Internalization of immunoglobulin G (IgG) opsonized particles by Fcγ receptors (FcγRs) on the surface of macrophages is the most widely used model of phagocytosis. FcγRs bind to bivalent or multivalent ligands to elicit downstream signaling (13). Recognition and binding of complementary IgG on the surface of a particle results in receptor clustering, which brings the cytosolic domain of each FcγR into close apposition and triggers signaling cascades. The first signaling event to occur after receptor-ligand binding is phosphorylation of tyrosine residues within conserved immunoreceptor tyrosine-based activation motifs (ITAMs) on the cytosolic tail of the FcγRs, mediated by Src family kinases (SFKs) ( Figure 3 ). Inhibition of SFKs or overexpression of their negative regulator, C-terminal Src kinase (Csk), eliminates particle engulfment, but not binding (19–21). Conversely, the heterologous expression of FcγRs in many non-phagocytic cells is able to elicit SFK signaling cascade and the engulfment of IgG-opsonized particles highlighting the robustness and this system.

Many immunoreceptors including FcγRs are found to reside in DRMs that also contain SFKs (22). The amino acid constituents of transmembrane domains and the S-acylation of juxtamembrane regions can partition proteins into liquid ordered regions of the plasma membrane (23). The current evidence suggests that FcγRIIa signaling is only partially dependent on S-acylation of Cys208; its absence prevents the intracellular mobilization of calcium without directly impairing the phosphorylation of the ITAMs (24). Additionally, mutations in the transmembrane domain, which ablate its localization to DRMs, prevent the activation of NF-kB downstream of FcγRIIa activation (25). Together these studies demonstrate that both the S-acylation and the composition of the transmembrane domain support proper receptor function.

It is possible that multiple activated FcγRs and SFKs associate in an ordered domain to support maximal signal output, and that this depends on the membrane lipids, the transmembrane domain of the receptor, and the recruitment of SFKs. We suspect that under suboptimal conditions the size of this domain could be restricted thereby limiting the overall strength and duration of signals. Similar observations have been seen with another phagocytic receptor, the scavenger receptor CD36 (26). In resting cells, CD36 resides in small clusters associated with the SFK Fyn. Upon activation the clusters increase in size recruiting more Fyn, leading to stronger signaling. Since CD36 is S-acylated by zDHHC4 and zDHHC5 (27), and Fyn is dually acylated with myristate and palmitate, it is possible that these modifications assist in lateral segregation and cluster density. Whether this mechanism explains the observations of FcγRIIa remains to be directly tested.

The role of these initial signaling events is to drive particle internalization through extensive remodeling of the actin cytoskeleton. In FcγR mediated phagocytosis, the small Rho-family GTPases Cdc42 and Rac, key regulators of actin dynamics, are activated and recruited to the site of phagocytosis (28, 29). GTPases function as molecular switches alternating between an active (GTP-bound) state and an inactive (GDP-bound) state (30), through the action of guanine nucleotide exchange factors (GEFs) and GTPase-activating proteins (GAPs), respectively. GEFs stimulate GDP dissociation to allow its replacement by GTP, while GAPs are essential to prompt GTP hydrolysis. The eventual outcome of cytoskeletal reorganization is the formation of pseudopod extensions and phagocytic cups that engulf the particle. Soon after sealing the phagocytic cup, the newly formed phagosome participates in successive fission and fusion interactions with endosomes and lysosomes to mature into a highly degradative phagolysosome.

