University at Buffalo - The State University of New York
Skip to Content
Cytokine Secretion via Cholesterol-rich Lipid Raft-associated SNAREs at the Phagocytic Cup

Cytokine Secretion via Cholesterol-rich Lipid Raft-associated SNAREs at the Phagocytic Cup*

  1. Jennifer L. Stow1
  1. Institute for Molecular Bioscience, University of Queensland, Brisbane, Queensland 4072, Australia
  1. 1 To whom correspondence should be addressed: Institute for Molecular Biosciences, the University of Queensland, Brisbane, Queensland 4072, Australia. Tel.: 61-7-3346-2034; Fax: 61-7-3346-2101; E-mail: j.stow{at}


Lipopolysaccharide-activated macrophages rapidly synthesize and secrete tumor necrosis factor α (TNFα) to prime the immune system. Surface delivery of membrane carrying newly synthesized TNFα is controlled and limited by the level of soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) proteins syntaxin 4 and SNAP-23. Many functions in immune cells are coordinated from lipid rafts in the plasma membrane, and we investigated a possible role for lipid rafts in TNFα trafficking and secretion. TNFα surface delivery and secretion were found to be cholesterol-dependent. Upon macrophage activation, syntaxin 4 was recruited to cholesterol-dependent lipid rafts, whereas its regulatory protein, Munc18c, was excluded from the rafts. Syntaxin 4 in activated macrophages localized to discrete cholesterol-dependent puncta on the plasma membrane, particularly on filopodia. Imaging the early stages of TNFα surface distribution revealed these puncta to be the initial points of TNFα delivery. During the early stages of phagocytosis, syntaxin 4 was recruited to the phagocytic cup in a cholesterol-dependent manner. Insertion of VAMP3-positive recycling endosome membrane is required for efficient ingestion of a pathogen. Without this recruitment of syntaxin 4, it is not incorporated into the plasma membrane, and phagocytosis is greatly reduced. Thus, relocation of syntaxin 4 into lipid rafts in macrophages is a critical and rate-limiting step in initiating an effective immune response.

In response to pathogens, activated macrophages produce and secrete tumor necrosis factor α (TNFα),2 a proinflammatory cytokine responsible for the activation and recruitment of cells necessary to mount a successful immune response (1). Lipopolysaccharide (LPS), a bacterial membrane component, is a potent activator of macrophages, binding to the CD14-MD2-TLR4 complex at the surface (2) that signals the induction of widespread gene transcription, including that of TNFα (3, 4). This ensures the rapid and abundant synthesis of the 26-kDa transmembrane form of TNFα, which initially accumulates in the Golgi complex (5) and is then trafficked to the cell surface for proteolytic cleavage by the TNFα-converting enzyme (TACE) (6). We have shown recently that TNFα is transported from the trans-Golgi network to the recycling endosome and is then delivered to the cell surface via fusion of the recycling endosome membrane at the site of the phagocytic cup formation (7, 8). Two distinct steps of membrane fusion are therefore required during the post-Golgi transport of TNFα.

Membrane fusion of TNFα transport vesicles is mediated by members of the soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) protein family (7, 8). Cognate pairing of an R-SNARE (generally on the vesicle) and a Q-SNARE complex (generally on the target membrane) is necessary at each trafficking step for specificity in vesicle docking and membrane fusion (9, 10). A variety of SNARE proteins is expressed in macrophages. The trans-Golgi Q-SNARE complex of syntaxin 6/syntaxin 7/Vti1b pairs with VAMP3 on the recycling endosome in the first step of post-Golgi transport of TNFα. Next the VAMP3 pairs with the cell surface Q-SNARE complex of syntaxin 4/SNAP-23 for TNFα delivery to the cell surface (7, 8, 11). Syntaxin 4, syntaxin 6, Vti1b, SNAP-23, and VAMP3 are up-regulated during LPS activation to accommodate the increased trafficking during TNFα secretion (7, 8, 11). During phagocytosis, the Q-SNARE syntaxin 4 concentrates at the phagocytic cup to mediate fusion of the recycling endosome, providing excess membrane for microorganism engulfment and rapidly delivering TNFα to the cell surface for secretion (7).

