The internalization of large phagocytic targets, such as apoptotic cells and fungal organisms, or numerous smaller targets, requires the delivery of internal membranes to the site of phagocytosis (Uribe-Querol and Rosales, 2020). Proteomic studies using latex beads and microorganisms demonstrated that isolated phagosomes contain a variety of ER-resident proteins (Gagnon et al., 2002). Furthermore, this study and a variety of others using electron microscopy could demonstrate that the ER comes into close contact with the PM (Guermonprez et al., 2003; Houde et al., 2003). However, another group revealed that despite the close apposition of the ER to the site of phagocytosis, the two organelles were not directly fusing (Touret et al., 2005) and that endosomes and lysosomes were the primary sources of the “extra” membrane required to support robust phagocytosis. However, given the ever-growing literature on ER contact sites, the earlier observations and proteomic data raise the question of whether ER contact sites are formed with a portion of the PM actively engaged in phagocytosis and if the MCS is important for the process.

Using J774 murine macrophage cells early studies supported the ER–PM fusion model of phagocytosis showed that the ER-localized SNARE proteins, Syntaxin18, and Sec22b promote phagocytosis (Becker et al., 2005; Hatsuzawa et al., 2006). However, no evidence has been provided to demonstrate that the ER can directly fuse with the PM in this or other cellular contexts (Saheki and De Camilli, 2017). One possibility is that Sec22b interacts with Syntaxin 1 or another SNARE, and together these two proteins form an ER-PM tether (Figure 1B). However, it has also been demonstrated that over-expression of Sec22b can act as an inhibitor of phagocytosis, implying that excess or mislocalized non-fusogenic SNARE interactions/tethering can block particle uptake (Hatsuzawa et al., 2009). Many questions remain regarding the role of these non-fusogenic SNARE interactions. Evidence from neurons demonstrates that Sec22b contributes to the expansion of the PM (Petkovic et al., 2014), a process that could conceivably assist with particle engulfment. This study revealed that the yeast Sec22 interacts with yeast oxysterol-binding proteins Osh2 and Osh3 to support lipid transfer and thus the membrane expansion. By analogy, it is conceivable that mammalian Sec22b may facilitate the localized exchange of lipids. In support of this concept, a recent preprint has been released examining the role of the mammalian oxysterol binding protein-related protein 8 (ORP8) and Sec22b in mouse embryonic fibroblasts made phagocytic through the expression of Fcγ2A receptor (Santos et al., 2022). This study demonstrates that Sec22b supports ER-phagosome contact sites, influences the metabolism of phosphoinositides, and phagosome maturation in this non-professional phagocyte. Together, while the exact roles of Sec22b remain unknown, the growing body of literature suggests that it functions as part of an MCS tether.

Both global and local calcium signals influence phagosome formation and maturation, including actin cytoskeleton remodeling and phagolysosome fusion (Nunes and Demaurex, 2010). Phospholipase C activation at the sites of phagocytosis results in the local hydrolysis of phosphatidylinositol 4, 5-bisphosphate (PI4,5P₂), producing diacylglycerol and inositol trisphosphate (IP₃). The IP₃ can then bind its receptor in the ER, causing a release of Ca²⁺ from its lumen. Upon depletion of the ER Ca²⁺ store, the ER-resident protein stromal interaction molecular 1 (STIM1) undergoes a series of conformational changes, allowing it to interact with plasmalemmal resident calcium release-activated calcium channel protein 1 (termed ORAI) (Wu et al., 2006). Light and electron microscopy, coupled with sophisticated genetic and molecular biology techniques, demonstrate that STIM1-mediated ER contacts promote the Ca²⁺ transients recorded during phagosome formation and early maturation (Nunes et al., 2012) (Figure 1B). In addition, the results suggest that binding of STIM1 to ORAI channels facilitates actin filament depolymerization from the base of the phagocytic cup (Garin et al., 2001; Buschow et al., 2012; Campbell-Valois et al., 2012). Importantly, these findings demonstrate that MCS form at the base of the phagocytic cup during the initial stages of phagocytosis. Since STIM1 and ORAI interact in trans at ER-PM MCS, their interaction may have a secondary role of stabilizing contact sites which may allow other proteins to function.

Phosphotyrosine signals drive most, if not all, forms of phagocytosis. There are numerous excellent reviews on the signal transduction pathways at play during phagocytosis, so we will only highlight a few general features. The best-studied phagocytosis pathway involves Fcγ Receptors, which bind to the opsonin immunoglobulin G on particles. Clustering of Fcγ Receptors (FcγR) activates Src family kinases (SFKs) which phosphorylate tyrosine residues in the cytosolic region of the FcγR. The phosphorylated tyrosine residues then serve as a docking site for Syk tyrosine kinase, which further phosphorylates itself and additional adaptors and transducers of the signal. The full complement of tyrosine phosphatases involved in turning off phagocytic signaling is not currently known.