Rac1 stimulates actin polymerization to promote the internalization of attached particles and microorganisms during phagocytosis (31). In its active GTP-bound state, Rac1 is targeted to membranes by prenylation (geranylgeranylation) and a C-terminal polybasic region (KKRKRK) (32, 33). Metabolic labeling in fibroblasts with tritiated palmitate revealed that Rac1 is modified on Cys178 (34, 35). This modification requires Rac1 to already be localized to the membrane as both prenylation and an intact C-terminal polybasic region was required to support S-acylation. Preventing the S-acylation of Rac1 by treatment with 2-bromopalmitate or mutation of the Cys178 residue results in a reduction in both plasma membrane localization and GTP loading of Rac1, which leads to a reduced activation of p21 activated kinase. Ultimately, cells with impaired Rac1 acylation display impaired spreading and defects in migration suggesting defects in the ability to polymerize actin. Whether the S-acylation of Rac1 occurs in macrophage or is important for pseudopod extension during phagocytosis is currently unknown. The identity of the of the zDHHC isoform(s) modifying Rac1 also needs to be determined.

Video-microscopy of phagocytosing macrophages has revealed that these cells have the ability to consume large volumes of particles without a noticeable diminution of their surface area or alteration in shape. Indeed, through the use of fluorescent spectroscopic and electrophysiological techniques (e.g., membrane capacitance) it was determined that the surface area of the cell actually increases immediately following phagocytosis (36, 37). Endocytic vesicles serve as the primary donor of this “extra” membrane used to support particle engulfment (38). This localized delivery at the site of phagocytosis is referred to as “focal exocytosis” (39). Fusion of endocytic vesicles with the base of the phagocytic cup is mediated by the soluble N-ethylmaleimide–sensitive factor attachment protein (SNAP) receptors (SNAREs). SNAREs are classified as Qa‐, Qb‐, Qc‐, Qbc‐ (which contain two SNARE motifs) or R‐SNAREs (40, 41). Specific Q‐SNAREs in the target membrane bind, via their SNARE motifs, to a partner R‐SNARE on vesicles to form a trans‐SNARE complex (42). This brings the two membranes into close proximity allowing fusion and delivery of cargo. On the plasma membrane and base of the phagocytic cup SNAP-23 (a Qbc) is in complex with a Qa SNARE such as Syntaxin2, 3 or 4. Unlike the Qa SNAREs, SNAP-23 does not have a transmembrane and instead associates with cellular membranes through acylation of a cysteine-rich domain present in the linker region connecting its Qb and Qc motifs (43, 44). SNAP-23 has five acylated cysteines and is found to partition into DRMs (45) which may also explain its preference for the plasma membrane and endosomes over the endoplasmic reticulum. Silencing of SNAP-23 in J774 macrophage cells results in impaired phagocytosis and a reduction in the delivery of NADPH oxidase 2 complex required to generate superoxide at the site of phagocytosis/nascent phagosome (46).

During phagocytosis recycling endosomes, early endosomes and late endosomes/phagosomes can all fuse at the base of the phagocytic cup. The R-SNAREs vesicle-associated membrane protein (VAMP) 3 (recycling endosomes) and VAMP7 (late endosomes/lysosomes) have been shown to be important for membrane fusion. VAMP3 has been identified in proteomics screens as an S-acylated protein in macrophages (16, 17). VAMP3 contains a sole cysteine residue, Cys76, located in the cytosol and adjacent to its transmembrane domain. To our knowledge direct examination of this acylation and any potential role has not been examined. Proteomic studies have also identified VAMP7 as an S-acylated protein in macrophages (17). Recent studies using Jurkat T cells have demonstrated that Cys183 can be acylated by zDHHC18 and zDHHC20 and that the fraction of VAMP7 that is acylated increases following T cell receptor stimulation (47). In T cells wild-type VAMP7 has a perinuclear localization which partly overlaps with the Golgi marker Giantin and possibly recycling endosomes. Conversely, the Cys183Ala mutant displays a more diffuse cytoplasmic signal demonstrating that S-acylation is critical for the proper subcellular distribution. To date the importance of S-acylation in macrophage or to phagocytosis has not been examined. However, given the similarity between T cell receptor and FcγR signaling and the role of VAMP7 in phagocytosis it is tempting to speculate that this modification will also support the process.

