As far as I understand, in complement system C3b gets deposited on pathogen surface but it can also be deposited on host cells. Host cells have some negative regulatory system such as membrane cofactor of proteolysis MCP, decay associating factor DAF etc which either removes C3b or cleave it using factor I to iC3b which can not form a C3 convertase or C5 convertase so no production of membrane attack complex on host cells.
However, phagocytes contain complement receptors CR2, CR3 etc that specifically recognise iC3b and engulfs the cells bearing it. While it is very useful for pathogens bearing iC3b (In those cases where pathogens contain sialic acid residue which attracts factor I and prevents forming MAC on those pathogens) wouldn't it be very dangerous for host cells?
So if there is receptors against iC3b on phagocytes then what is the point of host cells converting C3b into iC3b, they get killed one way or another anyway
There's a relatively new concept of self-associated molecular patterns (SAMPs), which much like DAMPs and PAMPs, help immune cells make a decision in a slurry of signals. There's at least a fair article on SAMPs here, and to go even further, the suggestion that tumor-associated SAMPs can dampen immune clearance of cancer cells.
In my head a few things are going on:
(1) The opsonins are potentially binding to everything. Of course, there are surfaces like ECM that don't have any receptors or patterns that you don't want getting damaged, and there's at least some evidence that complement clearance by factors I and H (paywall) may contribute to ablating that effect.
(2) The actual targets for the immune system are expressing patterns that specialized pattern recognition receptors on your phagocytes and so forth can recognize.
(3) Self cells are expressing receptors and patterns that tell the very same cells there's no threat, either through direct inhibition or degradation of kill signals.
Remember that when self cells are damaged, the expression of DAMPs help macrophages and the like come clear them away, so there's obviously a way for self cells to control that effect.
The early interaction of Leishmania with macrophages and dendritic cells and its influence on the host immune response
The complicated interactions between Leishmania and the host antigen-presenting cells (APCs) have fundamental effects on the final outcome of the disease. Two major APCs, macrophages and dendritic cells (DCs), play critical roles in mediating resistance and susceptibility during Leishmania infection. Macrophages are the primary resident cell for Leishmania: they phagocytose and permit parasite proliferation. However, these cells are also the major effector cells to eliminate infection. The effective clearance of parasites by macrophages depends on activation of appropriate immune response, which is usually initiated by DCs. Here, we review the early interaction of APCs with Leishmania parasites and how these interactions profoundly impact on the ensuing adaptive immune response. We also discuss how the current knowledge will allow further refinement of our understanding of the interplay between Leishmania and its hosts that leads to resistance or susceptibility.
Phagocytosis: A Fundamental Process in Immunity
One hundred years have passed since the death of Élie Metchnikoff (1845–1916). He was the first to observe the uptake of particles by cells and realized the importance of this process for the host response to injury and infection. He also was a strong advocate of the role of phagocytosis in cellular immunity, and with this he gave us the basis for our modern understanding of inflammation and the innate and acquired immune responses. Phagocytosis is an elegant but complex process for the ingestion and elimination of pathogens, but it is also important for the elimination of apoptotic cells and hence fundamental for tissue homeostasis. Phagocytosis can be divided into four main steps: (i) recognition of the target particle, (ii) signaling to activate the internalization machinery, (iii) phagosome formation, and (iv) phagolysosome maturation. In recent years, the use of new tools of molecular biology and microscopy has provided new insights into the cellular mechanisms of phagocytosis. In this review, we present a general view of our current knowledge on phagocytosis. We emphasize novel molecular findings, particularly on phagosome formation and maturation, and discuss aspects that remain incompletely understood.
Élie Metchnikoff (1845–1916) made his original observations in the 1880s while studying invertebrate marine organisms. He found special cells attacking small thorns placed into starfish larvae. Based on these findings, he later moved into immunology and championed the concept of cellular immunity. For his contributions he was awarded the Nobel Prize in 1908 . He shared the prize with Paul Ehrlich, a supporter of humoral immunity. Together they provided the bases for modern immunology.
Phagocytosis is an important process for nutrition in unicellular organisms, while in multicellular organisms it is found in specialized cells called phagocytes. Phagocytosis consists in recognition and ingestion of particles larger than 0.5 μm into a plasma membrane derived vesicle, known as phagosome. Phagocytes can ingest microbial pathogens, but importantly also apoptotic cells. In this way, they contribute to the clearance of billions of cells that are turned over every day. Thus phagocytosis becomes essential not only for microbial elimination, but also for tissue homeostasis. Professional phagocytes  include monocytes, macrophages, neutrophils, dendritic cells, osteoclasts, and eosinophils. These cells are in charge of eliminating microorganisms and of presenting them to cells of the adaptive immune system. In addition, fibroblasts, epithelial cells, and endothelial cells can also perform phagocytosis. These nonprofessional phagocytes cannot ingest microorganisms but are important in eliminating apoptotic bodies [3, 4].
Phagocytes must recognize a large number of different particles that could potentially be ingested, including all sorts of pathogens and also apoptotic cells. This recognition is achieved thanks to a variety of discrete receptors that distinguish the particle as a target and then initiate a signaling cascade that promotes phagocytosis. Receptors on the plasma membrane of phagocytes can be divided into nonopsonic or opsonic receptors. Nonopsonic receptors can recognize directly molecular groups on the surface of the phagocytic targets. Among these receptors there are lectin-like recognition molecules, such as CD169 and CD33 also related C-type lectins, such as Dectin-2, Mincle, or DNGR-1 scavenger receptors  and Dectin-1, which is a receptor for fungal beta-glucan . Other receptors, such as SR-A or CD36, can recognize both apoptotic and microbial polyanionic ligands, but their signaling capacity is not well described . Interestingly, toll-like receptors (TLRs)  are detectors for foreign particles, but they do not function as phagocytic receptors. However, TLRs often collaborate with other nonopsonic receptors to stimulate ingestion .
Opsonic receptors recognize host-derived opsonins that bind to foreign particles and target them for ingestion. Opsonins include antibodies, complement, fibronectin, mannose-binding lectin, and milk fat globulin (lactadherin) . The best characterized and maybe most important opsonic phagocytic receptors are the Fc receptors (FcR) and the complement receptors (CR). FcRs bind to the constant (Fc portion) of immunoglobulin (Ig) G [9, 10] or IgA antibodies . Complement receptors, such as CR3, bind to iC3b deposited on the particle after complement activation .
After recognition of the target particle, phagocytic receptors initiate signaling cascades that remodel lipids in the cell membrane and regulate the actin cytoskeleton in order to extend the cell membrane around the particle . During this part of the process, phagocytic receptors also engage in a sequential order and cooperate to complete the formation of the phagosome .
Once the particle is internalized inside the early phagosome, this vacuole can fuse with vesicles coming from the endoplasmic reticulum and the Golgi complex to form an intermediary phagosome [15–21]. The contribution of the endoplasmic reticulum to phagosome formation and maturation is not completely understood, particularly in relation to cross-presentation of antigens. This is the process by which MHC class I (MHC-I) molecules can also present peptides from extracellular proteins. MHC-I molecules are delivered to the phagosome, where they are loaded with peptide and then recycled back to the plasma membrane. At present, it is not possible to convincingly describe a trafficking pathway for MHC-I molecules leading to cross-presentation. While classic (endogenous) MHC-I loading is basically restricted to the secretory pathway, cross-presentation involves interaction between this pathway and the phagocytic pathway . A complete discussion of cross-presentation is beyond the scope of the present review. The reader is directed to recent excellent reviews on this topic [23, 24]. Similarly, the contribution of the Golgi complex to phagosome formation is a matter of debate. Despite the fact that a role for the Golgi complex during phagocytosis by macrophages has been ruled out consensually by several groups [25–27], it is important to notice that these reports are mainly focused on Fcγ receptor-mediated phagocytosis. In contrast, it was recently reported that recruitment of Golgi-derived secretory vesicles during phagosome formation was important for uptake of most particles, except IgG-opsonized ones . The formation of an intermediary phagosome is dynamic process involving fusion of endocytic vesicles and fission of secretory vesicles, resulting in remodeling of the membrane and progressive acidification of the phagosome . Later this intermediary phagosome turns into a microbicidal vacuole, the phagolysosome, by fusing with lysosomes and changing its membrane and interior characteristics through a process named phagolysosome maturation .
2. Particle Recognition
The first step in phagocytosis is the detection of the particle by phagocytes. This, as mentioned before, is accomplished by specialized receptors on the cell membrane. Foreign particles, such as microbial pathogens, can be recognized directly by receptors that bind molecules not found in higher organisms, or indirectly through opsonins. Several receptor types are found on a single phagocyte and they cooperate for recognition and ingestion of the particle. Some receptors can bind to pathogen-associated molecular patterns (PAMPs) but not necessarily initiate phagocytosis. TLRs and some G-protein coupled receptors prepare (prime) the cell for phagocytosis by inducing inside-out activation of phagocytic integrins.
2.1. Receptors for Foreign Particles
2.1.1. Pattern-Recognition Receptors
Some receptors that directly bind PAMPs and seem to be phagocytic receptors include Dectin-1, mannose receptors, CD14, and scavenger receptor A (SR-A) (Table 1). Dectin-1 binds to polysaccharides of some yeast cells . Mannose receptors bind mannan . CD14 binds to lipopolysaccharide-binding protein . SR-A can detect lipopolysaccharide (LPS) on some gram-negative bacteria  and on Neisseria meningitidis . Among these receptors, Dectin-1 has been clearly shown to be sufficient for activating phagocytosis. When it is expressed on heterologous cells that normally cannot perform phagocytosis, it gives the cells phagocytic capabilities [29, 34]. However, for other PAMP receptors the phagocytic potential is still a matter of debate. It may be that they induce phagocytosis indirectly by tethering the particle to the phagocyte surface, or by priming the phagocyte  to ingest the particle via other receptors.
2.1.2. Opsonic Receptors
Foreign particles can also be recognized by phagocytes through soluble molecules that will bind to the particles, tagging them for ingestion. Once on the surface of the target particle, these molecules, called opsonins, are in turn recognized by specific receptors on the membrane of phagocytes. In this manner, opsonins function as a bridge between the phagocyte and the particle to be ingested. Antibody (IgG) molecules and complement components are important opsonins that induce efficient phagocytosis, and their receptors have been studied extensively (Table 1). Fcγ receptors (FcγR) are a family of glycoproteins expressed on the membrane of leukocytes, capable of binding the Fc portion of IgG molecules [10, 36]. These receptors can bind to the various IgG subclasses with different affinities  and when crosslinked by multivalent antigen-antibody complexes can induce phagocytosis and other cellular responses . Complement receptors (CRs) recognize components of the complement cascade, deposited on the surface of phagocytic targets . There are now three recognized gene superfamilies of complement receptors: (i) the short consensus repeat (SCR) modules that code for CR1 and CR2, (ii) the β2 integrin family members CR3 and CR4, and (iii) the immunoglobulin Ig-superfamily member CRIg . Complement receptors, such as the integrin
β2 (also known as CD11b/CD18, CR3, or Mac-1), bind the complement component iC3b deposited on pathogens to promote phagocytosis [38, 39].
2.2. Receptors for Apoptotic Cells
In addition to foreign pathogens, in a normal organism there are millions of cells that die by apoptosis every day. These apoptotic bodies are constantly cleared by phagocytosis. Recognition of apoptotic bodies involves several signals. First, cells in apoptosis release molecules that normally do not exist outside cells. Some of these molecules include ATP, lysophosphatidylcholine, and sphingosine 1-phosphate. These soluble molecules function as chemoattractants for phagocytes. Also, apoptotic cells are displayed on their surface molecules, such as phosphatidylserine (PS) not normally present on a healthy cell . These surface molecules function as an “eat me” signal  for phagocytes. Some receptors such as TIM-1, TIM-4 , stabilin-2 , and BAI-1 (brain-specific angiogenesis inhibitor 1) directly recognize PS . Other receptors, for example, MFG-E8 (lactadherin), can connect PS to αVβ3 integrins . Apoptotic cells can also be recognized by scavenger receptors A (SR-A), MARCO, and CD36 . CD36 bind modified lipids, including oxidized PS . Many normal cells can also express some amounts of PS on their membranes. However, PS increases as much as 300-fold in apoptotic cells, creating a threshold that prevents phagocytosis of normal cells. There are some cells, for example, activated B and T cells, that may present large amounts of PS on their membrane. To prevent phagocytosis, these cells express molecules that deliver a “do not eat me” signal . CD31 is one such molecule. It prevents phagocytosis by promoting cell detachment after homotypic (self)-binding . Also, CD47 is another molecule that blocks phagocytosis of cells expressing it on their surface. CD47 binds to the receptor SIRPα (signal regulatory protein α), on the membrane of phagocytes, and delivers an inhibitory signal for actin assembly . Another level of complexity is the fact that multiple receptors bind apoptotic cells directly or indirectly and professional phagocytes coexpress many of these receptors. Thus, there are still many unidentified mechanisms for phagocytosis via apoptotic receptors. Because, it is now recognized that clearance of apoptotic cells is fundamental for tissue homeostasis , future research will bring us great surprises in this area.