Lipid rafts, specific membrane microdomains that are enriched in sterols and sphingolipids, play a number of important roles in immune cells; for instance, they contribute to B cell receptor signaling and antigen uptake (12, 13), ligand-mediated signaling from the T cell receptors (14, 15) and high affinity IgE receptors (16), and play a role in MHCII-mediated antigen presentation (17, 18). In macrophages, phagosomes are enriched in raft proteins (19-21). In other cell types lipid rafts have been shown to organize and regulate SNARE proteins (22, 23). We thus investigated a possible role for lipid rafts in SNARE-mediated delivery of TNFα to the cell surface. We show here that an inducible association of syntaxin 4 with lipid rafts and a cholesterol-rich membrane environment at specific exit sites are required for TNFα secretion.


Antibodies and Reagents—Rabbit polyclonal and rat monoclonal antimouse TNFα antibodies were purchased from Calbiochem and Auspep (Victoria, Australia), respectively. Anti-SNAP-23 and VAMP2 antibodies were purchased from Synaptic Systems (Goettingen, Germany); anti-Gαi-3 antibodies were purchased from DuPont, and antibodies specific for flotillin and syntaxin 6 were purchased from BD Biosciences. Antibodies specific for TACE were purchased from Chemicon (Temecula, CA), and anti-VAMP3 antibodies were purchased from Abcam (Cambridge, UK). Alexa-488 or Alexa-647 conjugated to phalloidin to label F-actin was from Molecular Probes (Eugene, OR), and filipin III was purchased from Sigma. Anti-actin antibody was a gift from Peter Gunning (Children's Hospital at Westmead, Sydney, Australia), and antibodies specific for syntaxin 4 and Munc18c were the kind gifts from David James (Garvan Institute of Medical Research, Sydney, Australia). Candida albicans was kindly provided by Robert Ashman (University of Queensland, Brisbane, Australia).

Cell Culture, Activation, and Cholesterol Depletion—RAW264.7 murine macrophages were grown in RPMI 1640 medium with 10% serum supreme and 1% l-glutamine as described previously (24). Cells were activated with 100 ng/ml LPS (Salmonella minnesota Re 595, Sigma) in the presence or absence of TACE inhibitor (BB-3103, British Biotech Pharmaceuticals). For cholesterol depletion, cells were incubated for 15-30 min in 7.5 mm methyl-β-cyclodextrin (MβCD) (Sigma) in serum-free RPMI 1640.

Phagocytosis—In some experiments macrophages were primed for 18 h in the presence of 500 pg/ml IFNγ (R & D Systems) prior to their incubation with live C. albicans at a ratio of 10:1 (yeast:macrophage) for 10-40 min as described previously (7).

Sucrose Density Gradient Separation—Sucrose density gradient separation of membrane extracts was performed according to published protocols (25). Briefly, macrophages were lysed by 20 passages through a 27-gauge needle in homogenization buffer (10 mm Tris, pH 7.5, 150 mm NaCl, 5 mm EDTA) containing either 0.2 or 1% Triton X-100, and the lysate was centrifuged at 2000 × g for 2 min at 4 °C to pellet unbroken cells and nuclei. The supernatant was loaded onto a 45 to 5% discontinuous sucrose gradient and centrifuged at 200,000 × g in a TLS-55 rotor at 4 °C for 18 h. Eleven fractions (150 μl) were collected from the top of the gradient, plus an additional fraction of 600 μl representing the load fraction. Samples were subjected to SDS-PAGE separation and analyzed by immunoblotting as described previously (5).

Immunofluorescence Staining—Immunofluorescence staining was performed as described previously (5). In some experiments cells grown on coverslips were partially air-dried, and the apical membranes were removed by application (30 s) and removal of damp filter paper (Millipore) (26). The remaining adherent plasma membrane and cell fragments were fixed and immunostained. Cholesterol in cell membranes was stained using 250 μg/ml filipin III in phosphate-buffered saline containing 0.5% bovine serum albumin. Epifluorescence microscopy was performed using an Olympus Provis AX70 microscope equipped with a 100× oil objective and MTI digital camera and NIH Image software. Confocal microscopy was performed using an LSM 510 META (Carl Zeiss Microscope Systems). Three-dimensional reconstructions were generated using LSM 510 META software.

Assays for TNFα Trafficking and Secretion—The trafficking of TNFα from the Golgi complex to the cell surface was measured as described previously using an immunofluorescence-based assay (11). To determine levels of secreted TNFα, a commercial enzyme-linked immunoabsorbent assay kit (OptEIA, BD Biosciences) was used according to the manufacturer's instructions.