PTP-1B is bound to the ER via a C-terminal transmembrane anchor (Frangioni et al., 1992). PTP-1B is best known for its ability to dephosphorylate SFKs, epidermal growth factor receptor (EGFR), and insulin receptor (IR) (Bakke and Haj, 2015). Using substrate-trapping mutants and fluorescence energy transfer, researchers have demonstrated that PTP-1B functions at ER–endosome contact sites and potentially ER–PM contact sites (Anderie et al., 2007; Eden et al., 2010; Yue et al., 2019). As the ER-phagosome contact site resembles the ER-endosome contact site, the biological significance of PTP-1B could be to downregulate FcγR signaling or other phagocytic receptors. In this regard, failure to efficiently dephosphorylate the phospho-ITAMs of the FcγR could potentially lead to prolonged signaling from the nascent or early phagosomes. Additionally, and by analogy to the EGFR, PTP-1B may impact the relative abundance of FcγRs since their targeted degradation in multivesicular bodies requires timely dephosphorylation. Whether this would enhance or hinder the rates of IgG-mediated phagocytosis in macrophages is unclear. Continued signaling by Src family kinases and Syk would result in the phosphorylation and activation of a variety of effectors and scaffolding proteins (Lowell, 2011), which may be limiting if the proper spatial and temporal coordination is lost. Syk kinase is also a substrate of PTP-1B, but whether this occurs at an ER-phagosome contact site has not been deduced (Figure 1C). Given our knowledge of MCS, PTP-1B, and phagocytosis, it is reasonable to hypothesize that PTP-1B may be dephosphorylating FcγR, SFKs, Syk, or other substrates at an ER–phagosome contact site.

Phagocytic receptors are not the only putative PTP-1B substrates in the phagosome. Través and colleagues previously demonstrated that murine or human macrophage deficient in PTP-1B have greater responses to the polyinosinic:polycytidylic acid (a TLR3 ligand) and lipopolysaccharide (a TLR4 ligand) (Través et al., 2014). This study used only the purified forms of these ligands. However, it is conceivable that TLR4 and other receptors could be locally activated at the site of phagocytosis and that PTP-1B may downregulate this signal during or following internalization. Clearly, the role of PTP-1B in phagosome formation and maturation is worth investigating and may reveal unappreciated aspects of signal transduction in macrophages and potentially other phagocytic cells.

In eukaryotic cells, most lipids are synthesized in the ER, with a few others produced in the mitochondria or Golgi apparatus. However, many lipids reside primarily in other compartments, especially the PM. The cell has two ways to move lipids between organelles; vesicular and non-vesicular transport. For the sake of brevity, we will focus on the non-vesicular transport pathways. Most lipids are too hydrophobic to spontaneously leave one membrane bilayer, transit an aqueous environment, and insert into another bilayer. Even the comparatively short distances at MCS are insufficient for the free diffusion of lipids to be biologically relevant. Instead, eukaryotes have evolved a variety of cytosolic proteins, which can move lipids one-by-one between organelles and others that act mostly as channels or troughs to allow for faster, less selective transfer (Jansen et al., 2011; Toulmay and Prinz, 2011; Egea, 2021). Many of these proteins function at ER-PM contact sites or ER-endosome/lysosome contact sites. Unfortunately, very little work on these proteins exists in the context of the life cycle of the phagosome. However, we anticipate that many proteins will likely have a role in the process.