2.3. Receptor Cooperation
For an efficient recognition of the target particle, multiple receptors on the phagocyte must engage multiple ligands on the particle. This interaction depends on the relative affinity of the molecules involved and also on their density on the surface of both the leukocyte and the particle. In addition, the relative mobility of the receptors on the membrane of the phagocyte affects the avidity of the interaction . Because phagocytic receptors get activated when they aggregate in the plane of the membrane, only receptors capable of fast lateral diffusion are more likely to form multimers and get activated than immobile receptors (see section on phagosome formation). Aggregation (also called crosslinking) of the receptors is additionally promoted by the active nature of phagocytes, which constantly form membranous projections to probe their environments [61, 62]. Thus, particle recognition by receptor binding and activation are very active processes.
Another aspect of receptor cooperation is observed when integrin receptors, such as the CR3, increase their affinity for their ligand only after the phagocyte gets extra stimuli through TLRs , Fc receptors , or CD44 . These receptors initiate intracellular signaling that activates the small GTPase Rap1 , which in turn provokes conformational changes in the integrin, leading to its increased affinity. This process is called inside-out signaling because the signal that activates the integrin comes from inside the cell. During the phagocytic process integrins get activated to promote efficient receptor binding all around the target particle (see later).
3. Particle Internalization
When a particle interacts with phagocyte receptors, a series of signaling events are triggered to activate phagocytosis. Important changes in membrane remodeling and the actin cytoskeleton take place leading to the formation of pseudopods that cover the particle. At the point of contact, a depression of the membrane (the phagocytic cup) is formed. Then, the membrane surrounds the target particle and within few minutes it closes at the distal end, leaving a new phagosome. The signaling cascades are known in great detail for the Fc receptors and the complement receptors, since these are the best-studied phagocytic receptors [38, 67, 68]. Signaling for other phagocytic receptors is just beginning to be explored. Great interest exists in this area and research will certainly be fruitful in the near future.
3.1. Fcγ Receptor Signaling
Fcγ receptors get activated in the plane of the phagocyte membrane when they aggregate after binding to their IgG ligands that cover the particle to be ingested. In humans there are several types of activating FcγRs that are coexpressed by professional phagocytes along with the only inhibitory FcγRIIb. The clustering of activating FcγRs results in the phosphorylation of immunoreceptor tyrosine-based activation (ITAM) motifs present within the cytoplasmic domain of the receptor (as is the case with FcγRIIa and FcγRIIc), or in an associated FcR common γ-chain (as with FcγRI and FcγRIIIa) [9, 10, 69]. ITAM phosphorylation is carried out by Src-family kinases (Lyn, Lck, and Hck specifically), creating a docking site for the SH2 domains of the tyrosine kinase Syk, which can itself phosphorylate neighboring ITAM tyrosines [38, 70]. The mechanism by which receptor aggregation induces phosphorylation of the ITAM tyrosines remains elusive. Aggregation may induce accumulation of the FcγRs in cholesterol-enriched lipid rafts, where Src-family kinases are concentrated. This model is supported by the fact that FcγRIIa becomes associated with detergent-resistant membranes (DRMs) upon activation by aggregation [71, 72] and that depletion of cholesterol with methyl-β-cyclodextrin inhibits FcγRII phosphorylation in response to aggregation . Association of FcγRIIa with DRMs depends on its palmitoylation on a cysteine residue . Despite these reports, the model of lipid rafts presents some limitations that need to be considered. For example, not all Fcγ receptors are palmitoylated (like FcγRIIa is) thus other receptors may not associate with lipid rafts or they would do by another mechanism. Interestingly, a transmembrane mutant form of FcγRIIa that failed to associate with lipid rafts was still able to trigger phagocytosis . Also, the use of methyl-β-cyclodextrin to eliminate cholesterol from the cell membrane may be a very harsh treatment and the functional condition of the cell afterwards is not clear. Moreover, lipid rafts disruption by cholesterol depletion did not inhibit phagocytosis in macrophages . In addition, there is still a debate whether DRMs really reflect the segregation of lipids in membranes or are artificially induced by the detergents used in their preparation. Thus, the model of lipid rafts needs to be considered with caution .
As mentioned above, different phagocytes express more than one activating FcγR, and at the same time they also express the inhibitory FcγRIIb. The coexpression of both activating and inhibitory FcγR results in simultaneous triggering of activating and inhibitory signaling pathways . Thus, a particular phagocyte will initiate phagocytosis when the sum of activating and inhibiting signals reaches a threshold of activation that is determined by the relative expression of both types of FcγR . The importance of the inhibitory FcγRIIb in regulating many IgG-mediated responses in different leukocytes was made evident in FcγRIIb-deficient mice, which showed enhanced activity of many IgG-mediated cell responses including phagocytosis . Another molecule that negatively regulates phagocytosis of macrophages is CD47 via SIRPα [78, 79]. Ligation of CD47 leads to phosphorylation of the immunoreceptor tyrosine-based inhibition (ITIM) motif in the cytoplasmic tail of SIRPα, which in turn recruits the phosphatase SHP-1 . By super-resolution microscopy, it has become evident that many receptors are found in clusters at the plasma membrane on a nanometer scale . In the case of resting macrophages, it was recently found that nanoclusters of Fcγ RI are constitutively associated with nanoclusters of SIRPα. Upon Fc receptor activation, Src-family kinase signaling leads to segregation of FcγRI and SIRPα nanoclusters , and co-ligation of SIRPα with CD47 prevented nanocluster segregation. Thus, when the balance of signals favors activation, FcγRI nanoclusters are separated from the inhibitory signal .
After FcγR phosphorylation, Syk binds to the ITAM motifs and gets also activated. Syk has also been shown to be required for phagocytosis [38, 70] and it is responsible for activation of several additional signaling proteins that get recruited to the FcγR signaling complex (Figure 1). The transmembrane protein LAT (linker for activation of T cells) is phosphorylated by Syk. Phosphorylation of LAT induces docking of additional adaptors: Grb2 binds to LAT, and in turn it recruits Gab2 (Grb2-associated binder 2). Gab2 is also phosphorylated by Syk. Other proteins are then also recruited to the complex. Among them is phospholipase C (PLC) γ1, which produces inositoltrisphosphate (IP3) and diacylglycerol (DAG). These second messengers cause calcium release and activation of protein kinase C (PKC), respectively. PKC leads to activation of extracellular signal-regulated kinases (ERK and p38) . The guanine nucleotide exchange factor (GEF) Vav activates GTPases of the Rho and Rac family, which are involved in regulation of the actin nucleation complex Arp2/3, which induces the actin polymerization that drives pseudopod extension. Other enzymes such as phosphatidylinositol 3-kinase (PI 3-K) activate the GTPase Rac and nuclear factors like NF-κB (Figure 1).
3.1.1. Lipid Signals
Signaling events regulating phagosome formation have also been examined by fluorescence imaging techniques. Detection of lipids and several activating proteins has shown that different molecules associate and dissociate from phagosomes in an orderly fashion (Figure 2). Phosphatidylinositol-4,5-bisphosphate [PI(4,5)P2] is present in large amounts in the inner leaflet of the plasma membrane of resting phagocytes. During phagocytosis, the concentration of PI(4,5)P2 increases in the pseudopods that form the phagocytic cup but then decreases abruptly . The drastic disappearance of PI(4,5)P2 following its modest initial accumulation is essential to allow particle internalization, probably by facilitating actin disassembly . Several pathways contribute to the disappearance of PI(4,5)P2. PLCγ is phosphorylated and recruited to the phagocytic cup in a Syk-dependent manner, probably by interaction with LAT [83, 85]. PLCγ activity is critical because its inhibition prevents DAG production and blocks phagocytosis . In addition, DAG leads to activation of PKCε, which enhances phagocytosis . PI(4,5)P2 is also consumed when it becomes phosphorylated by PI-3K, producing PI(3,4,5)P3 at the phagocytic cup . PI-3K is recruited and activated by Syk , or by adaptor proteins such as Gab2  (Figure 1). These dramatic changes in membrane lipid composition during Fcγ receptor-mediated phagocytosis demonstrate that distinct molecules are activated and recruited in a carefully orchestrated manner to induce phagosome formation.
3.1.2. Small GTPases
Small GTPases of the Rho family are important regulators of the actin cytoskeleton. These enzymes function as molecular switches alternating between an active (GTP-bound) state and an inactive (GDP-bound) state . For activation, they need to release GDP and replace it with GTP. This action is catalyzed by guanine nucleotide exchange factors (GEFs). Later, GTP is hydrolyzed to GDP returning the GTPase to its inactive state. This last step is enhanced through interactions with GTPase-activating proteins (GAPs). The GTPases Rac and Cdc42 are activated and recruited to the forming phagosome during Fcγ receptor-mediated phagocytosis (Figure 1) . Cdc42 is activated early in phagocytosis mostly at the rims of the phagocytic cup  (Figure 2). Rac1 is activated throughout the entire nascent phagosome, whereas Rac2 is activated later, mostly at the base of the phagocytic cup  (Figure 2). Cdc42 and Rac participate in regulating the localized formation of actin fibers, necessary for pseudopod extension, by activating the nucleation-promoting factors WASp (Wiskott-Aldrich Syndrome protein) and Scar/WAVE, respectively  (Figure 1). WASp and Scar, in turn, activate the Arp2/3 complex for actin polymerization  (Figure 1).
3.2. Complement Receptor Signaling
The integrin CR3 is the best-studied phagocytic complement receptor. For a long time, it has been recognized that engagement of CR3 on macrophages triggers a distinct form of phagocytosis, characterized by “sinking” of the particle into the cell without forming the characteristic pseudopods of FcγR phagocytosis . However, this idea has been questioned by recent microscopy observations that showed membrane protrusions encircling the targets during CR3-mediated phagocytosis [62, 96]. Still, it is thought that integrin CR3 signaling for phagocytosis is very different from FcγR signaling. Early reports demonstrated that phagocytosis of complement- opsonized zymosan and of complement-opsonized erythrocytes was unaffected by tyrosine kinase inhibitors . This ruled out the participation of tyrosine kinases in this type of phagocytosis. In addition, macrophages from Syk −/− mice showed normal levels of CR-mediated phagocytosis . However, β2 integrin stimulation by adhesive ligands, or by artificial integrin cross-linking with antibodies induced various cellular responses in a Src and/or Syk kinase-dependent manner . More recently, it was shown that Syk is phosphorylated during CR3-mediated phagocytosis and its inhibition prevents particle ingestion . Also, Syk can be indirectly activated by integrins via the ITAM-bearing FcR γ chain and/or DAP12 . The reason Syk −/− macrophages are capable of CR-mediated phagocytosis while the other experimental systems clearly implicate Syk in integrin signaling remains a mystery. It might be possible that genetically deficient cells have upregulated other molecules, for example, Zap70, that allow the bypass of Syk during CR-mediated phagocytosis.