SNARE Recruitment to Lipid Rafts—Cholesterol-rich membrane domains and their associated proteins, typically insoluble in nonionic detergents, can be isolated by flotation on sucrose density gradients (27). Macrophage membranes extracted with 1% Triton X-100 were investigated by sucrose density gradient separation and immunoblotting to analyze the distribution of SNARE proteins involved in TNFα secretion (Fig. 1, A and B). Cholesterol-rich lipid raft domains (fractions 5-8) were characterized by the enrichment of known raft marker proteins flotillin (25, 28) and the heterotrimeric Gαi-3 subunit (29) (Fig. 1A). The cell surface Q-SNARE protein SNAP-23 distributed throughout both raft and nonraft fractions, although its Q-SNARE partner syntaxin 4 was recovered almost entirely in nonraft fractions (Fig. 1A). Munc18c, a protein known to bind and regulate syntaxin 4 (30), also partitioned in nonraft fractions (Fig. 1A). The trans-Golgi Q-SNARE syntaxin 6 (8) segregated into nonraft fractions, although the R-SNAREs VAMP2 and VAMP3 were detected in discrete fractions from both raft and nonraft fractions (Fig. 1A).


Syntaxin 4 recruitment to cholesterol-rich lipid raft fractions. A, unstimulated RAW264.7 macrophages lysed in buffer containing 1% Triton X-100 were fractionated by discontinuous sucrose density gradient centrifugation. Fractions 1-12 were analyzed by immunoblotting. Lipid raft fractions 5-8 are denoted by the presence of the raft markers flotillin and Gαi-3. Individual SNARE proteins fractionate in lipid raft and/or nonraft fractions. B, macrophages activated with LPS were extracted and fractionated as above. C, bar chart, with means ± S.E. (n = 4), shows the relative amounts of each protein in lipid raft fractions before and after cell activation; of note is the recruitment of syntaxin 4 to lipid rafts upon activation.

LPS activation of macrophages induces increased expression of syntaxin 4 (11) and initiates a 4-fold recruitment of syntaxin 4 to raft domains, resulting in over 40% of cellular syntaxin 4 now being concentrated in lipid rafts (Fig. 1, B and C). The distributions of other proteins analyzed were unchanged after LPS activation (Fig. 1, B and C). Thus, the differential distributions of the Q-SNARE components SNAP-23 and syntaxin 4 in lipid raft domains and the LPS-induced recruitment of syntaxin 4 to rafts may represent a mechanism for modulating vesiclemediated delivery to the cell surface during TNFα secretion.

Syntaxin 4 Localizes to Cholesterol-dependent Puncta with LPS Activation—We next examined the distribution of syntaxin 4 on the plasma membrane by immunostaining. Because SNARE proteins are typically on the cytoplasmic face of the plasma membrane, staining was performed on patches of ripped open cells exposing the cytoplasmic face of the plasma membrane (26). Staining of F-actin provided orientation for identifying and viewing cell patches. Prior to macrophage activation, syntaxin 4 was localized to distinct puncta (∼0.5-1 μm diameter) randomly distributed on the cytoplasmic face of the plasma membrane (Fig. 2A). Upon macrophage activation, syntaxin 4-stained puncta on the main cell body increased in size (∼1-2 μm), whereas smaller syntaxin 4 puncta (less than 500 nm) also appeared de novo on filopodia (Fig. 2B). Treatment of cells with MβCD depletes cholesterol from cell membranes disrupting cholesterol-dependent lipid raft domains (31, 32). Cholesterol depletion with MβCD resulted in a significant loss of syntaxin 4 puncta from the cell body in activated cells and a complete loss of small syntaxin 4 puncta on filopodia (Fig. 2C). Disruption of cholesterol with MβCD in LPS-activated macrophages also resulted in depletion of syntaxin 4, and the raft markers flotillin and Gαi-3 from sucrose density gradient fractions corresponding to lipid rafts (Fig. 2D). Thus, syntaxin 4, the SNARE responsible for TNFα surface delivery, is found in cholesterol-dependent clusters corresponding to lipid rafts at the plasma membrane. The increased number and size of syntaxin 4 clusters on activated cells is consistent with LPS-induced recruitment of syntaxin 4 to lipid rafts.


