Phagocytosis: How the PhagosomeBecame the Phag-ER-some

Dispatch

Colin Watts

Phagocytosis is a fundamental mechanism forengulfment of pathogens, dead and dying host cellsand other particulate material. Surprisingly, theinvolvement of the endoplasmic reticulum inmacrophage-mediated phagocytosis has only justbeen discovered.

It was the Russian embryologist Elie Metchnikoff whofirst observed the phenomenon of phagocytosis. Hisinitial observations, made in 1882 on starfish larvaeand water fleas, led him to perform a series of pio-neering studies in cellular immunology on the battlebetween higher animals and invading pathogenicmicroorganisms. Today most biology textbookscontain images of phagocytic cells engulfing whatoften seem to be impossibly large and/or impossiblynumerous foreign particles such as microorganisms,red blood cells or synthetic beads. How can the cellafford to lose the large amounts of surface membraneneeded to package all this material? Fresh insight intothis process has come from a paper recently pub-lished in Cell. Gagnon et al. [1] show that whenmacrophages phagocytose various objects most ofthe enveloping membrane derives not from theplasma membrane but remarkably from the endoplas-mic reticulum (ER).

Earlier studies provide valuable but apparentlyincomplete information on the membrane dynamics ofphagocytosis. Fusion of newly formed phagosomeswith pre-existing elements of the endosome/lysosomesystem allows recycling of membrane to the cellsurface, the acquisition of hydrolytic enzymes and thecapacity to acidify as these organelles mature tobecome degradative phagolysosomes [2–4]. Recentwork demonstrating focused exocytosis of membraneat the site of phagosome formation also offers apartial solution to the ‘membrane economy’ problemassociated with phagocytosis [5]. Recently, thechanges that take place during phagosome matura-tion have been catalogued by proteomic analysis ofhighly purified phagosomes isolated at differentstages of their life history. The emerging picture is ofa highly dynamic organelle whose compositionchanges with age.

The most recent and comprehensive proteomicanalysis of purified phagosomes revealed some ratherunexpected constituents. In a study published lastyear, Garin and colleagues [6] demonstrated that iso-lated phagosomes loaded with latex beads contained

several proteins normally found in the ER. Severalcontrols suggested that contaminating ER per se wasnot likely to be the source of these proteins. Instead,it was suggested that phagosomes had fused withelements of the ER, providing another mechanism forthe cell to control its membrane economy. Thusplasma membrane spent ingesting large phagocyticmeals could be paid back following replacement withendoplasmic reticulum, an abundant reservoir ofmembrane but not one known previously to communi-cate directly with organelles on the endocytic/phago-cytic pathway.

In a follow up to this work, Gagnon et al. [1] useimmunoelectron microscopy to show directly that ERcomponents are integral to the phagosome mem-brane, removing any lingering doubts that the ERcomponents are contaminants. Remarkably, they alsoshow that ER proteins are introduced at the very ear-liest stage of phagosome formation in macrophages.Striking images reveal the fusion of ER tubules at thesite of the forming phagosome (Figure 1). This tran-sient intermediate was best visualised when phos-phatidylinositol 3-kinase, known to be required forcompletion of phagocytosis [7], was inhibited. Gagnonet al. [1] describe the process as one whereby the par-ticle ‘slides’ into the lumen of the ER. It is a phago-some predominantly made up of ER that then beginsa programme of maturation.

This strikingly different view of phagocytosis inmacrophages will stimulate a reassessment of manyaspects of the cell biology of both the ER and phago-somes. For example, several pathogenic microorgan-isms find a safe niche within phagocytic vacuoles andare able to control phagosome maturation to theiradvantage (reviewed in [8]). These new data canexplain how pathogens such as Brucella survive in anessentially ER-like vacuole [8]. Presumably Brucella isable to impose an embargo on communication withother organelles, while other pathogens may allowlimited interaction with endosomes but not enough toendanger their survival. The results of Gagnon et al. [1]may also explain the appearance of ER proteins suchas calreticulin on the cell surface where it apparentlyworks in conjunction with CD91 to bind and internaliseapoptotic cells [9]. However, in this case the mecha-nism of uptake appears to involve a variant form ofphagocytosis whereby the particle stimulates mem-brane ruffling and the formation of a vacuole consid-erably larger than necessary to accommodate theparticle. It will be interesting to confirm whether this‘spacious phagocytosis’, which is more akin tomacropinocytosis [10], also involves recruitment of ERmembrane.