Other differences between FcγR- and CR-mediated phagocytosis seem to be the cytoskeleton requirements for particle internalization. The actin cytoskeleton is required for FcγR-mediated phagocytosis, whereas the actin and microtubule cytoskeletons are required for CR-mediated phagocytosis [97, 102]. Moreover, in complement phagocytosis F-actin accumulation and particle ingestion depend on RhoA, but not on Rac or Cdc42 [103, 104], and binding of iC3b-opsonized erythrocytes increased levels of Rho-GTP but not of Rac-GTP . However, ingestion of iC3b-opsonized erythrocytes is reduced in cells where Rac1 and Rac2 were deleted . Together these findings challenge the classical model that CR3-mediated phagocytosis depends only on RhoA .
Rho, in turn, leads to actin polymerization via two mechanisms (Figure 3). First, Rho can activate Rho kinase, which phosphorylates and activates myosin II . Inhibition of Rho kinase activity also prevents accumulation of Arp2/3 and actin assembly at the phagocytic cup . Second, Rho can induce accumulation of mDia1 (mammalian diaphanous-related formin 1) and polymerized actin in the phagocytic cup. Interfering with mDia activity inhibits CR3-mediated phagocytosis while having no effect on FcγR-mediated phagocytosis . Also, mDia1 binds directly to the microtubule-associated protein CLIP-170 and induces its accumulation at the phagocytic cup . This pathway also provides a link to the microtubule cytoskeleton required for CR-mediated phagocytosis [97, 102]. Thus, microtubules and actin seem to function cooperatively in CR-mediated phagocytosis (Figure 3).
The signaling pathway for Rho activation is not clearly defined. Two regions in the cytosolic domain of the β2 subunit of the integrin receptor are important for Rho activation during phagocytosis , but it is not clear how the integrin connects to a Rho GEF for activation. In addition, Vav (a Rho/Rac GEF) originally reported to participate in FcγR-mediated phagocytosis, but not in CR-mediated phagocytosis , can also activate Rho . Since, Rho participates in Arp2/3 activation and actin polymerization by CR3  and Vav is a substrate for Syk , it is possible that a connection exists for Rho activation via Syk and Vav  (Figure 3).
4. Phagosome Formation
As indicated before, phagocytosis commences by interaction of phagocytic receptors with ligands on the surface of target particles. Then, receptors must aggregate to initiate signaling pathways that regulate the actin cytoskeleton, so that the phagocyte can produce membrane protrusions for involving the particle. Finally, the particle is enclosed in a new vesicle that pinches out from the plasma membrane.
4.1. Initial Interactions
The initial interactions of phagocytic receptors with the particle are not easy, since receptor ligands do not usually cover the particle uniformly and receptors are not freely accessible on the cell membrane. In fact, most phagocytic receptors are short molecules that extend only around 5 nm from the surface of the cell (Figure 4(a)) and are found among many much longer, usually rigid, transmembrane glycoproteins present throughout the membrane. These glycoproteins form a thick layer, known as glycocalyx, covering the cell membrane, that can effectively conceal short receptors . Mucins, high molecular weight, heavily glycosylated proteins, CD44 and hyaluronan, and transmembrane phosphatases such as CD45 and CD148 are components of the glycocalyx that can reduce ligand access to receptors on the phagocyte membrane (Figure 4(a)). In addition, the lateral diffusion of receptors on the cell membrane can be effectively reduced by glycocalyx components that are tethered to cytoskeletal structures. These glycoproteins effectively act as the “pickets” of a cytoskeletal “fence” [13, 14] that impedes free diffusion of other membrane molecules. This is the case for phagocytic receptors, which move only in discrete areas on the cell membrane among these immobile picket fences (Figure 4(b)).
Phagocytes improve interactions of receptors with possible targets by (i) creating active membrane protrusions that allow the cell to explore larger areas, increasing the chances for receptors to engage their ligands [61, 113], and by (ii) selectively removing some of these larger glycoproteins allowing the receptors to diffuse more freely on the membrane . The phosphatase CD45 can extend more than 40 nm from the cell membrane , and it is a real steric obstacle for phagocytic receptors. Removing these large molecules could greatly improve receptor binding. Indeed, removal of CD45 was first observed during Dectin-1-mediated phagocytosis in a structure that was called “phagocytic synapse” , for its similarity to the T lymphocyte immune synapse . When T cell receptor (TCR) molecules on the T lymphocyte interact with MHC/peptide molecules on an antigen-presenting cell, a central cluster of engaged TCR is formed. The TCRs are surrounded by a ring of integrin LFA-1 (lymphocyte-function-associated antigen-1) molecules, and CD45 is excluded from the central area. TCR interactions span around 15 nm, while integrin interactions span around 30–40 nm between the two cells. Thus removal of the larger molecules helps an efficient TCR interaction. A similar situation for FcγR-mediated phagocytosis has also been elegantly described recently by Sergio Grinstein’s group .
Besides its steric interference, there is another reason for removing CD45 from FcγRs. The tyrosine phosphatase CD45 must be taken away from sites of FcγR engagement to allow full activation of Src tyrosine kinases, which phosphorylate ITAM sequences needed for activation of phagocytosis signaling . First, CD45 must be allowed to diffuse more on the membrane. The lateral diffusion of CD45 is restricted by interactions between its cytoplasmic domain with ankyrin and spectrin molecules that connect to the actin cytoskeleton . These interactions can be reduced by signals that alter the cytoskeleton and prime the cell for phagocytosis. TLR ligands, for example, LPS and bacterial DNA, can reduce the restricted diffusion of immunoreceptors . Second, the more motile CD45 molecules need to be kept away from the engaged phagocytic receptor. This is achieved by the creation of a diffusion barrier made of activated integrins . FcγRs (and also G-protein coupled receptors or TLR) deliver signals for inside-out activation of integrins. Inactive integrins exist in a bent conformation that does not bind ligands. The signal from FcγR can produce DAG and Ca 2+ , which together activate CalDAG-GEF1 (a GEF for Rap). The small GTPase Rap in its GTP form is then able to recruit RIAM and talin to the cytoplasmic tail of the β subunit of integrins  (Figure 4(c)). This triggers the unfolding of the integrin into a high affinity “active” state. Kindlin-3 is another molecule that also binds to the β subunit of integrins causing their activation [121, 122]. The extended active integrin can then bind to many different ligands on the target particle . Thus, integrins participate in FcγR-mediated phagocytosis by promoting adhesion to the opsonized particle . In addition, the integrin molecules that get engaged by ligands get also tethered to the actin cytoskeleton and, with this, they form a diffusional barrier for CD45 molecules. The extended integrin bound to the target particle effectively pushes out the larger glycocalyx components, such as CD45 (Figure 5). As more integrin molecules get engaged they function as a progressive wave migrating ahead of the engaged FcγRs, allowing new receptors to aggregate in microclusters  (Figure 5).
4.2. Actin Remodeling in Membrane Protrusions
After a target particle is detected, the phagocytic process requires remodeling of the actin cytoskeleton to promote changes of the plasma membrane. The process is very complex and we have only a partial understanding of it. However, several important steps directed by actin remodeling, to form the pseudopodia that will cover the particle, can be identified. First, the membrane-associated cortical cytoskeleton, of the resting phagocyte, needs to be disrupted. Second, nucleation of actin filaments takes place in order to initiate F-actin polymerization and extension of pseudopodia. Third, actin gets depolymerized from the base of the phagocytic cup and the phagosome is closed at the distal end . These steps of the precise temporal and spatial activation and inactivation of multiple proteins that govern F-actin dynamics are described next and presented in Figure 6.
A resting phagocyte presents a membrane-associated cortical cytoskeleton that provides cell shape. Upon activation, this cytoskeleton is disrupted by the action of coronins (F-actin debranching proteins)  and cofilin  and gelsolin  (F-actin-severing proteins). Coronin 1 rapidly accumulates at the nascent phagosome during both FcγR- and CR-mediated phagocytosis , and, in macrophages, it can interact with F-actin and inhibit the Arp2/3 complex . Coronin 1 debranches F-actin leaving linear fibers that can be severed by cofilin and gelsolin (Figure 6, step (b)). Their activity is controlled by modulating their association with filaments, or by sequestering them away from filaments by binding to phosphoinositides, such as PI(4,5)P2 [127, 128]. In addition, the vesicular OCRL phosphatase activity to hydrolyze PI(4,5) P2 seems to contribute to the step of actin depolymerization . The role for these enzymes in phagocytosis is much more complex than just described, and future research is needed in this area . This initial disruption of the cytoskeleton has two consequences: it provides G-actin monomers for incorporation into new filaments and increases the mobility of nonligated receptors on the membrane (see previous section). The second step is the nucleation of actin filaments to initiate F-actin polymerization and extension of pseudopodia (Figure 6, step (c)). This is achieved mainly by the action of the Arp2/3 protein complex, which can be stimulated by different pathways. In fact, as indicated above the signaling pathways triggered by the best-studied phagocytic receptors, namely, FcγRs and CRs, are very different (see Figures 1 and 3). For FcγR-mediated phagocytosis, Arp2/3 is recruited to the nascent phagocytic cup, where its actin-nucleating activity is stimulated by WASp and N-WASp [130, 131], which in turn are activated by Cdc42-GTP and PI(4,5)P2 . In the case of CR-mediated phagocytosis, actin polymerization is associated with RhoA . This GTPase recruits and stimulates mDia formins , which in turn also activate the Arp2/3 complex (Figure 3). However, other GTPases, such as Rap, seem to play a role in CR-mediated phagocytosis, independently of RhoA . Rap-GTP also activates profilin, which is essential for actin polymerization via formins . Rap can also activate the GTPase Rac . But as discussed earlier, the role of Rac in complement-mediated phagocytosis remains a subject of debate.
4.3. Phagosome Sealing
The last step in phagosome formation is characterized by elimination of F-actin from the base of the phagocytic cup, just before the membrane protrusions fuse at the other end to seal the nascent phagosome (Figure 6, panel (d)). Depolymerization of actin filaments from the phagocytic cup may also facilitate curving of the membrane around the particle and provide room for fusion of internal vesicles, a source of endomembranes . The mechanism for actin removal from the forming phagosome has been poorly defined, and much more research is needed in this topic. The mechanism for removing F-actin must include the termination of actin polymerization and the detachment and depolymerization of existing filaments. Both steps seem to be controlled by phosphoinositides, in particular PI(3,4,5)P3, the product of PI-3K. Inhibition of this enzyme prevents depolymerization of actin at the base of the phagocytic cup and arrests extension of pseudopods . PI(3,4,5)P3 can activate Rho-family GAPs, which will induce deactivation of the GTPases stimulated during phagocytosis [137, 138]. Supporting this idea is the fact that PI-3K inhibition causes accumulation of activated Cdc42 and Rac at the phagocytic cup [92, 137]. However, because inhibition of PI-3K blocks phagocytosis even when GTPases are constitutively activated , this enzyme must control other molecules important for phagocytosis. One such molecule is PI(4,5)P2, which decreases by the action of PI-3K, but also by the action of PLCγ. Since PI(4,5)P2 sequesters cofilin and gelsolin and it is required for WASp activation, its reduction will increase F-actin severing (by liberation of cofilin and gelsolin) and reduce actin polymerization (by inhibition of WASp) . Other molecules regulated by PI(3,4,5)P3 are myosins. Myosins exert contractile activity that functions as a purse string to facilitate phagosome closure [139–142] (Figure 6, step (d)).
Recently, the process of phagosome formation and closure has been revisited thanks to live microscopy with the technique of total internal reflection fluorescent microscopy (TIRFM) . In this way, an important role for dynamin-2 in phagosome formation was revealed. Dynamin-2, which mediates the scission of endocytic vesicles, was recruited along with actin during phagosome formation, and depolymerization of actin led to impaired dynamin-2 recruitment or activity. Also, dynamin-2 accumulated at the site of phagosome closure . Thus, it seems there is a cross-talk between actin and dynamin for phagosome formation and closure before dynamin functions for scission .
5. Phagolysosome Maturation
The phagosome changes its membrane composition and its contents, to turn into a phagolysosome, a vesicle that can destroy the particle ingested. This transformation is known as phagosome maturation (Figure 7) and consists of successive fusion and fission interactions between the new phagosome and early endosomes, late endosomes, and finally lysosomes. At the end, the mature phagosome, also called phagolysosome, has a different membrane composition, which allows it to contain a very acidic and degradative environment [145, 146].