Syntaxin 4 localization to cholesterol-dependent puncta. A, staining for syntaxin 4 and F-actin was performed on patches of ripped open macrophages that expose the cytoplasmic face of the plasma membrane. At least 50 cells were analyzed in each of three different experiments. Staining consistently showed syntaxin 4 in distinct puncta; these were 0.5-1 μm and distributed over the cell body. Bar, 10 μm. B, macrophages were activated with LPS for 2 h and treated as above. Syntaxin 4 puncta of increased size (1-2 μm) appeared on the cell body, whereas many smaller puncta (less than 500 nm) now appear especially on filopodia. C, macrophages were activated with LPS for 2 h in the presence of MβCD for the final 30 min and treated as above. The syntaxin 4 puncta seen on macrophage cell body rip offs did not change size or appear on filopodia in cholesterol-depleted cells. D, macrophages activated with LPS for 2 h in the presence or absence of 1% MβCD for the final 30 min were lysed in buffer containing 1% Triton X-100, fractionated by discontinuous sucrose density gradient centrifugation, and analyzed by immunoblotting. Cholesterol depletion resulted in the loss of syntaxin 4 and raft marker proteins from lipid raft fractions 5-8.

Surface Delivery of TNFα Is Cholesterol-dependent—Treatment of LPS-activated macrophages with MβCD revealed that secretion of TNFα, but not its synthesis, is cholesterol-dependent. Secretion of soluble TNFα, as measured by an enzyme-linked immunosorbent assay, was greatly reduced (>60%) after cholesterol depletion in activated cells (Fig. 3A). This reduction in surface delivery of TNFα was confirmed by using an immunofluorescence-based assay that measures the transport of newly synthesized TNFα from the Golgi complex to the cell surface (11). TNFα is labeled on the surface of LPS-activated macrophages (Fig. 3B), although depletion of cholesterol effectively blocks this surface delivery of TNFα (Fig. 3B). LPS signaling through the CD14-MD2-TLR4 complex occurs in lipid rafts, and pretreatment of monocytes with MβCD has been shown to inhibit TNFα secretion (33); however, the p38 and c-Jun N-terminal kinase signaling pathways are reportedly initiated (34). Consistent with this, TNFα staining is present at the level of the Golgi complex, suggesting TNFα surface delivery rather than its synthesis is disrupted (Fig. 3C). In addition, depletion of cholesterol had no effect on the total levels of TNFα present in LPS-activated cells (Fig. 3D). Thus, disruption of cholesterol-dependent lipid rafts effectively reduced TNFα secretion by blocking its surface delivery, without affecting earlier steps in its synthesis or trafficking.


Surface delivery of TNFα is cholesterol-dependent. A, TNFα secretion from unstimulated macrophages, LPS-activated macrophages (2 h), and LPS-activated macrophages (2 h) treated with MβCD for the final 30 min was quantified by an enzyme-linked immunosorbent assay. The results from three experiments are shown in the graph along with the means ± S.E. B, macrophages were treated with LPS for 2 h in the presence of TACE inhibitor with or without MβCD for the final 30 min and immunostained for surface TNFα. In the absence of MβCD TNFα was delivered to the surface of activated cells; however, in cells treated with MβCD surface delivery of TNFα was greatly reduced. C, macrophages were treated with LPS for 2 h in the presence of TACE inhibitor with MβCD for the final 30 min, fixed, permeabilized, and stained to detect intracellular TNFα. TNFα is still being synthesized and is present in the Golgi complex. D, cell lysates from unactivated macrophages, LPS-activated macrophages (1 h), and LPS-activated macrophages with MβCD for the final 30 min (LPS + MβCD) were analyzed by immunoblotting.


TNFα is delivered to syntaxin 4 puncta on filopodia. A, macrophages incubated in the absence or presence of with LPS for 2 h with TACE inhibitor were lysed in buffer containing 0.2% Triton X-100, fractionated by discontinuous sucrose density gradient centrifugation, and analyzed by immunoblotting. Lipid raft fractions 5-8 are denoted by the presence of the raft marker flotillin. B, macrophages stimulated with LPS for 30 min in the presence of TACE inhibitor were fixed and stained for syntaxin 4, TNFα, and F-actin. Three-dimensional reconstructed, surface-rendered, and pseudocolored composite images show the outline of the cell with F-actin staining (green), surface TNFα staining (red), and syntaxin 4 (white). Initial points of delivery of newly synthesized TNFα are concentrated on the surface of filopodia. TNFα and syntaxin 4 staining are seen to colocalize in adjacent images, both appearing at the same points and on the same surface features (arrows).

