our results identify a new role for the actin cytoskeleton in clathrin-mediated endocytosis.
Our previous study (22) showed that disruption of the actin cytoskeleton enhanced the rate and extent of fMLP-stimulated exocytosis of neutrophil gelatinase, specific, and azurophil granules. Whereas stimulation with fMLP alone increased plasma membrane expression of the secretory vesicle marker CR1, pharmacological disruption of the actin cytoskeleton resulted in a time-dependent reduction in CR1 expression after fMLP stimulation. The ability of hypertonic sucrose, chlorpromazine, and MDC to prevent the loss of CR1 expression under these conditions suggests that the reduced CR1 expression observed previously was due to clathrin-mediated endocytosis of CR1 in the absence of interaction with its ligand. The colocalization of CR1 and transferrin-containing endosomes by confocal microscopy is consistent with this conclusion. Using immunogold labeling and electron microscopy, Berger and colleagues (3, 4, 52) demonstrated that fMLP and ionomycin stimulation of human neutrophils with an intact actin cytoskeleton induced a 6- to 10-fold increase in plasma membrane CR1 expression due to exocytosis of secretory vesicles. This translocation was rapidly followed by ligand-independent endocytosis and subsequent degradation of CR1. However, we were unable to detect endocytosis of CR1, albumin, or transferrin without disruption of the actin cytoskeleton by confocal microscopy.
Three agents that inhibited clathrin-mediated endocytosis by different mechanisms were used to determine the relationship between endocytosis and exocytosis of specific and azurophil granules. Both chlorpromazine and MDC significantly inhibited specific granule exocytosis stimulated by fMLP or ionomycin with or without latrunculin A pretreatment. On the other hand, hypertonic sucrose pretreatment inhibited specific granule exocytosis stimulated by fMLP or ionomycin alone but failed to alter exocytosis stimulated in the presence of latrunculin A. This observation may be explained by the ability of osmotic stress to induce the rigid subcortical actin polymerization that would be expected to inhibit exocytosis (43). As we have previously reported (22), azurophil granule exocytosis could only be stimulated after disruption of the actin cytoskeleton. All three inhibitors of clathrin-mediated endocytosis dramatically reduced azurophil granule exocytosis stimulated by ionomycin, whereas sucrose and MDC significantly inhibited azurophil granule exocytosis stimulated by fMLP. The reason for the failure of chlorpromazine to inhibit exocytosis of this granule subset is not clear. Taken together, these data suggest that the extent of clathrin-mediated endocytosis is one determinant of the level of specific and azurophil granule exocytosis. The data also suggest that the regulation of exocytosis by the actin cytoskeleton may be through control of endocytosis rather than a direct effect on granule access to the plasma membrane. Although the mechanism by which endocytosis regulates exocytosis was not examined in the present study, the ability of inhibition of endocytosis to block both fMLP- and ionomycin-stimulated azurophil granule exocytosis indicates that the effect of endocytosis is not mediated by internalization of ligand-bound receptors. Fittschen and Henson (13) previously proposed a model for the interaction between endosomes and azurophil granules. By ultrastructural analysis, they showed that endosomes fused with azurophil granules and suggested that endosomal membranes provided the contact sites for recognition and fusion with the plasma membrane. The failure of transferrin-containing endosomes to colocalize with a marker of azurophil granules in the present study, however, does not support this hypothesis.
Our observation that endocytosis regulates azurophil granule exocytosis provides one explanation for the observation that disruption of the actin cytoskeleton is required for stimulated exocytosis of azurophil granules (22). Botelho and colleagues (5) described pinocytosis occurring in the vicinity of FcγR-mediated phagocytosis and reported that disruption of the actin cytoskeleton blocked phagocytosis but enhanced pinocytosis. Similarly, Bauer and Tapper (2) described pinocytosis at the sites of neutrophil phagocytosis of gram-positive bacteria. Both groups determined that exocytosis of azurophil granules occurred at the sites of phagocytosis and that pinocytosis correlated with azurophil granule exocytosis. These reports, coupled with our data, suggest that under physiological conditions, actin reorganization at sites of phagocytosis leads to clathrin-mediated endocytosis, and this localized endocytosis contributes to azurophil granule exocytosis at sites of phagosome formation. We speculate that the ability of the actin cytoskeleton to limit endocytosis-dependent azurophil granule exocytosis may serve to focus the release of the toxic granule enzymes only at sites of phagocytosis.