5.1. Early Phagosome
The new phagosome rapidly gets the properties of early endosomes, by fusing with sorting and recycling endosomes . Its interior becomes a little acidic (pH 6.1–6.5) but it is not very destructive. Membrane fusion events between the phagosome and early endosomes are regulated by the small GTPase Rab5 [147, 148]. This membrane GTPase is required for the transition from an early to a late phagosome. Rab5 functions through the recruitment of EEA1 (early endosome antigen 1), which promotes fusion of the new phagosome with early endosomes . Rab5 also recruits class III PI-3K human vacuolar protein-sorting 34 (hvPS34), which, in turn, generates phosphatidylinositol 3-phosphate [PI(3)P] . This lipid then helps fix EEA1 to the cytosolic face of the phagosome and promotes recruitment of other proteins involved in phagosome maturation, including Rab7, a marker of late endosomes [151, 152]. EEA1 functions as a bridge that tethers early endosomes to incoming endocytic vesicles  and binds to syntaxin 13, a SNARE (soluble NSF-attachment protein receptor) protein required for membrane fusion . Despite fusion with multiple early endosomes, the new phagosome does not seem to change size. This is due to the retrieval of vesicles to endosomes and the trans-Golgi network. Acidification of the phagosome lumen results from the gradual accumulation of active V-ATPases on the phagosome membrane. This V-ATPase is a multimeric protein complex that translocates protons (H
) into the lumen of the phagosome using cytosolic ATP as an energy source [155, 156] (Figure 7). In order to keep an electrical balance across the phagosome membrane, negative anions (mainly Cl
) also move inside, while cations (such as K + and Na + ) move outside [157, 158].
5.2. Intermediate Phagosome
As maturation proceeds, Rab5 is lost, and Rab7 appears on the membrane. The vpsC-homotypic protein-sorting (HOPS) complex mediates the transition from Rab5 to Rab7 endosomes  and may function in a similar fashion in phagosome maturation. Rab7 mediates the fusion of the phagosome with late endosomes . At the same time, intraluminal vesicles are now formed. They contain membrane-associated molecules that are intended for degradation. These vesicles seem to arise from inwards budding and pinching of the limiting membrane of the phagosome . The membrane proteins marked for degradation are ubiquitinated and associate with the endosomal-sorting complex required for transport (ESCRT) . This complex forms a circular array that directs the vesicles into the lumen of the phagosome  (Figure 7).
5.3. Late Phagosome
Once the intermediate phagosome eliminates the proteins that will be recycled or degraded, it continues maturation to a late phagosome. Rab7 accumulates and becomes a marker for this stage. Rab7 recruits new proteins to the membrane. One such protein is Rab-interacting lysosomal protein (RILP), which binds to the dynein-dynactin complex [162, 163] and brings the phagosome in contact with microtubules. This mediates the centripetal movement of late phagosomes and lysosomes [162, 163] that brings the organelles in close contact so that SNARE proteins, such as VAMP (vesicle-associated membrane protein) 7 and VAMP8 can complete membrane fusion [164, 165]. At this stage, the lumen gets more acidic (pH 5.5–6.0), thanks to more V-ATPase molecules on the membrane  (Figure 7). In addition, lysosomal-associated membrane proteins (LAMPs) and luminal proteases (cathepsins and hydrolases) are incorporated from fusion with late endosomes or from the Golgi complex [145, 146].
The last stage in the maturation process involves fusion of late phagosomes with lysosomes, to become phagolysosomes. Phagolysosomes are the ultimate microbicidal organelle . Phagolysosomes count with many sophisticated mechanisms directed to eliminate and degrade microorganisms. They are highly acidic (pH as low as 4.5) thanks to the large number of V-ATPase molecules on their membrane . Phagolysosomes are also characterized by a PI(3)P-enriched internal membrane [166, 167] and by the lack of mannose-6-phosphate receptors . They also contain a number of hydrolytic enzymes, including various cathepsins, proteases, lysozymes, and lipases . Other microbicidal components of the phagosome are scavenger molecules, such as lactoferrin that sequesters the iron required by some bacteria  and the NADPH oxidase that generates superoxide (
)  (Figure 7). Superoxide can dismutate to H2O2, which can in turn react with to generate more-complex reactive oxygen species (ROS), such as hydroxyl radicals and singlet oxygen . In addition, H2O2 can be combined with Cl − ions into hypochlorous acid by the enzyme myeloperoxidase .
Phagocytosis is an elegant and very complex process for the ingestion and elimination of pathogens and apoptotic cells. It is performed by a series of cells we call professional phagocytes. They are monocytes, macrophages, neutrophils, dendritic cells, osteoclasts, and eosinophils. It is evident that phagocytosis is fundamental for tissue homeostasis, controlling important aspects of inflammation and the immune response. Clearly, the many cell types that can perform phagocytosis and the overwhelming number of different phagocytic targets require more than one mechanism to complete this cellular function. We have presented the main four steps of phagocytosis to provide a general view of the whole process. Still, we have to keep in mind that this description corresponds primarily to opsonic receptors. We have very little knowledge of the signaling pathways other phagocytic receptors activate. Similarly, the process of phagosome maturation has gained much information from studies on vesicular traffic. Yet, important gaps remain in every step. Also, how the final phagolysosome completes its antimicrobial or degradative functions is not completely clear. But, the fact that several microbial pathogens have developed special ways for interfering with phagolysosome function gives us another opportunity to learn from them novel aspects on phagocytosis. In addition, the resolution of the phagolysosome, after the infection or the inflammation processes have terminated, is an area that has brought very little attention. What are the molecular details and functional implications of ingesting different particles? How the various phagocytic receptors on the same phagocyte cooperate? And how the various phagocytes participate in tissue homeostasis? These are important questions that future research in this exciting area will have to address. An improved understanding of phagocytosis is essential for the clear implications it has for antigen presentation and autoimmune disease.
Conflicts of Interest
The authors declare that they do not have any conflicts of interest in the subject discussed in this review.
The authors thank Lilian Araceli González Hernández for preparing the list of references. Research in the authors’ laboratory was supported by Consejo Nacional de Ciencia y Tecnología, Mexico (Grant 254434 to Carlos Rosales), and by Dirección General de Asuntos del Personal Académico, Universidad Nacional Autónoma de México, Mexico (Grant PAPIIT IA202013-2 to Eileen Uribe-Querol).
Copyright © 2017 Carlos Rosales and Eileen Uribe-Querol. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
How LAP is distinguished from macroautophagy
The use of autophagy-related proteins in LAP blurs the border between phagocytosis and autophagy. As both autophagy and phagocytosis can degrade self or foreign cargo, the original source of the cargo cannot be used to distinguish between these pathways. However, given the different origins of phagosome and autophagosome membranes (Figure 1), there are differences between the molecular mechanisms of LAP and autophagy. Proteins involved in phagocytosis, such as engulfment receptors on the plasma membrane, are not required for autophagy (Sanjuan et al., 2007 Martinez et al., 2011 Fazeli et al., 2016 ). LAP employs receptors on the cell surface to selectively recognise cargo. Activation of receptor signalling leads to remodelling of the cortical actin cytoskeleton to extend the plasma membrane and wrap the cargo, a process that is not required for autophagy.
Although many proteins of the autophagy pathway are used during LAP, a few autophagy markers can be used to distinguish autophagy from LAP. As LAP does not involve formation of a double-membrane phagophore, the autophagy pre-initiation complex is not required for LAP. ULK1, FIP200 and Atg13 are subunits of this complex and are each dispensable for LAP (Martinez et al., 2011 Henault et al., 2012 ). The subunit composition of the PI3K complex involved in LAP also differs from that of autophagy. The LAP complex contains the RUN domain protein Rubicon, whereas the autophagy complex includes Atg14 (Martinez et al., 2015 ). Thus, molecular distinctions exist between autophagy and LAP.
Given the discovery of non-canonical forms of autophagy, detection of LC3-positive double-membrane vesicles or increased levels of lipidated LC3 (LC3-II) cannot be considered ultimate proof of autophagy induction. However, since LC3 is found on both the inner and outer autophagosome membrane (Figure 1A), but only on the outer surface of LC3-associated phagosomes (Figure 1B), more LC3 is expected to localise to autophagosomes than to LAPosomes. In fact, LC3 reporters appear relatively dim on midbody LAPosomes in comparison with autophagosomes degrading P-granules and paternal mitochondria in Caenorhabditis elegans embryos (Fazeli et al., 2016 ). In summary, these molecular differences can be used to distinguish between LAP and autophagy.
R ituximab T argeting CD20
Rituximab (IDEC-C2B8, Rituxan® IDEC Pharmaceuticals, San Diego, CA, and Genentech, Inc, San Francisco, CA) is a genetically engineered chimeric mAb that specifically binds to CD20, which contains human γ-1 and κ constant regions with murine variable regions . Rituximab therapy alone or in combination with chemotherapy has remarkably improved the treatment outcome of patients with B-cell NHL. Its single-agent efficacy with low toxicity was proven by an overall response rate of 50% with a median time to progression in responders of 13.2 months, but some responsive patients develop resistance to further rituximab treatment [5, 6]. The mechanism of rituximab unresponsiveness remains unclear, and rituximab resistance requires more extensive investigation.
The rituximab-targeted molecule, CD20, a 32-kDa nonglycosylated phosphoprotein, provides a universal target for immunotherapy, which is especially expressed on the surface of normal precursor and mature B cells and, importantly, not on early pre-B cells, stem cells, or antigen-presenting dendritic reticulum cells . More than 90% of B-cell NHLs express this surface protein [8, 9]. The CD20 protein has four transmembrane domains and does not modulate from the cell surface in response to antibody binding, thus providing an excellent target for immunotherapeutic strategies . So far, the natural ligand for CD20 has not been identified, and the biological function of CD20 remains unclear. CD20 knockout mice show normal B-cell development and function . Via phage display libraries, Binder et al.  showed that rituximab binds a discontinuous conformational epitope on CD20, comprising (170)ANPS(173) and (182)YCYSI(185), with both strings brought in steric proximity by a disulfide bridge between C(167) and C(183). This structural interaction was further proven by crystal structure analysis of the complex of rituximab Fab fragment and CD20 fragment (aa163-aa187) . Several distinct antitumor activities of rituximab have been suspected in B-cell NHL therapy, including complement-dependent cytotoxicity (CDC), antibody-dependent cellular cytotoxicity (ADCC), apoptosis, and direct growth arrest . In addition, complement-dependent cellular cytotoxicity (CDCC) has often been included in ADCC .
The Complement System
The complement system, a central innate system, is the effector of adaptive immunity. It spontaneously identifies any potential pathogens and efficiently protects the host from intruding pathogen attack through a wide range of cellular responses [15–18]. This system is composed of >30 soluble plasma proteins and membrane proteins that can trigger three distinct protease cascades known as the classical, mannose-binding lectin (MBL), and alternative pathways (Fig. 1). The classical pathway is triggered by antigen-bound antibody molecules and is initiated by the binding of a specific part of the antibody (Fc) to C1q . The MBL pathway is initiated when plasma MBL in complex MBL and MBL-associated protease (MASP)-1/2 bind to arrays of carbohydrates (specifically mannose and fucose residues) on the surface of many pathogens such as bacteria, viruses, fungi, yeasts, and parasites MASP-1/2 then cleaves C4 and triggers the subsequent complement cascade . The alternative pathway is capable of spontaneous autoactivation, termed the “tickover” of C3 at a rate of ∼1% of total C3 per hour [19, 20], thereby identifying any potential pathogens. All three pathways converge at the C3 level, which allows C3b to target the pathogens, and at the C5 level, which leads to polymerization of C9 by C5b-8 binding and to assembly of a membrane attack complex (MAC) with a diameter of 5–10 nm (Fig. 1) . If the MAC inserted into the cell membrane remains open, it will directly induce targeted cell lysis through subsequent influx of ions and water that leads to lethal colloid-osmotic swelling, CDC. The CDC effect is characterized by swelling of mitochondria, dilation of the rough endoplasmic reticulum, disruption of the Golgi complex and of the plasma and nuclear membranes, and heterochromatin disappearance at the ultrastructural level involving the generation of reactive oxygen species [21, 22]. For targeted cell lysis, a single MAC is enough for erythrocytes but not for nucleated cells, because nucleated cells can endocytose MAC and repair the damage unless multiple MACs (12- to 16-mers) are assembled .