Cholesterol-dependent Syntaxin 4 Puncta Represent the Delivery Sites for TNFα—Additional sucrose density experiments were performed using 0.2% Triton X-100 (35), producing a cleaner separation of the raft maker flotillin and again showing the LPS-induced recruitment of syntaxin 4 to lipid rafts (Fig. 4A). Activation of macrophages induced the synthesis of full-length TNFα, which accumulates in the presence of TACE inhibitor and is partially lipid raft-associated (Fig. 4A). Confocal imaging was used to examine the relationship between syntaxin 4 and newly synthesized TNFα delivery to the cell surface. To capture and stain TNFα at the surface, its proteolytic release was prevented by using a TACE inhibitor, and the external aspect of the cell surface was imaged on intact macrophages shortly after LPS activation. Three-dimensional composite confocal images were enhanced by surface rendering to remove background staining and reveal TNFα surface staining (Fig. 4B). Under these conditions, the earliest appearance of newly synthesized TNFα was observed in discrete patches on the macrophage surface. TNFα patches (Fig. 4B, red labeling) were notably concentrated on filopodia where they colocalized with the syntaxin 4 puncta (Fig. 4B). At later times (1-2 h), TNFα staining becomes more widely spread across the cell surface (data not shown and Fig. 3B), either because of additional delivery sites being recruited or because of retention and spreading of uncleaved TNFα. Thus, these syntaxin 4 patches correspond to the cholesterol-dependent syntaxin 4 puncta labeled on filopodia in ripped open cells and can now be pinpointed as the initial delivery sites for TNFα on the macrophage surface. Interestingly, TACE, the enzyme responsible for cleaving TNFα at the cell surface (6), is totally excluded from lipid raft fractions (36) (Fig. 4A), suggesting that surface delivery and TNFα cleavage occur in different membrane domains.


TNFα is delivered to phagocytic cups in a cholesterol-dependent manner. A, IFNγ-primed macrophages incubated with C. albicans in the presence of TACE inhibitor for 15 min were stained for TNFα and F-actin. Uncleaved TNFα concentrates at the phagocytic cup. Bar,10 μm. B, IFNγ-primed macrophages treated with MβCD for 30 min were incubated with C. albicans in the presence of TACE inhibitor and stained for TNFα and F-actin. TNFα no longer localizes to the phagocytic cup, although it is present at the level of the Golgi complex. C, IFNγ-primed macrophages incubated with C. albicans for 20 min were stained for TACE and F-actin. TACE localizes to the phagocytic cup. D, IFNγ-primed macrophages incubated with C. albicans for 40 min were stained with filipin III revealing the presence of cholesterol-rich membrane domains in the phagocytic cups.

Cholesterol Dependence of SNARE Accumulation and TNFα Delivery to the Phagocytic Cup—During the phagocytosis of C. albicans, TNFα is delivered to the actin-rich phagocytic cup (7) (Fig. 5A) where TACE also accumulates (7) (Fig. 5C). Disruption of cholesterol-dependent lipid rafts inhibited TNFα surface delivery to the phagocytic cup without affecting TNFα synthesis (Fig. 5B). Macrophages were stained for cholesterol using filipin, which revealed the presence of cholesterolrich membrane domains in the phagocytic cups (Fig. 5D). Syntaxin 4 also concentrates in the cups (37) (Fig. 6A); however, depletion of cholesterol ablated syntaxin 4 staining at the phagocytic cup (Fig. 6A). Likewise, staining of the cognate R-SNARE for TNFα delivery, VAMP3, which is normally enriched at the phagocytic cup (7, 38) (Fig. 6B), is abolished by pretreatment of macrophages with MβCD (Fig. 6B). Thus, cholesterol-rich lipid rafts in the phagocytic cup are required for the concentration of SNARE complexes necessary for TNFα delivery at this site.