One lipid transfer protein studied in the context of phagosome maturation is oxysterol binding protein (OSBP) related protein 1 L (ORP1L). ORP1L is unique among the OSBP/ORP family as it contains an N-terminal ankyrin domain that allows it to bind to active Rab7 found on late endosomes, lysosomes, and the phagosome equivalents (Johansson et al., 2005). Like many ORPs, ORP1L also contains an FFAT-motif, which allows it to bind to the ER-resident proteins VAMP-associated protein (VAP) A and B (Loewen et al., 2003; Loewen and Levine, 2005; Rocha et al., 2009) By binding active Rab7 and VAPs, ORP1L mediates ER-lysosome contact sites. One role of ORP1L is to act as a cholesterol sensor and regulator of lysosomal transport (Van Der Kant et al., 2013; Wijdeven et al., 2016; Zhao and Ridgway, 2017). When cholesterol levels in late endosomes/lysosomes are low, ORP1L is in an open conformation, allowing contact with Rab7 and VAP (Van Der Kant et al., 2013). However, when cholesterol accumulates in the endosomal membrane, ORP1L dissociates from ER and disrupts the MCS. During maturation, the lysosomes and phagosomes move centripetally toward the microtubule organizing center (MTOC) (Harrison et al., 2003; Eng et al., 2007). This dynein-mediated process is required for lysosome fusion. Rab7 connects the phagosome to microtubules via its effectors, Rab7-interacting lysosomal protein (RILP) and ORP1L (Harrison et al., 2003; Rocha et al., 2009). RILP recruits the dynein-dynactin complex with the aid of ORP1L to promote minus-end transport along microtubules. This centripetal migration of phagosomes is essential for the phagolysosome maturation (Jordens et al., 2001; Van Der Kant et al., 2013). Additionally, the Rab7-RILP-dynein-dynactin complex facilitated the movement of these compartments to support the maturation process of phagosomes since it brings them into proximity to lysosomes (Harrison et al., 2003; Wijdeven et al., 2016). To what extent does the phagosome movement toward the MTOC require ORP1L? Still, based on the analogy to other organelles, including autophagosomes and lysosomes, we suspect that it is a minimum playing a supportive role. However, it is abundantly clear that active Rab7 and ORP1L are associated with the phagosome at this stage.

ORP1L has been shown to mediate the transport of cholesterol and PI4P at various ER-contact sites with late endosomes, lysosomes, and phagolysosomes. For instance, in non-phagocytic cells, in the absence of low-density lipoprotein (LDL), ORP1L mediates cholesterol transport from the ER, where it is synthesized, to late endosomes (Eden et al., 2016). Alternatively, when cells are supplied with LDL cholesterol, ORP1L can facilitate its transfer in the opposite direction (Zhao et al., 2020). In macrophages, ORP1L localizes at the MCSs between the ER-phagolysosome contact sites, where it mediates the transport of PI4P from phagolysosomes to the ER (Levin-Konigsberg et al., 2019) (Figure 1D). Live imaging revealed that the disappearance of PI4P from phagolysosomes was impaired in ORP1L-deficient macrophages. This study could not find clear evidence of cholesterol counter-transfer. However, it has been reported that cholesterol tends to increase with phagosome maturation (Rai et al., 2016). In this study, the authors noted that late phagosomes have more cholesterol than early phagosomes and that cholesterol facilitated the clustering of flotillin, Rab7, and dynein (Rai et al., 2016). The authors suggest that clustering of dynein may facilitate interactions with microtubules and thus promote movement towards the MTOC. Currently, the relative importance of ORP1L as a cholesterol sensor and regulator of dynein activity and the ability of cholesterol transfer proteins to generate cholesterol-rich nanodomains that also facilitate dynein-mediated transport are unclear. This and other possible functions of cholesterol in the late phagosome and phagolysosome warrant further investigations.

The examination of PI4P and ORP1L in the phagolysosome revealed striking heterogeneity; regions containing ORP1L and in contact with the ER were devoid of PI4P, whereas the areas free of ORP1L and the ER were rich in PI4P (Figure 2). Live cell imaging using lattice light sheet and electron microscopy revealed robust and long tubules emanating from the phagolysosome once the prey had been degraded. The PI4P positive domains on the phagolysosome serve as “hotspots” to recruit SKIP/PLEKHM2, further recruiting Arl8b and kinesin motors that collectively draw tubules away from the phagolysosome. Importantly, in cells devoid of ORP1L, the PI4P remained continuous on the phagolysosomes, which impaired tubule formation and phagolysosome breakdown (Levin-Konigsberg et al., 2019). Inhibiting the formation of ORP1L-mediated contacts also inhibited phagolysosome resorption, suggesting a role for the MCS in the conclusion of phagocytosis (Figure 2). The findings demonstrate that ORP1L-VAP contacts promote resolution in two distinct ways: the generation of PI4P-rich and depleted microdomains spatially enriches tubulation machinery; and by providing a physical tether that immobilizes the phagolysosome to mechanically support tubulation and fission (Zhao and Ridgway, 2017; Levin-Konigsberg and Grinstein, 2020). In addition to repopulating lysosomes consumed during phagosome maturation, the tubules can carry processed antigens and fuse with MHC Class II compartments for subsequent antigen presentation (Mantegazza et al., 2013). Consequently, it is plausible that phagolysosomal membrane resolution is coupled to antigen presentation mechanisms, especially in dendritic cells.