Three complement activation cascades.
Abbreviations: C1 INH, C1 inhibitor C4BP, C4 binding protein CR, complement receptor DAF, decay-accelerating factor MAC, membrane attack complex MBL, mannose-binding lectin MCP, membrane cofactor protein MASP, MBL-associated proteases.
The byproducts produced in complement activation, such as C1q, C3b, iC3b, and C4b, are critical opsonins for host defense against infection and for disposal of immune complexes and dead cell debris by the phagocytosis/lysis effect of the immune effector cells (macrophages, neutrophils, natural killer [NK] cells, etc.) through their surface receptor binding to these byproducts (Fig. 1 and 2). On the other hand, the small fragment byproducts such as C3a, C4a, and C5a (Fig. 1), termed anaphylatoxins, also play an important role in inflammation and especially in host defense against parasites. These anaphylatoxins can cause mast cell and basophil degranulation, with the release of histamine and other substances that increase vascular permeability, stimulate smooth muscle constriction, and induce chemotaxis. The anaphylatoxins can also activate these immune effector cells by binding to cell surface receptors [24, 25]. Among these anaphylatoxins, C5a has the most potent biological activity.
Rituximab-mediated CDC effect and strategies for enhancing the CDC effect. C1q interacts with the rituximab Fc region exposed after binding to CD20 on the B-cell surface, thus activating the classical complement cascade, and a MAC is inserted into the cell membrane, with multiple MACs (12- to 16-mers) leading to cytolysis. Strategies to overcome resistance to the CDC effect include: (a) inhibitors of the mCRPs DAF and especially CD59, which leads to more complement activation and more MAC formation on the cell surface (b) heteroconjugates of rituximab to CVF or C3b, and other Ag-Ab complexes targeting tumor cells, which enhance complement activation and (c) other drugs that can upregulate the expression of CD20, which include the histone deacetylase inhibitor trichostatin A and protein kinase C activator bryostatin-1.
Abbreviations: Ag-Ab, antigen-antibody CDC, complement-dependent cytotoxicity CVF, cobra venom factor DAF, decay-accelerating factor MAC, membrane attack complex MCP, membrane cofactor protein mCRP, membrane complement regulatory protein.
To prevent the potentially harmful effect of complement activation on autologous cells, >10 plasma- and membrane-bound inhibitory proteins have evolved for restricting complement activation at different stages of activation pathways (Fig. 1 and 2) [26, 27]. The soluble plasma complement regulatory proteins include C1 inhibitor, which regulates C1 factor H and factor I, which regulate the cleavage of C3b and C3/C5 convertases C4 binding protein, which splits C4 convertase and assists factor I in the cleavage of C4b and S-protein, clusterin, and serum lipids, which compete with membrane lipids for reacting with nascent C5b67 . Membrane complement regulatory proteins (mCRPs) include: (a) CD46 (membrane cofactor protein), which regulates C3 activation by functioning as a cofactor protein for factor I–mediated cleavage of C3b and C4b  (b) CD55 (decay-accelerating factor [DAF]), which inactivates the convertases of C3 (C4b2a and C3bBb) and C5 (C4b2a3b and C3bBb3b) by preventing the formation of new and accelerating the decay of activated convertase via binding to C3b and C4b  (c) CD59 (membrane inhibitor of reactive lysis), which restricts MAC assembly on the cell membrane through binding to C8α and C9 (d) CD35 (complement receptor 1), which has both CD46 and CD55 functions in humans [26, 27, 30] and is a major immune adherence receptor that plays a role in immune-complex processing and clearance  and (e) complement-receptor 1-related protein y, present only in rats and mice, which possesses both CD46 and CD55 biological functions [32–35]. CD55 and CD59 attach to the cell surface via a glycosylphosphatidylinositol (GPI) anchor, whereas CD46 and CD35 associate with the plasma membrane via their C-terminal transmembrane domains . Structurally, CD55, CD46, and CD35 belong to the regulators of complement activation (RCA) family and contain a variable number of short consensus repeat domains. CD59 is a much smaller protein with no sequence or structural resemblance to the RCA family of proteins . Several lines of evidence from human and animal studies indicate that CD59 is more relevant than CD55 and CD46 in protecting normal cells from MAC formation and MAC-induced phenomena [37–42].
There is a delicate balance between complement activation and regulation on autologous cells that is subject to perturbation by either increased complement activation or decreased regulation, which may cause a variety of immune diseases. For example, immune-complex activation of the classical complement pathway results in autoimmune diseases such as systemic lupus erythematosus and glomerulonephritis [30, 43]. Conversely, deficiency of GPI-anchored DAF and CD59 in circulating cells—resulting from an acquired somatic PIG-A gene mutation—is responsible for the hemolytic anemia and thrombosis that characterize paroxysmal nocturnal hemoglobinuria [44–46]. However, the high expression level of mCRPs also confers protection on cancer cells, invading microorganisms including HIV for complement attack, and may lead to resistance to antibody therapy.
The formation of MACs and subsequent cytolysis are usually considered as the activity of CDC via activation of the classical pathway initiated by C1q binding to antibody Fc fragment in rituximab therapy (Fig. 2). The consequence of the formation of MACs generally depends on the copy numbers assembled on the targeted tumor cells. Only at high copy numbers (12- to 16-mers) per cell (a lytic dose), can MACs induce a loss of membrane integrity and rapid necrotic-type cell death, possibly by a caspase-independent mechanism [21–23, 47]. At low copy numbers of MACs per cell (nonlytic or sublytic dose), MACs exhibit a wide range of effects leading to cellular responses such as secretion, adherence, aggregation, chemotaxis, and even cell proliferation activity through various cell-signaling pathways [48–50].
There is much evidence to highlight the importance of CDC in rituximab therapy of B-cell NHL. Complement depletion by cobra venom factor (CVF) significantly reduced the antitumor activity of rituximab in severe combined immunodeficiency (SCID), athymic nude mice [51, 52]. This activity was not affected by depletion of macrophages, NK cells, and/or neutrophils, which are irrelevant to CDC, in either a disseminated or s.c. tumor mouse model. It was also completely abolished in a C1q-deficient syngeneic mouse model and in a CVF complement-depleted mouse model [53, 54]. Furthermore, use of inhibitors abrogating mCRP (such as CD55 and CD59) functions can facilitate rituximab therapy in B-cell NHL, confirming the importance of CDC. However, many groups have claimed that ADCC plays a pivotal role in the antiproliferative effect of rituximab. B-cell lymphomas were depleted in vivo mainly by innate monocytes/macrophages in an FcγR-dependent manner of isotype-specific mAb interactions with distinct FcγRs. This rituximab-mediated ADCC effect on B-cell lymphomas is completely effective in C3-, C4-, or C1q-deficient mice [55–57]. The explanation for the discrepancy between rituximab-mediated CDC and ADCC is complex and far from convincing . A number of factors, including tumor status, mCRP expression, tumor cell type, tumor-inoculating methods, and tumor growth period, may influence the contributions of CDC and ADCC differently. However, it is widely accepted that ADCC and CDC synergistically affect cytotoxicity in cancer cells through the ability of complement to promote inflammation and change the activation status of innate effectors . The byproduct of anaphylatoxin C5a during CDC is pivotal in activating these effector cells and can upregulate the expression of activating FcγR and downregulate the expression of inhibitory FcγRIIb via interaction with the C5a receptor, which enhances the ADCC effect in tumors [58–61]. Moreover, CDC-resistant cells are sensitive to ADCC and vice versa .
The potency of CDC in rituximab therapy is partially correlated with the level of CD20 expression on the B-cell NHL cell membrane [22, 63]. The higher the CD20 level, the more MACs can be assembled. Subsequently, targeted cells are directly lysed. With B-cell NHL cells freshly isolated from patients, a level of CD20 expression of >50 × 10 3 CD20 molecules per cell is a prerequisite for the lytic response arising from the mechanism of CDC . A sigmoidal or linear correlation between the CD20 expression level and the rituximab-mediated killing effect further demonstrates the important role of CDC in the rituximab-mediated anticancer effect [22, 62, 63]. Other reports indicate that the efficacy of CDC in rituximab therapy may be determined by the expression levels of CD20 and mCRPs (CD55, and especially CD59) , but not of individual CD20 or mCRPs [65, 66].
Overcoming Resistance to Rituximab Through Enhancing the CDC Effect
Abrogating mCRP Function
The mCRPs CD46, CD55, and CD59 play a critical role in tumor resistance to rituximab-mediated CDC (Fig. 2). These mCRPs are expressed widely in almost all cancer cells independent of their tissue of origin . The various expression levels of mCRPs may be regulated by the selective stress of complement attack , the stage of differentiation [68, 69], and host factors of neighboring tumor or stromal cells. These mCRPs in tumor cells restrict complement activation cascades in distinct stages and reduce MAC formation, thereby profoundly protecting the tumor from rituximab-mediated CDC (Fig. 1 and 2). The rituximab-resistant B-cell NHL cell lines of Raji and Ramos generated under the stress of rituximab and normal human serum (complement resource) exhibit upregulation of CD52 and the complement regulators CD55 and CD59, as well as downregulation of CD20 [67, 70]. The use of neutralizing antibodies abrogating the function of CD46, CD55, and CD59 markedly enhanced the antitumor activity of rituximab in vitro and in vivo [13, 14, 22, 63, 71, 72]. However, these antibodies are not applicable to clinical treatment because they would initiate undesired complement attack by binding to the host cells expressing mCRPs. To overcome this problem, Tedesco's group  developed two miniantibodies (MBs), MB-55 (against CD55) and MB-59 (against CD59), containing the human IgG1 hinge-CH2-CH3 domains, two single-chain variable fragments (scFv) without Fc fragment, a region necessary for activating the classical pathway. Furthermore, they generated two heteroconjugates of bio-MB55 and bio-MB59  through a three-step biotin–avidin system, which is currently employed in patients to target radionuclides to cancer cells for enhancing the anticancer effect [75–77]. They documented that these modified antibodies against CD55 and CD59 significantly increased the antitumor activity of rituximab in vitro  and in vivo . Alternatively, enhanced susceptibility can be achieved by downregulating the expression of mCRPs. Knockdown of CD55 expression via siRNA transfection attenuated the resistance of B-cell lymphoma or breast cancer cells to complement-mediated cytolysis by treatment with rituximab or trastuzumab . The chemotherapeutic drug fludarabine showed a synergistic effect with rituximab treatment in NHL tumors, likely through the downmodulation of the membrane expression of CD55 [78, 79]. Consistent with the critical role of CD59 for restricting MAC in normal cells, CD59 is also more relevant to protect tumor cells from immunotherapy including rituximab than CD46 and CD55. Therefore, it is imperative to develop effective and practicable inhibitors against mCRPs, especially CD59, for facilitating rituximab and even other antibodies in treating tumors. For this reason, we have searched for an inhibitor for human CD59. Intermedilysin (ILY), a cytolytic pore-forming toxin secreted by Streptococcus intermedius, lyses human cells exclusively because the toxin binds to human CD59 (hCD59) with high specificity [80–82]. The specificity for hCD59 is derived from binding of ILY domain 4 (ILYd4) to amino acids 42–58 in hCD59, which also participate in binding to C8 and C9 . Thus, we hypothesize that truncated ILYd4 would abrogate hCD59 function and thereby increase antibody-mediated CDC in cancer cells. If this is the case, ILYd4 may represent an innovative adjuvant for cancer immunotherapy in general and for antibody-resistant cancers in particular.