Cholesterol-dependent localization of SNARE proteins at the phagocytic cup. A, IFNγ-primed macrophages incubated in the presence or absence of MβCD for 30 min prior to incubation with C. albicans were stained for syntaxin 4 and F-actin. Syntaxin 4 localization to the actin-rich phagocytic cup is dependent upon cholesterol. Bar, 10 μm. B, IFNγ-primed macrophages were treated as in A and stained for VAMP3 and F-actin. VAMP3 localization to the actin-rich phagocytic cup is dependent upon cholesterol. C, IFNγ-primed macrophages incubated in the presence or absence of MβCD prior to incubation with C. albicans were stained for F-actin. Macrophages were examined under the microscope for the presence of internalized yeast. Bar chart, with the means ± S.E., shows the percentage of macrophages containing yeast. A minimum of 500 cells was counted per condition.

The cholesterol-rich lipid rafts at the site of phagocytic cup formation are also required for efficient phagocytosis of yeast. Treatment of cells with MβCD reduced the number of macrophages with internalized yeast from 60 to 11% (Fig. 6C), confirming the importance of cholesteroldependent lipid rafts in phagocytosis. This reinforces the dual function of recycling endosome membrane delivery to the phagocytic cup for both cytokine secretion and phagocytosis. Clustering of SNAREs in cholesterol-rich lipid rafts at these sites permits the delivery of this membrane for these innate immune functions.


We show here that upon macrophage activation, syntaxin 4, a Q-SNARE responsible for the surface delivery of TNFα, is recruited to discrete cholesterol-dependent lipid rafts on the plasma membrane, predominantly on filopodia, which then become the site of delivery for TNFα. TACE, the enzyme responsible for cleavage of TNFα at the cell surface, is excluded from these raft domains suggesting that TNFα is translocated to other membrane subdomains prior to its cleavage. During the initial stages of phagocytosis, syntaxin 4 recruitment to lipid rafts is concentrated at the phagocytic cup and is crucial for the delivery of the excess membrane required to engulf a microorganism in addition to the rapid delivery and secretion of TNFα.

It has been suggested that spatial distribution of SNAREs in the plasma membrane may play a prominent role in regulating exocytosis. In other cell types (35, 39-42) a number of SNARE proteins are located in cholesterol-dependent lipid raft domains. Previous studies in macrophages have shown that syntaxin 4 and SNAP-23 are also associated with lipid rafts in these cells (21, 22). SNAREs display different levels of raft association depending on the cell type (23); however, to date no other cell type has shown the stimulus coupled relocation of SNARE proteins into lipid rafts that we demonstrate occurs with syntaxin 4 in LPS-activated macrophages. The precise role of lipid rafts in SNARE-mediated fusion is currently unclear. The lipid rafts may act as sites for transmembrane pairing and membrane fusion, or they may function in a regulatory fashion by spatially separating the Q-SNARE partners in the membrane until they are required for fusion. Consistent with lipid rafts as fusion sites in PC12 cells (39, 41), neutrophils (43), synaptosomes (44), and now macrophages, disruption of lipid rafts leads to a decrease in secretion. In contrast, the decreased lipid raft association of SNAP-23/25 increased the secretion of recombinant growth hormone from PC12 cells (45) suggesting a more regulatory role for lipid rafts in this case. There are of course many other examples of molecules being recruited to lipid rafts in immune cells. The relocation of receptors for antibody complex or antibody in B cells, T cells, mast cells, and basophils to lipid rafts coordinates signaling (46).

The Sec/Munc18 family of SNARE-binding proteins regulates SNARE complex formation, and because in macrophages Munc18c is completely excluded from raft domains, it may function to restrict the movement of syntaxin 4 into lipid rafts for SNARE complex formation (47, 48). Similar results are seen in mast cells where syntaxin 3/Munc18b complexes were found in nonraft fractions, whereas syntaxin 3-containing SNARE complexes were found within lipid rafts domains (49). Lipid rafts may also orchestrate the fate and biological activities of TNFα. TNFα can be retained on cells as an active 26-kDa transmembrane protein at the cell surface, otherwise it is cleaved by the enzyme TACE and released as a soluble 17-kDa cytokine for other roles in immunity (50-52). We found TACE in nonraft fractions regardless of the activation state of the macrophages, implying that TNFα must first exit the lipid raft domains in order to be cleaved and released from the membrane, although this remains to be formally shown. Retaining TNFα in lipid rafts could potentially preserve the 26-kDa transmembrane form at the surface, and the release from lipid rafts would allow cleavage and release of the 17-kDa form, thus dictating the physiological functions of TNFα.