Gagnon et al. [1] also highlight the possible role thatER phagocytosis might play in permitting antigen-pre-senting cells to display class I major histocompatibil-ity complex (MHC) associated peptides frompathogenic organisms. Class I MHC molecules in the

Current Biology, Vol. 12, R666–R668, October 1, 2002, ©2002 Elsevier Science Ltd. All rights reserved. PII S0960-9822(02)01163-6

Division of Cell Biology & Immunology, Wellcome TrustBiocentre, School of Life Sciences, University of Dundee,MSI/WTB Complex, Dow Street, Dundee, DD1 5EH, UK.Email: c.watts@dundee.ac.uk

ER membrane usually capture peptides delivered intothe lumen of the ER by the TAP transporter complexfollowing generation in the cytosol by the proteasome.Transport of the peptide–MHC complex to the cellsurface alerts the immune system to the presence of‘endogenous’ pathogens, especially viruses. However,considerable evidence has accumulated that peptidesfrom so-called ‘exogenous’ antigens, including thosephagocytosed by antigen-presenting cells, can alsobe presented on class I MHC molecules (reviewed in[11]). Does ER-mediated phagocytosis help us tounderstand the mechanisms behind this unconven-tional but important mode of class I MHC loading? Itis worth remembering that proteasome and TAP func-tion is required for the processing of some T-cell epi-topes generated from exogenous antigens but notothers [11].

Although ER-mediated phagocytosis would bringthe ‘antigen’ into direct proximity with nascent class IMHC molecules, it is not clear how the acidic prote-olytic environment needed for peptide generation canco-exist with the complex multi-chaperoned environ-ment required for folding nascent class I MHC mole-cules. Indeed, Gagnon et al. [1] show that most ERproteins are degraded within 30 minutes of phago-some formation. A further problem is that many T-cellepitopes from exogenous antigens appear to requireproteasomal, i.e. cytosolic, processing before theycan be loaded onto class I MHC molecules.

Simply wrapping up a phagocytosed antigen in ERinstead of plasma membrane does not seem to fullyresolve the problem of how exogenous antigens areprocessed and loaded onto class I MHC moleculesbut a closer look at the data from Gagnon et al. [1]suggests other possibilities. Although the initial inputof ER proteins appeared to be rapidly degraded, ERmarkers were re-acquired by phagosomes later onsuggesting that repeated contacts with the ER aremade during phagosome maturation [1]. In that casepeptides generated in phagosomes may be able toaccess the ER proper and, perhaps following trimming[12,13], bind to class I MHC molecules located there

(route 1 in Figure 1). This mode of class I MHC loadingwould be proteasome and TAP independent. But whatabout proteasome- and TAP-dependent loading? Oneof the problems here has been to understand howexogenous material enters the cytosol for proteaso-mal processing. ER-targeted proteins that misfold orfail to oligomerise are re-exported for degradation viathe translocon channel [14,15]. Conceivably, exoge-nous antigen material that reaches the ER could alsobe recognised as aberrant by the ER quality controlmachinery and be exported to the cytosol. Followingproteasomal processing, such epitopes could then re-enter the ER via TAP (route 2 in Figure 1). Of course,direct loading of class I MHC molecules within phago-somes may also occur instead of or as well as the pro-posed pathways above. In addition, there is plenty ofevidence that non-phagocytic routes of uptake canlead to presentation of exogenous antigens on class IMHC molecules, suggesting that other pathways mustexist. Sorting out which ones might be exploited, forexample for vaccination purposes, will keep immunol-ogists busy for a while longer.

Two signalling pathways associated with phagocy-tosis have been described — one triggered byimmunoglobulin-opsonised particles and dependenton the GTPases Cdc42 and Rac and the other trig-gered by complement-opsonised particles and depen-dent on Rho [16]. Interestingly, Gagnon et al.’s results[1] indicate that in macrophages both modes ofphagocytosis utilise ER-mediated uptake. Neutrophils,however, apparently use a more conventional plasma-membrane-mediated type of phagocytosis [1]. Futurework will focus on what determines the use of distinctmodes of phagocytosis, which mechanism is used byother phagocytic cell types and what the biologicalconsequences are for the internalised particle findingitself wrapped up in ER versus plasma membrane.

References1. Gagnon, E., Duclos, S., Rondeau, C., Chevet, E., Cameron, P.H.,

Steele-Mortimer, O., Paiement, J., Bergeron, J.J. and Desjardins, M.(2002). Endoplasmic reticulum-mediated phagocytosis is a mecha-nism of entry into macrophages. Cell 110, 119-131.

Current BiologyR667

Figure 1. ER-mediated phagocytosis andthe possibilities for peptide loading onclass I MHC molecules.