In addition to supporting focal exocytosis SNAP23 is also critical for mediating membrane fusion events during phagosome maturation. The activity of SNAP23 is necessary for the for the recruitment of several proteins essential for phagosome maturation including v-ATPase (46), MHCI (55) and NOX2 (46, 56, 57) from recycling endosomes and lysosomes. Focal exocytosis of recycling and late endosomes during particle engulfment results in the delivery of a variety of membrane proteins found on these compartments including Syntaxin7 and 13 (also referred to as Syntaxin12) (58). This delivery of endocytic Q-SNAREs to the nascent phagosome allows for further fusions events with recycling and late endosomes/lysosomes containing the appropriate R-SNAREs.

In RAW macrophage, Syntaxin13 colocalizes with the transferrin receptor on recycling endosomes (58). Expression of the cytosolic domain of Syntaxin13, which is known to act as a dominant negative mutant, did not have any impact of the uptake of particles. However, this dominant negative construct impaired the ability of the phagosome to mature to a late phagosome/phagolysosome stage. Syntaxin13 is not reported to be S-acylated and it does not contain any juxtamembrane cysteine residues. However, its cognate partner Syntaxin4 has been found to be S-acylated (59). The presence of Syntaxin7 and 8 on maturing phagosomes supports the fusion with late endosomes and lysosomes via interactions with their cognate pattern VAMP7. Syntaxin7 and 8 are modified with palmitate on Cys239 and Cys214, respectively, adjacent to their transmembrane domains (60). In HeLa cells the Syntaxin 8 is partly colocalized with both CD63 and LAMP1 consistent with a late endosome/lysosome distribution. The Cys214Ala mutant that lacks S-acylation has the same subcellular localization. In contrast in HeLa cells Syntaxin7 localizes primarily with the early endosomal marker EEA1 while the Cys239Ala mutant is retained on the plasma membrane of HeLa. This suggests that S-acylation is needed for partitioning into an endocytic carrier. Whether this mode of regulation is important in macrophage is yet to be examined.

Fusion with late endosomes, lysosomes and carriers derived from the TGN lead to the accumulation of v-ATPases in the membrane of the late phagosome that results in the acidification of the phagosomal lumen. Fusion of the maturing phagosome with these compartments also delivers a variety of hydrolytic enzymes to the lumen of the phagolysosome to facilitate destruction of the internalized prey. The v-ATPase is a large multimeric protein complex consisting of two functional subcomplexes, V₁ and V₀, that together control the import of protons into the lumen of endosomes and phagosomes (61). The cytosolic V₁ subcomplex mediates ATP hydrolysis, while the V₀ subcomplex, integral to the membrane, constitutes the pore through which protons are pumped at the expense of ATP (62). The phagosomal acquisition of the membrane embedded V₀ subcomplex requires membrane trafficking either from pre-existing active complex (i.e., fusion with lysosomes) or the delivery of newly synthesized V₀ from the ER via the Golgi ( Figure 4 ).

Batten disease is a group of rare, fatal, genetically inherited disorders that impact the nervous system which all manifest as neuronal ceroid lipofuscinoses (63). Infantile Batten disease, or ceroid lipofuscinosis, neuronal, 1 (CLN1) is a severe form of the disease caused by mutations in the thioesterase PPT1 (64). PPT1 is a soluble protein thioesterase enzyme that resides in the lumen of lysosomes involved in the degradation of lipid-modified proteins by removing thioester-linked fatty acyl groups from cysteine residues (65). Delivery of recombinant PPT1 via fluid phase endocytosis is able to reverse the accumulation of proteinaceous deposits in the lysosomes lacking PPT1 function. Paradoxically, recent evidence suggests that the activity of the v-ATPase also requires PPT1 to directly remove an acyl chain from the v-ATPase in the cytosol.