Upon activation, macrophages assume a highly ruffled cell surface with the extension of exaggerated filopodia (53) rich in cholesterol and lipid rafts (54). We now show syntaxin 4 in resting macrophages localizes to discrete puncta mainly on the cell body similar to the syntaxin 1 cholesterol-dependent puncta found in neuroendocrine cells (41) and the punctate staining of SNAP-23 on adipocytes (35). Upon activation, additional cholesterol-dependent syntaxin 4 puncta appear on the surface, particularly on filopodia forming the sites for the initial delivery of TNFα. Larger cholesterol-dependent clusters also emerge on the surface of macrophages after LPS activation, the significance of which is not yet clear. Individual lipid raft microdomains are typically below 50 nm diameter (55, 56), and TNFα delivery occurs at the many smaller syntaxin 4-labeled clusters; nevertheless, raft clustering has been demonstrated at sites of T cell receptor stimulation during T cell activation (57).

The formation of preferential fusion sites on filopodia is important in the context of macrophages where excess membrane is required for the formation of the nascent phagocytic cup (58, 59). We have recently shown the delivery of recycling endosome membrane containing TNFα to the nascent phagocytic cup serves to provide extra membranes to engulf a microbe while simultaneously delivering TNFα to surface delivery for rapid secretion (7). The relocation of syntaxin 4 to lipid rafts in the phagocytic cup ensures the recycling endosome and its cargo TNFα are delivered to these specific sites. Efficient phagocytosis of particles, such as mycobacteria, opsonized red blood cells, and now yeast, requires cholesterol (20, 60). Our data suggest this reliance on cholesterol is because of the requirement of SNAREs to be associated with lipid rafts for delivery of extra membrane to the phagocytic cup.

The family of serum cholesterol-reducing drugs termed statins (3-hydroxy-3-methylglutaryl-coenzyme A reductase inhibitors) has been reported to have anti-inflammatory properties, such as reducing inducible nitric-oxide synthase and proinflammatory cytokines, including TNFα, in macrophages treated with lovastatin (61-64). However, other reports show that statins can increase the pro-inflammatory response; for example, simvastatin can increase the promoter activity and production of IL-12p40 and TNFα in macrophages (65-67). How the requirement of cholesterol for the delivery to the cell surface of cytokines is involved in the statin influence on inflammatory response has yet to be determined.

In conclusion, the data presented here show that the delivery of TNFα to the plasma membrane in LPS-activated and phagocytosing macrophages is dependent upon the movement of syntaxin 4 into cholesterol-dependent lipid rafts to be fusion-competent. Our findings are in keeping with the participation of lipid rafts in forming surface delivery and exit sites for cytokine secretion.


We thank I. Morrow and S. Martin for expert advice and J. Venturato, T. Khromykh, and D. Brown for technical assistance. Our thanks to all members of the Stow laboratory for useful discussions.


  • 2 The abbreviations used are: TNFα, tumor necrosis factor α; LPS, lipopolysaccharide; MβCD, methyl-β-cyclodextrin; SNARE, soluble N-ethylmaleimide-sensitive factor attachment protein receptor; SNAP-23 or SNAP-25, synaptosomal associated protein of 23 or 25 kDa; VAMP, vesicle-associated membrane protein; IFNγ, interferon γ; TACE, tumor necrosis factor α-converting enzyme.

  • * This work was supported by a grant from the National Institutes of Health and by a fellowship and a grant (to J. L. S.) from the National Health and Medical Research Council of Australia. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

  • Received January 27, 2006.
  • Revision received March 2, 2006.



  1. 17.
  2. 18.
  3. 19.
  4. 20.
  5. 21.
  6. 22.
  7. 23.
  8. 24.
  9. 25.
  10. 26.
  11. 27.
  12. 28.
  13. 29.
  14. 30.
  15. 31.
  16. 32.
  17. 33.
  18. 34.
  19. 35.
  20. 36.
  21. 37.
  22. 38.
  23. 39.
  24. 40.
  25. 41.
  26. 42.
  27. 43.
  28. 44.
  29. 45.
  30. 46.
  31. 47.
  32. 48.
  33. 49.
  34. 50.
  35. 51.
  36. 52.
  37. 53.
  38. 54.
  39. 55.
  40. 56.
  41. 57.
  42. 58.
  43. 59.
  44. 60.
  45. 61.
  46. 62.
  47. 63.
  48. 64.
  49. 65.
  50. 66.
  51. 67.
  • Advertisement