According to Gagnon et al. [1], the ER(red) is involved in the process of phago-cytosis at several stages. Initially it medi-ates the phagocytic event but may makerepeated contacts that allow fresh ERcomponents to be incorporated. At thesame time, interaction with the endo-some/lysosome system (blue) delivershydrolytic enzymes (blue dots). The figuresuggests that there is two-way trafficbetween ER and maturing phagosomesand that material from the phagosomecan also reach the ER. This could includedigestion products (grey shapes) suitablefor loading onto class I MHC molecules,perhaps after trimming by ER aminopep-tidases (route 1) or suitable for export viathe translocon, processing by the protea-some and re-import via TAP (route 2). Theplasma membrane is shown in green.

Current Biology Translocon

Endosomesystem

Proteasome

TAP complex

Plasmamembrane

Endoplasmicreticulum

Class I MHC

2

1

DispatchR668

2. Pitt, A., Mayorga, L.S., Stahl, P.D. and Schwartz, A.L. (1992). Alter-ations in the protein composition of maturing phagosomes. J. Clin.Invest. 90, 1978-1983.

3. Beron, W., Alvarez-Dominguiz, C., Mayorga, L. and Stahl, P.D.(1995). Membrane trafficking along the phagocytic pathway. TrendsCell Biol. 5, 105.

4. Desjardins, M., Huber, L.A., Parton, R.G. and Griffiths, G. (1994).Biogenesis of phagolysosomes proceeds through a sequentialseries of interactions with the endocytic apparatus. J. Cell Biol. 124,677-688.

5. Bajno, L., Peng, X.R., Schreiber, A.D., Moore, H.P., Trimble, W.S.and Grinstein, S. (2000). Focal exocytosis of VAMP3-containingvesicles at sites of phagosome formation. J. Cell Biol. 149, 697-706.

6. Garin, J., Diez, R., Kieffer, S., Dermine, J.F., Duclos, S., Gagnon, E.,Sadoul, R., Rondeau, C. and Desjardins, M. (2001). The phagosomeproteome: insight into phagosome functions. J. Cell Biol. 152, 165-180.

7. Araki, N., Johnson, M.T. and Swanson, J.A. (1996). A role for phos-phoinositide 3-kinase in the completion of macropinocytosis andphagocytosis by macrophages. J. Cell Biol. 135, 1249-1260.

8. Meresse, S., Steele-Mortimer, O., Moreno, E., Desjardins, M., Finlay,B. and Gorvel, J.P. (1999). Controlling the maturation of pathogen-containing vacuoles: a matter of life and death. Nat. Cell Biol. 1,E183-E188.

9. Ogden, C.A., deCathelineau, A., Hoffmann, P.R., Bratton, D., Ghe-brehiwet, B., Fadok, V.A. and Henson, P.M. (2001). C1q andmannose binding lectin engagement of cell surface calreticulin andCD91 initiates macropinocytosis and uptake of apoptotic cells. J.Exp. Med. 194, 781-795.

10. Swanson, J.A. and Baer, S.C. (1995). Phagocytosis by zippers andtriggers. Trends Cell Biol. 5, 89.

11. Yewdell, J.W., Norbury, C.C. and Bennink, J.R. (1999). Mechanismsof exogenous antigen presentation by MHC class I molecules invitro and in vivo: implications for generating CD8+ T cell responsesto infectious agents, tumors, transplants and vaccines. Adv.Immunol. 73, 1-77.

12. Serwold, T., Gaw, S. and Shastri, N. (2001). ER aminopeptidasesgenerate a unique pool of peptides for MHC class I molecules. Nat.Immunol. 2, 644-651.

13. Fruci, D., Niedermann, G., Butler, R.H. and van Endert, P.M. (2001).Efficient MHC class I-independent amino-terminal trimming ofepitope precursor peptides in the endoplasmic reticulum. Immunol-ogy 15, 467-476.

14. Wiertz, E. J., D. Tortorella, M. Bogyo, J. Yu, W. Mothes, T. R. Jones,T. A. Rapoport and H. L. Ploegh. (1996). Sec61-mediated transfer ofa membrane protein from the endoplasmic reticulum to the protea-some for destruction Nature 384, 432-438.

15. Johnson, A.E. and Haigh, N.G. (2000). The ER translocon and retro-translocation: is the shift into reverse manual or automatic? Cell102, 709-712.

16. Caron, E. and Hall, A. (1998). Identification of two distinct mecha-nisms of phagocytosis controlled by different Rho GTPases.Science 282, 1717-1721.