Neurons, macrophage and microglia express the V₀a1 isoform as part of the membrane embedded V₀ complex of the v-ATPase. A recent study has demonstrated that the V₀a1 containing Vo complex requires S-acylation for proper trafficking from the Golgi to lysosomes (66). This feature is unique to the a1 isoform compared to the other tissue-selective isoforms (a2-a4). Cys25 of the a1 subunit was shown to be S-acylated by acyl-biotin exchange and that this modification is important for the recognition of the V₀ complex by the adaptor complex 3 in the trans-Golgi and its subsequent delivery to lysosomes. Curiously, despite its role in trafficking, the modification itself impairs maximal proton translocation activity (66). Mutations in PPT1 result in a significantly higher lysosomal pH in neurons compared to those from littermate controls. It is currently unclear how lysosomal PPT1 can remove fatty acids attached to a cytosolically modified Cys residue. One possibility is that a small fraction of the PPT1 also resides in the cytosol (66). Whether this is through inefficient insertion into the ER during biosynthesis or escape via some sort of transient “leak” in limiting membrane of the lysosome is unclear. Since some bacterial pathogens and inert particles can rupture phagolysosomes it would be worth investigating to determine if the release of PPT1 into the cytosol impacts the S-acylome of macrophages.

Phosphatidylinositol 4-kinase IIa (PI4KIIα) resides primarily in the trans-Golgi network (TGN) and endosomes/lysosomes and plays essential roles in membrane transport including: cargo trafficking between the TGN and endosomes via direct interactions with adaptor protein complexes, AP-1 and AP-3 (67–69). Importantly both the activity and the presence of PI4KIIα supports the formation of vesicular carriers (68). PI4KIIα is a peripheral membrane protein that is S-acylated within its catalytic core (70). All four cysteine residues within its CCPCC motif are capable of being S-acylated by zDHHC3 and/or zDHHC7 (71). However, mutational analysis indicates that the first two are the preferred sites of this modification (70).

Recently it was also demonstrated that following the acquisition of Rab7, the maturing late phagosome, acquires PI4KIIα by fusion with lysosomes or vesicles from the TGN. This leads to the appearance of PI4P on this late stage phagosome and is required for the maximal acidification of the now phagolysosome (72). Additionally, in dendritic cells PI4K2a and PI4P have been shown to be critical for the localization and activity of toll-like receptor 4 (73). Following the digestion of the prey within the phagolysosome, the organelle is eliminated by a tubulovesicular dissolution process, which concomitantly regenerates lysosomes. The precise mechanisms of this phagolysosomal dissolution or resolution are unclear, however, it is known that if the phagolysosome fails to acquire PI4P its dissolution in inhibited (74). Moreover, similar PI4P/PI4K2a positive tubules in dendritic cells may account for the requirement for PI4K2a to support optimal presentation of antigens by the major histocompatibility complex II (73). Thus, we would predict that loss of both zDHHC3 and zDHHC7 would result in an inability of phagolysosomes to degrade prey and be broken down into lysosomes.

In non-myeloid cells, Rab7 is S-acylated on Cys83 and Cys84 and this regulates the ability of Rab7 to interact with and recruit the retromer complex to late endosomes (75). Prevention of Rab7 acylation results in the aberrant secretion of lysosomal pro-enzyme to the extracellular medium. Although the S-acylation of these two residues has now been demonstrated, it is still not fully understood how this regulates the retromer. While the role of Rab7 acylation has not been investigated in phagosome maturation, it is tempting to speculate that it will also be required for retrograde transport of cargo receptors such as the mannose 6-phosphate receptor or sortillin back to the Golgi. In a separate investigation identified zDHHC15 as a protein required for optimal retrograde lysosome to Golgi transport of sortillin and the cation-independent M6PR (76). This study also demonstrated that acylation was required for these two cargo proteins to associate with the retromer complex. This raises the possibility that zDHHC15 may directly modify Rab7 to mediate retrograde transport from lysosomes/phagosomes to the trans-Golgi Network.