Increasing CD20 Expression
It is of note that the level of CD20 expression is related to the number of MACs formed on rituximab-targeted lymphocytes via binding to CD20, which determines whether targeted tumor cells can be lysed by CDC or not. Thus, an increased level of CD20 expression might be a means to overcome resistance to CDC (Fig. 2). However, rituximab–CD20 complex formed by rituximab binding to CD20 on the targeted cell membrane can be shaved by monocytes/macrophages expressing FcγRI, a phenomenon called antigen modulation [83–85] (Fig. 2). This shaving reaction starts within 1–40 hours after rituximab infusion, which could substantially compromise the therapeutic efficacy of rituximab . Interestingly, the histone deacetylase inhibitor suberoylanilide hydroxamic acid modulated the expression of apoptosis-related genes , and another histone, deacetylase inhibitor trichostatin A, could epigenetically increase CD20 mRNA and protein expression in an established CD20-negative cell line to sensitize rituximab therapy . Bryostatin-1, a protein kinase C activator, can also enhance expression of CD20 at the level of both mRNA and protein in human tumor B cells through extracellular signal–related kinase (ERK)-dependent mechanisms, which in turn increases the susceptibility of the tumor to rituximab . Moreover, synthetic CpG oligodeoxynucleotides are currently being tested in clinical trials as a vaccine adjuvant and for rituximab immunotherapy of B-cell NHL . Synthetic CpG oligodeoxynucleotides resembling sequences found in bacterial DNA specifically increase primary malignant B-cell expression of CD20, thereby resulting in enhanced sensitivity to rituximab treatment [89, 90].
Enhancing Complement Activation
Besides abrogating mCRP function and increasing the CD20 expression level, an alternative strategy that can be employed to overcome resistance to CDC is to directly enhance complement activation on tumor cells. A complement-activating protein such as CVF or C3b can be conjugated to the antitumor antibody for the enhancement of complement activation  (Fig. 2). This effect was achieved in vitro by either anti–epithelial cell adhesion molecule–CVF heteroconjugates targeted to colorectal cancer cells  or anti–GD2-CVF heteroconjugates targeted to neuroblastoma cells [93, 94]. Additionally, an antibody against iC3b was demonstrated to increase iC3b deposition, which allowed it to interact with effector cells containing both Fc and complement receptors, and therefore significantly enhancing rituximab-mediated killing of Raji and DB cells in a cynomologous monkey model  (Fig. 2).
Effect of ADCC in Rituximab Therapy
Together with CDC, ADCC also plays an important role in rituximab antitumor activity. ADCC triggers tumor cell killing through interaction between the Fc region of CD20 binding rituximab and FcγRs, particularly FcγRI and FcγRIII, activating receptors expressed on immune effector cells such as monocytes/macrophages, granulocytes/neutrophils, and NK cells (Fig. 3). This interaction, mediated by rituximab on activated effector cells, initiates a series of signaling pathways that lead to the release of inflammatory and/or cytotoxic immune modulators including cytokines, chemokines, proteases, and reactive oxygen species. Eventually, the activated monocytes/macrophages and granulocytes/neutrophils phagocytose the targeted cancer cells, whereas activated NK cells eliminate targeted lymphoma cells using the granzyme-perforin system. Generally, the resting effector cells where the inhibitory FcγRIIB is dominantly expressed cannot function unless activated. The inhibitory FcγRIIB is a potent regulator of ADCC in vivo, modulating the activity of FcγRIII on effector cells . The antitumor activity of rituximab is greatly reduced in FcγRI/FcγRIII-deficient mice, whereas disruption of the gene that encodes the inhibitory receptor FcγRIIB substantially enhanced antitumor activity. In addition, rituximab can induce the β-chemokine CCL3 in vivo, activating innate immune cells such as NK cells, macrophages, and polymorphonuclear cells, which increases the ADCC effect . These results suggest that rituximab-mediated ADCC is important for killing cancer cells [96, 97].
Rituximab-mediated ADCC and strategies to enhance the ADCC effect. CDCC is also classified here, which normally has little efficacy in rituximab therapy. This kind of cellular cytotoxicity can occur after binding of either the rituximab Fc region (in ADCC) or C1q, C3b, C4b, and iC3b (in CDCC) to their respective receptors, resulting in either phagocytosis or cell-mediated lysis of B-lymphocytes, depending on the effector cell type. The Fc region of cell-bound rituximab is recognized principally by either the activating receptors FcγRI/FcγRIII or the inhibitory receptor FcγRIIB, whereas the byproducts of complement activation are recognized by C1qR, CR1, or CR3 on effector cell surfaces. The strategies to overcome resistance to the ADCC/CDCC effect include: (a) anaphylatoxins (C5a, C3a) (b) some cytokines (IL-2, IL-12, M-CSF) that are able to activate effector cells and (c) CR3-specific polysaccharides such as β-glucan, which primes CR3 and therefore triggers cellular cytotoxicity.
Abbreviations: ADCC, antibody-dependent cellular cytotoxicity CDCC, complement-dependent cellular cytotoxicity CR, complement receptor IL, interleukin.
CDCC may also be engaged in destroying invading pathogens. In CDCC, deposition of C3b, iC3b, C1q, and C4b bound on the surface of targeted cells is able to opsonize phagocytosis through binding to their receptors on monocytes/macrophages, NK cells, and polymorphonuclear leukocytes [95, 98, 99] (Fig. 3). This process can be further enhanced by chemotactic and cell-activating anaphylatoxins such as C5a and C3a through their receptors on the above phagocytes [24, 25], which are pivotal mediators of the host defense against infection and of the disposal of immune complexes and products of inflammatory injury . To date, there has been little research on the effect of CDCC on rituximab therapy, perhaps because of its relative ineffectiveness .
Antibody-targeted tumor resistance to innate immune cells such as macrophages, NK cells, and neutrophils in ADCC may arise from the suppression of these immune effectors after long-term expansion of the tumor. Resistance to rituximab in some patients may be linked to a polymorphism in FcγRIIIa affecting isotype preference . Genetically modified NK cells carrying a chimeric antigen receptor that consists of a CD20-specific scFv antibody fragment conferred enhanced cytotoxic activity and could overcome NK-cell resistance of lymphoma and leukemia cells . In addition, the effect of some cytokines on ADCC-related resistance was tested. M-CSF in vitro enhanced the cytotoxicity of monocytes on Daudi B-cell lymphomas through upregulation of monocyte FcγRI and FcγRIII by 1.5-fold and 1.6-fold, respectively, whereas the expression of FcγRII remained unchanged . Interleukin (IL)-2 is a pleiotropic cytokine that activates selective immune effector cell responses associated with antitumor activity, including ADCC. Golay et al.  demonstrated in vitro that IL-2–activated NK cells strongly enhanced the therapeutic activity of rituximab through ADCC in primary B-cell NHL cells freshly isolated from patients. Furthermore, IL-2 synergistically enhanced rituximab efficiency in killing B-cell NHL cells in a xenograft model, in part, through activation and trafficking of monocytes and NK cells to tumors  a similar result was obtained with an anti-CD20–IL-2 immunocytokine . Additionally, IL-12 synergizes the rituximab ADCC effect through upregulating γ-interferon and interferon-inducible protein 10 expression and increasing NK cell lytic activity in vitro  (Fig. 3). The concomitant use of IL-12 and rituximab had only a modest effect in treating patients with B-cell NHL. The response rate in patients treated with IL-12 in combination with rituximab did not seem to be better than that seen with rituximab alone. Also, the sequential administration of IL-12 after rituximab did not result in additional clinical responses .
To enhance the CDCC effect in rituximab treatment, β-glucan can be applied to promote CR3-dependent cellular cytotoxicity. Generally, β-glucan is naturally exposed on the cell wall of yeasts and fungi. Moreover this CR3-mediated cytotoxicity that is responsible for host protection against yeast and fungus infection requires the binding of CR3 to both iC3b and β-glucan. Normally, this CDCC effect does not happen in rituximab therapy because tumor cells lack β-glucan . However, together with deposited iC3b generated by rituximab-mediated complement activation, the administration of β-glucan can trigger cytotoxic phagocytosis and degranulation of iC3b-coated tumor cells [109, 110], thus regressing tumor growth [110, 111] (Fig. 3). The application of β-glucan has consistently been documented to markedly increase the therapeutic efficacy of rituximab in B-cell NHL-inoculated SCID mice .
Complement receptors CR3 (CD11b/CD18) and CR4 (CD11c/CD18)
Complement receptors CR3 (αMβ2, CD11b/CD18, Mac-1) and CR4 (αXβ2, CD11c/CD18, p150,95) are members of the β2-integrin family, expressed on most white blood cells. Both receptors bind multiple ligands – for example iC3b, fibrinogen, ICAM-1 or pathogen-related ligands like LPS – and thereby play an important role in phagocytosis, adherence and migration. Although it is clear that the two receptors exhibit nonoverlapping functions, comparative studies are barely available  . CR3 and CR4 are important phagocytic receptors, leading either effective antimicrobial responses against pathogens or noninflammatory phagocytosis of apoptotic cells under physiological conditions of a healthy individual. Some pathogens however evolved to hijack these functions of CR3 and CR4. Extracellular pathogens are likely to avoid phagocytosis by these receptors, by blocking their function or even cleaving them from the surface of phagocytes. In contrast, intracellular pathogens often use them as an effective entry route into host cells, thereby causing a more severe infection.
As viruses are obligate intracellular pathogens, they can benefit from their ability to exploit complement-mediated phagocytosis. For instance, the opsonization of HIV-1 by complement causes an up to 10-fold higher productive infection of human dendritic cells compared to nonopsonized or only antibody-opsonized virus particles. Complement-dependent HIV infection is mediated by CR3  , which also modulates the signal transduction of Toll-like receptor 8 (TLR8). Modulated TLR8 signalling resulted in a lower expression of antiviral and inflammatory factors such as IL-1β, IL-6, TNF-α, IFN-β, myxovirus resistance protein A and IFN-stimulated genes, leading to enhanced infection [47, 48] . In addition, CR3 has been shown to mediate the complement-dependent enhancement of West Nile virus replication in mouse macrophages [49, 50] , and the infection of human monocytes with the Dengue fever virus  . The capsid of Rotavirus contains integrin ligand motif bearing viral peptides, which are shown to bind to β1 and β2 integrins including CR4, and thereby promoting viral entry to host cells  . A recent experiment suggests that complement receptors CR3 and CR4 also act as Hantavirus entry receptors  .
As phagocytic receptors, both CR3 and CR4 have antimicrobial roles during the immune response to infections. Both CR3 and CR4 are involved in the uptake and killing of Escherichia coli  or M. tuberculosis  . Out of the two receptors, CR3 was shown to be the dominant mediator of phagocytosis over CR4 in the case of iC3b opsonized S. aureus on human monocytes, neutrophils, monocyte-derived macrophages and dendritic cells [56, 57] . CR3 is also involved in the phagocytosis of Salmonella enterica  , Borrelia burgdorferi  or Mycobacterium kansasii  , and has a key role in the effective immune response against Listeria monocytogenes  .
Phagocytosis via CR3 and CR4 is an effective antimicrobial immune response however, some pathogens try to evade it. For instance, Group A Streptococcus (GAS), avoids phagocytosis by secreting the CR3 homologue GAS Mac-1-like protein (Mac). The Mac binds to CD16 (FcγRIII) on the surface of human polymorphonuclear cells (PMNs) and blocks receptor–antibody interactions as well as the binding of iC3b to CD11b, as CR3 and CD16 are physically and functionally linked. By that Mac inhibits opsonophagocytosis and also the production of reactive oxygen species, thereby decreasing pathogen killing  .
While some pathogens evolved the ability to avoid phagocytosis, others use this mechanism to enter host cells. As CR3 and CR4 have a main role in noninflammatory phagocytosis of apoptotic cells, it is an ideal target for intracellular pathogens to exploit for their entry, especially if additional danger signals are blocked by the microbe.
Uptake of F. tularensis, which is one of the most virulent pathogens known, is mediated by CR3 and CR4 in human dendritic cells. The internalization of F. tularensis is followed by its rapid growth inside cells resulting in cell death  . The same receptors are involved in the infection of human macrophages, while in human neutrophils it is mediated by CR1 and CR3  .
The RrgA adhesin containing pili of Str. pneumoniae enhances the CR3 dependent uptake of pneumococci by murine and human macrophages through a direct interaction with CR3. Macrophages harbouring higher numbers of viable bacteria are more likely to be destroyed prior to the complete eradication of the ingested particles. Moreover, the interaction between RrgA and CR3 leads to increased motility and migratory behaviour of macrophages, resulting in an earlier onset of septicaemia and a more rapid disease progression  .