Enhanced endocytosis in the presence of disruption of the actin cytoskeleton suggested that another cytoskeletal element provided the mechanical force necessary for endosome formation and separation from the plasma membrane. Our data suggest that microtubule reorganization is involved in early stages of endocytosis in human neutrophils. Additionally, pharmacological disruption of microtubules significantly reduced fMLP-stimulated azurophil granule exocytosis and CR1 endocytosis in the presence of latrunculin A. These observations are in agreement with those of Botelho et al. (5), who showed that phagocytosis-induced pinocytosis was reduced in colchicine-treated neutrophils. Our results support a model in which clathrin-mediated endocytosis in neutrophils undergoes antagonistic regulation by the actin and microtubule cytoskeletons. The microtubule cytoskeleton provides the mechanical force necessary for endocytosis, whereas the actin cytoskeleton acts to inhibit endocytosis. Endocytosis is necessary for optimal exocytosis to proceed after neutrophil stimulation.
A number of signaling pathways activated by G protein-coupled receptors regulate membrane trafficking, including those containing p38 MAPK, ERK1/2, PI3K, and nonreceptor tyrosine kinases (8, 15, 24, 30, 36, 39, 51). Botelho et al. (5) showed that inhibition of PI3K with wortmannin significantly reduced FcγR-mediated pinocytosis in human neutrophils. In the present study, pretreatment with a different inhibitor of PI3K, LY-294002, almost completely blocked transferrin uptake in fMLP-stimulated neutrophils in which the actin cytoskeleton was disrupted.
Ligand-induced, clathrin-mediated endocytosis has been demonstrated previously for a number of neutrophil chemoattactant receptors, including those for formylated peptides, platelet-activating factor, and leukotriene B4 (12, 17, 30, 55). It has been reported that the platelet-activating factor receptor requires clathrin-mediated endocytosis and recruitment of β-arrestin-1 and the p38 MAPK signalosome to transduce its signal (30). Endocytosis of ligand-receptor complexes stimulates ERK1/2 activation in many cell types (38), and inhibition of ERK1/2 activity significantly inhibited fMLP-stimulated endocytosis in the present study. Although fMLP-stimulated ERK1/2 activation was modestly increased under conditions that enhanced clathrin-mediated endocytosis, inhibition of clathrin-mediated endocytosis had no effect on ERK1/2 activation. These data suggest that formyl peptide receptor-induced ERK1/2 activation occurs independently of, but is required for, clathrin-mediated endocytosis.
Our data using pharmacological inhibitors suggest that tyrosine kinases and PKC also play a role in clathrin-mediated endocytosis in human neutrophils, whereas p38 MAPK and Src family tyrosine kinases are not involved. Previous studies (33–35, 57, 58) have shown that p38 MAPK and Src tyrosine kinases, but not ERK1/2, play a significant role in granule exocytosis in neutrophils. Although a more detailed analysis of the signal transduction pathways regulating endocytosis and exocytosis is required, our data suggest that distinctly different signal transduction pathways regulate exocytosis and endocytosis in human neutrophils.
This work was supported by a Department of Veterans Affairs Merit Review Grant (to K.R. McLeish), by National Institute of Diabetes and Digestive and Kidney Diseases Grant R21-DK-62389 (to R. A. Ward and K. R. McLeish), and American Heart Association Ohio Valley Affiliate Predoctoral Fellowship Grant 0415191B (to N. R. Jog) and Beginning Grant-In-Aid 0765387B (to S. M. Uriarte).
We thank Karen Brinkley and Terri Manning for the excellent technical assistance.
Present address of N. R. Jog: Sect. of Rheumatology, Dept. of Medicine, Temple Univ., Philadelphia, PA 19140.
↵* S. M. Uriarte and N. R. Jog contributed equally to this work.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2009 the American Physiological Society