The obligate intracellular pathogen M. leprae invades host cells via phagocytosis and proliferates within mononuclear phagocytes. Both CR3 and CR4 (and CR1) were shown to be involved in that process  .
The causative agent of Legionnaires' disease, Legionella pneumophila, multiplies in human monocytes and alveolar macrophages. CR1 and CR3 were shown to mediate the adherence of Leg. pneumophila, thereby contributing to the entry and intracellular proliferation of the bacteria, as these processes could be inhibited by specific antibodies for CR1 and CR3  .
Hajishengallis et al. described in detail the immune evasion mechanism of Porphyromonas gingivalis. The fimbriae of this bacteria serve as a ligand for CR3, which mediates its phagocytosis, thus proactively promoting its binding and entry into the host cells. The fimbriae activate the high-affinity conformation of CR3, which does not promote the killing but the persistence of Po. gingivalis after internalization and induces a selective suppression of IL-12 production [66-68] . Therefore, Po. gingivalis enhances its survival by exploiting CR3, as pharmacological blockade of CR3 promotes its killing and suppresses Po. gingivalis-induced periodontal bone loss in a mouse model [66, 69] .
CR3 and CR4 were shown to be involved in the binding of Cryptococcus neoformans  and are dominant receptors in the uptake and killing of Candida albicans  . However, the fungal pathogen Histoplasma capsulatum evades antimicrobial defences and proliferates intracellularly in macrophages infected through CR3, CR4 and LFA-1. The host macrophages are destroyed by the multiplying yeast, and the released microbes are phagocytosed by other macrophages attracted to the infected site  . The major H. capsulatum ligand for CR3 on macrophages was identified as heat shock protein 60 (hsp60), whereas dendritic cells recognize it via a different ligand  . Similar to H. capsulatum, the related dimorphic fungal pathogen Blastomyces dermatitidis also expresses a CR3-interacting protein, BAD1 (blastomyces adhesin 1). BAD1 helps pathogen survival by binding via CR3 and CD14 that mediates its internalization and the suppression of TNF-α production of host cells  .
The fungal pathogen Aspergillus fumigatus avoids opsonization and phagocytosis by expressing proteases that degrade complement proteins and CR3  . As surface-bound factor H can enhance the antifungal activity via binding to CR3 and CR4, C. albicans avoid phagocytosis by releasing the secreted aspartic protease 2 (Sap2) to cleave both FH and the complement receptors CR3 and CR4 on macrophages  . Additionally, C. albicans expresses a CR3-like structure that mediates adhesion of the yeast to human endothelium [76-78] .
CR3 and CR4 also contribute to the phagocytosis of P. falciparum-infected erythrocytes  , and Leishmania ssp. are able to enter and survive in host macrophages in a CR3-mediated manner [80, 81] . Moreover, Leishmania is known to inhibit IL-12 production in macrophages  , which is also mediated by CR3, as signalling via CR3 by L. major reduces IL-12 production  . At the same time, dendritic cells were shown to take up L. major in a CR3-independent, FcγRI- and FcγRIII-mediated manner, which leads to a more effective antigen presentation, indicating that CR3-mediated uptake is likely to represent a ‘decoy’ mechanism for this pathogen  .
The complement receptor immunoglobulin [CRIg, also known as V-set and Ig domain-containing 4 (VSIG4) and Z39Ig] is a phagocytic receptor expressed on macrophage subpopulations. First, Helmy et al. proved the expression of CRIg in CD68 + Kupffer cells in the liver, interstitial macrophages in the heart, adrenal gland macrophages, alveolar macrophages, Hofbauer cells, synovial macrophages and lamina propria histiocytes by immunohistochemistry  . In dendritic cells, two groups showed the expression of CRIg, but only at the mRNA level [86, 87] . CRIg belongs to the immunoglobulin (Ig) superfamily. In humans, it has two splice variants. The long form (huCRIg(L)) contains a V- and a C2-type Ig domain, and a short form (huCRIg(S)) encodes only a V-type Ig domain  . The murine muCRIg receptor comprises only of a single V-type Ig domain  .
CRIg binds C3b and iC3b, providing the first line of defence in the liver and spleen by quickly eliminating opsonized pathogens. This receptor was shown to swiftly resurface through recycling endosomes after internalization, thus providing the means for a continuous phagocytosis  . In contrast to CR3 and CR4, which also bind iC3b-opsonized particles, CRIg is regarded to have an anti-inflammatory role as well. It is hypothesized that Kupffer cells internalize opsonized microbes and apoptotic cells first through CRIg, without the induction of inflammation [89, 90] . A higher number of microbial agents in the blood will prompt the engagement of other pattern recognition and complement receptors, leading to the initiation of an immune response including leucocyte recruitment and inflammation.
CRIg is additionally involved in the promotion of immunological tolerance through the inhibition of the alternative complement pathway convertases [91, 92] and the suppression of T-cell activation. However, the tolerogenic function of CRIg might support the progression of cancer with keeping T cells in an unresponsive state [93, 94] . Recently, the downregulation of CRIg in chronic Hepatitis B virus (HBV) infection was shown to lead to a poor prognosis in hepatocellular carcinoma patients probably due to reduced virus clearance [95, 96] .
Studies on the CRIg-mediated phagocytosis proved that this receptor is indispensable in the rapid internalization of complement opsonized Adenovirus particles  , S. aureus  , Li. monocytogenes  and C. albicans [97, 98] . The intracellular pathogen Li. monocytogenes survive inside macrophages by delaying phagosome maturation and escaping into the cytoplasm  . Kim et al. proved a multistep counter mechanism involving the engagement of CRIg. Signalling through CRIg facilitates phagosome acidification and fusion with lysosomes, enhancing the killing of internalized bacteria  . In addition, the ligation of CRIg with opsonized Li. monocytogenes or an agonistic mAb induces autophagosome formation, enabling macrophages to eliminate cytoplasmic bacteria already escaped from the phagolysosome system  .
Zeng et al.  proposed that the CRIg receptor is able to clear bacteria directly without opsonization, through the recognition of the gram-positive wall constituent, lipoteichoic acid (LTA). However, Broadley et al.  proved in a C3 knockout mouse, that Kupffer cells still internalize both gram-positive and gram-negative bacteria strains, but instead of CRIg, they use pattern recognition receptors, that is scavenger receptors for LTA. Further studies are required to clarify the individual participation of complement and pattern recognition receptors expressed by Kupffer cells, with consideration of the shear stress conditions present in the liver  .
Primary Centers of the Immune System
Although the immune system is characterized by circulating cells throughout the body, the regulation, maturation, and intercommunication of immune factors occur at specific sites. The blood circulates immune cells, proteins, and other factors through the body. Approximately 0.1 percent of all cells in the blood are leukocytes, which encompass monocytes (the precursor of macrophages) and lymphocytes. The majority of cells in the blood are erythrocytes (red blood cells). Lymph is a watery fluid that bathes tissues and organs with protective white blood cells and does not contain erythrocytes. Cells of the immune system can travel between the distinct lymphatic and blood circulatory systems, which are separated by interstitial space, by a process called extravasation (passing through to surrounding tissue).
The cells of the immune system originate from hematopoietic stem cells in the bone marrow. Cytokines stimulate these stem cells to differentiate into immune cells. B cell maturation occurs in the bone marrow, whereas naïve T cells transit from the bone marrow to the thymus for maturation. In the thymus, immature T cells that express TCRs complementary to self-antigens are destroyed. This process helps prevent autoimmune responses.
On maturation, T and B lymphocytes circulate to various destinations. Lymph nodes scattered throughout the body, as illustrated in Figure 12, house large populations of T and B cells, dendritic cells, and macrophages. Lymph gathers antigens as it drains from tissues. These antigens then are filtered through lymph nodes before the lymph is returned to circulation. APCs in the lymph nodes capture and process antigens and inform nearby lymphocytes about potential pathogens.
Figure 12. (a) Lymphatic vessels carry a clear fluid called lymph throughout the body. The liquid enters (b) lymph nodes through afferent vessels. Lymph nodes are filled with lymphocytes that purge infecting cells. The lymph then exits through efferent vessels. (credit: modification of work by NIH, NCI)
The spleen houses B and T cells, macrophages, dendritic cells, and NK cells. The spleen, shown in Figure 13, is the site where APCs that have trapped foreign particles in the blood can communicate with lymphocytes. Antibodies are synthesized and secreted by activated plasma cells in the spleen, and the spleen filters foreign substances and antibody-complexed pathogens from the blood. Functionally, the spleen is to the blood as lymph nodes are to the lymph.
Figure 13. The spleen is similar to a lymph node but is much larger and filters blood instead of lymph. Blood enters the spleen through arteries and exits through veins. The spleen contains two types of tissue: red pulp and white pulp. Red pulp consists of cavities that store blood. Within the red pulp, damaged red blood cells are removed and replaced by new ones. White pulp is rich in lymphocytes that remove antigen-coated bacteria from the blood. (credit: modification of work by NCI)
4 SIA SENSING BY SIGLECS
SIGLECs belong to the class of I-type lectins that are characterized by three major components: (a) a set of 1–16 immunoglobulin domains which determine how far the N-terminal head group extends from the plasma membrane, (b) an extracellular carbohydrate recognition domain at the N-terminus which may recognize Sia residues on the glycocalyx of other cells (trans) or on the same cell (cis), and (c) an intracellular signaling tail. For some SIGLECs, SIGLEC-14, -15, -16 in humans and SIGLEC-3 and SIGLEC-H in mouse, the intracellular signaling tail contains a positively charged amino acid residue that can associate with activating adaptor proteins such as TYROBP/DAP12 bearing an immunoreceptor tyrosine activation motif (ITAM Figure 4 Duan & Paulson, 2020 ). Upon ligand binding, such ITAM-containing adaptor proteins are recruited and become phosphorylated by a member of the spleen tyrosine kinase (SYK) family triggering a number of downstream signaling cascades similar as for other ITAM-SYK-signaling receptors like complement receptor 3 (CR3) or triggering receptor expressed on myeloid cells 2 (TREM2 as reviewed in Linnartz-Gerlach, Kopatz, & Neumann, 2014 ).
However, most SIGLEC receptors contain an intracellular immunoreceptor tyrosine-based inhibition motif (ITIM Figure 4), which upon Sia binding lead to inhibitory downstream signaling pathways. Besides conserved SIGLEC-2, these SIGLECs include SIGLEC-E, -F and -G in mice and SIGLEC-3, -5, -6, -7, -8, -9, -10, -11, and -12 in humans (Duan & Paulson, 2020 ). Upon ligand binding to extracellular Sia residues, these SIGLECs containing intracellular ITIMs can counteract activatory signals emanating from receptors containing ITAMs (Crocker, Paulson, & Varki, 2007 ). The counteraction is initiated via phosphorylation of ITIMs by Src family kinases after ligand binding. The tyrosine-phosphorylated ITIMs then recruit tyrosine phosphatases, such as Src homology region 2 domain-containing phosphatase-1 (SHP-1/PTPN6) or Src homology region 2 domain-containing phosphatase-2 (SHP-2/PTPN11) (as reviewed in Duan & Paulson, 2020 ), which may dephosphorylate signaling molecules in the ITAM-signaling cascade of associated activatory receptors to suppress the activation of immune cells (Linnartz, Wang, & Neumann, 2010 ).
The ITAM-SYK-signaling pathway of TREM2/DAP12 or CR3/DAP12 (Figure 4b) induces phagocytosis in mononuclear phagocytes such as microglia. Multiple studies have shown how ITIM signaling of inhibitory SIGLECS could modulate this pathway. One study showed that SIGLEC-5 inhibited macrophage-mediated phagocytosis of apoptotic bodies upon binding of sialooligosaccharide ligands or monoclonal antibodies (Rapoport, Sapot'ko, Pazynina, Bojenko, & Bovin, 2005 ). Another study discovered that SIGLEC-11 was able to prevent microglial phagocytosis of apoptotic neuronal material and to attenuate LPS-induced inflammation (Wang & Neumann, 2010 ). In addition mouse SIGLEC-E prevented the removal of neural debris and the phagocytosis-associated oxidative burst of microglia (Claude et al., 2013 ). Interestingly, most inhibitory SIGLECs act as endocytic receptors after binding of small sized sialoglycoconjugates, while they inhibit phagocytosis of larger structures (Duan & Paulson, 2020 ). Beside inhibition of phagocytosis, the inhibitory ITIM-containing SIGLECs may also inactivate the inflammasome via inhibiting SYK signaling triggered by ITAM-signaling receptors. It is reported that both c-Jun NH2-terminal protein kinase (JNK) and SYK are necessary kinases for inflammasome activation (Linnartz & Neumann, 2013 ). Toll-like receptor (TLR) signaling contributes to the JNK kinase activation, whereas ITAM receptors stimulate SYK kinase. Thus, inhibitory SIGLECs may also prevent inflammatory activation of mononuclear phagocytes. There is evidence in B-cells that ligand binding sequester inhibitory SIGLECs away from activatory receptors, thus increasing the signaling effect of the activatory receptors (Duan & Paulson, 2020 ). However, so far this phenomenon has not been observed for microglial cells.
SIGLECs are mainly expressed on immune cells, such as microglia, macrophages and monocytes with a few exceptions, such as for SIGLEC-4 (expressed in oligodendrocytes and Schwann cells) and SIGLEC-6 (also expressed in placental trophoblasts) as nicely summarized by Duan and Paulson ( 2020 ). All inhibitory SIGLECs expressed on microglia including mouse SIGLEC-2, human SIGLEC-3, SIGLEC-11, mouse SIGLEC-E, and SIGLEC-F have trans interaction with Sia residues on neighboring cells and several SIGLECs including SIGLEC-3 and SIGLEC-E have been demonstrated to also bind Sia residues in cis (Figure 4a Bornhöfft, Goldammer, Rebl, & Galuska, 2018 ).
The first functional studies on human SIGLEC-11 were carried out by ectopic expression of this human receptor on murine microglia (Wang & Neumann, 2010 ). The expression of SIGLEC-11 was able to inhibit the proinflammatory response of LPS-challenged microglia and prevented microglial phagocytosis of apoptotic neuronal material. This neuroprotective effect of SIGLEC-11 was mediated by trans interaction with polySia on neighboring neurons, but not by cis interaction with its own glycocalyx (Wang & Neumann, 2010 ). Similar neuroprotective effects were shown for SIGLEC-E (Claude et al., 2013 ). SIGLEC-E inhibited the phagocytosis of neural debris. Furthermore, SIGLEC-E on microglia prevented the neural debris-triggered release of superoxide and production of proinflammatory cytokines. Although on a low level cis interaction can be found for SIGLEC-E, our data indicated that the neuroprotective role of SIGLEC-E was mediated in trans via recognition of Sia residues on neurons (Claude et al., 2013 ). Interestingly, the release of the microglia-intrinsic polySia pool from Golgi upon inflammatory stimulation acted as a trans-activating ligand of SIGLEC-E, thus inhibiting the inflammatory response in traumatic brain injury (Thiesler et al., 2020 ). In contrast, CRISPR/Cas9-mediated Siglec-E knockout of the murine microglial cell line BV2 resulted in a more pronounced increase of LPS-induced proinflammatory cytokines on transcription level. Furthermore, Siglec-E knockout of BV2 prevented the anti-inflammatory effect of exogenously added polySia (Thiesler et al., 2020 ).
Both neurodegeneration and aging have been associated with dysfunction or deletion of SIGLEC receptors (Duan & Paulson, 2020 Schwarz et al., 2015 ). SIGLEC receptors on microglial cells interact with Sia to inhibit microglial activation, inflammation, phagocytosis and oxidative burst (Figure 4a). Siglec-E KO mice showed oxidative damage to cellular DNA, proteins and lipids in all organs that was related both to an unbalanced ROS metabolism, and to a secondary impairment in detoxification of reactive molecules (Schwarz et al., 2015 ). Consequently, Siglec-E KO mice exhibited accelerated aging and their life span was reduced (Schwarz et al., 2015 ). Interestingly, an opposite phenotype was observed after deletion of the crucial activatory microglial TREM2 receptor (Linnartz-Gerlach et al., 2018 ). Deletion of the main counter-balancing activatory TREM2 receptor in mice led to decreased age-related inflammatory signs and reduced neuronal loss in the substantia nigra and the hippocampus of 24 months old mice (Linnartz-Gerlach et al., 2018 ). The absence of age-related neurodegeneration in Trem2 KO mice was associated with decreased microglial numbers and less accumulation of oxidized lipids (Linnartz-Gerlach et al., 2018 ).
Meanwhile, it also has been shown that human SIGLEC-3 prevents microglial uptake of amyloid-β 42 (Aβ42) (Griciuc et al., 2013 ). Human SIGLEC-3, also known as CD33, is abundantly expressed on microglia. In contrast to other SIGLECs, SIGLEC-3 is capable to interact with Sia in both trans and cis. Two splice variants of human CD33 are relevant for AD: the full-length CD33M isoform and the shorter CD33m isoform, which lacks the exon 2-encoded Sia ligand-binding domain (Malik et al., 2013 ). Several genome-wide association studies (GWAS) indicated that the full length CD33M is a risk factor for AD, whereas a decreased AD risk was associated with the shorter CD33m isoform (Malik et al., 2013 ). Furthermore, increased expression of CD33M was detected in microglial cells of AD brains, further supporting the GWAS findings (Hollingworth et al., 2011 Naj et al., 2011 ). Interestingly, the phagocytosis of Aβ42 by microglia was inhibited with increasing human CD33M levels. Concomitantly, the Aβ pathology was attenuated in amyloid precursor protein (APP)-transgenic mice by inactivating CD33, and thus possibly increasing the uptake of Aβ42 (Bradshaw et al., 2013 Griciuc et al., 2013 ). These data suggest that any weakening of the Sia-SIGLEC-3 axis, either by desialylation or variants of SIGLEC-3/CD33m lacking the Sia-binding domain might increase microglial phagocytosis of Aβ42, and hence potentially could be beneficial in AD pathology (see Box 1). Importantly, the human and the mouse SIGLEC-3 share the same name but not the exact same function. While the human SIGLEC-3 is an inhibitory receptor with an ITIM domain, the mouse SIGLEC-3 has only an ITIM-like domain. Consequently, both receptors fulfill different functions as shown previously (Bhattacherjee et al., 2019 ). While the human SIGLEC-3 inhibits monocyte and microglial phagocytosis, no effect on phagocytosis was found for the mouse SIGLEC-3 (Bhattacherjee et al., 2019 ).
Most recently, it was reported that SIGLEC-2, also called CD22, downregulates microglial phagocytic ability during aging and SIGLEC-2 blockade promotes removal of α-synuclein fibrils, Aβ oligomers and myelin debris in vivo (Pluvinage et al., 2019 ). Furthermore, SIGLEC-2 was upregulated in microglia during aging (Pluvinage et al., 2019 ), and in the brains of AD patients (Friedman et al., 2018 ). By activating CD22 with synthetic glycopolymers bearing α2–6-linked Sia, it was shown that SIGLEC-2 on microglia inhibited phagocytosis (Pluvinage et al., 2019 ). Interestingly, inhibition of SIGLEC-2 with SIGLEC-2 blocking antibody or genetic ablation showed the opposite effect and increased the phagocytosis of Aβ oligomers, myelin debris and α-synuclein fibrils in vivo. Long-term blockade of SIGLEC-2 using blocking antibody of SIGLEC-2 via CNS-delivery was even able to restore microglial homeostasis and to attenuate the cognitive decline of aged mice (Pluvinage et al., 2019 ).
In this review we discussed the structure/function relationships of complement FH, with a focus on the emerging roles of this protein in the pathophysiology of infectious diseases, inflammatory conditions and cancer. FH is the major soluble inhibitor of the AP C3 convertase, an enzyme complex that is at the very center of the complement cascade. Therefore, alterations of FH protein structure, due to mutations and polymorphisms, likely lead to aberrant functional outcomes. This is the case for FH mutations associated with aHUS, AMD and DDD all diseases of local complement dysfunction where a loss of appropriate self-recognition, e.g. within the extracellular matrix, appears to be a common factor. Also, FH is a preferential target for complement evasion by a number of pathogens. This has paved the way towards complement vaccines, which are being designed and developed based on FH microbial ligands, and novel FH-based therapeutics. Finally, a dual role is emerging for FH in cancer, where it can either be hijacked by cancer cells to avoid the complement attack, or can be used to dampen cancer-related inflammation.
3 HOW MUCH INFORMATION?
3.1 The information in chemotactic communication systems
In communication theory, information is the “measure of one's freedom of choice when one selects a message”. 34 For example, each character in the Latin alphabet (26 characters) possesses much more information than each character of Morse code (3 characters: dot, dash, and space), because one has significantly more choice in how to express oneself when choosing from the larger set of Latin characters. 123 Indeed, it takes a sequence of dots and dashes to specify a single letter of the alphabet, and thus more dots and dashes than letters to specify the same word, showing how symbols vary in the amount of information they encode.
Like Morse code and the Latin alphabet, chemoattractant ligands encode different amounts of information to coordinate chemotaxis in bacteria, D. discoideum, and immune cells. In this section, we ask the following questions: How diverse are the sets of chemoattractant symbols used by these systems, and how might that change the amount of information each system is able to convey in the context of chemotaxis? How diverse are the sets of cell-surface receptors that receive and transmit the messages encoded in chemoattractants, and how might their diversity and type alter ability to transmit the full meaning of the encoded messages to the inside of the cell? How are the chemoattractant alphabets used by these systems strung together into sentences that convey more complex instructions? Finally, why do bacteria, D. discoideum, and immune cells use such different alphabets to accomplish the same task?
In this section, we address these questions with the help of 4 principles: (1) message alphabet size, (2) message complexity, (3) transmitter capacity, and (4) combinatorial complexity (Fig. 4). These principles will help inform the way we understand how different organisms encode information to direct complex chemotactic processes. In each subsection, we will highlight illustrative examples of how these principles shed light on cellular chemotaxis in different systems.
3.2 Understanding chemotactic information by message alphabet size
The idea of information as a function of ‘message alphabet size’ was developed by Ralph Hartley. Hartley developed an equation that defines the amount of information in a message as the base 2 logarithm of the number of symbols from which it was selected (i.e., the alphabet size). 34, 124 For instance, a symbol chosen at random from a set of 32 symbols carries 5 bits of information per symbol (number of possible symbols = 2 amount of information per symbol , or 32 = 2 5 ). With this equation, we see that larger alphabets contain more information per symbol (Fig. 4A). If we compare chemoattractant message alphabet sizes of E. coli, D. discoideum, and neutrophils, we find widespread variation in the information content among chemoattractant ligands or messages. For E. coli, whose alphabet contains ∼10 messages (see Section 2.2), each individual message is estimated to contain only ∼3 bits of information per ligand. Similarly, the 2 chemoattractant messages for D. discoideum (i.e., cAMP and folate) each carry only 1 bit of information. The chemoattractant message alphabet for neutrophils is by far the largest, with more than 100,000 estimated messages. 100 Each neutrophil chemoattractant message thus contains ∼17 bits of information, significantly more information than either E. coli or D. discoideum chemoattractant messages.
Chemoattractant message alphabet size can be defined in two other ways, the number of ligands belonging to a particular class (e.g., chemokines vs. formylated peptides, see Fig. 3) or the number of ligands binding a single chemoattractant receptor. Considering these definitions of message alphabet size, we find that E. coli and D. discoideum possess restrictive alphabets by all definitions. For example, E. coli MCP receptors recognize exclusively small molecule chemoeffectors and each MCP receptor typically interacts with only 1–2 ligands (though some MCP receptors have been reported to bind upward of 5). 36 D. discoideum principally respond to a single nutrient (i.e., folate) and a single communication signal (i.e., cAMP), both of which are small molecule chemoeffectors. The alphabet size of ligands shared by a single receptor is similarly limited, with cAMP and folate constituting the only canonical ligands binding the respective receptors. 52
Comparatively, mammalian neutrophils encode significantly larger chemoattractant alphabets. Formylated peptides, which act as bacterial pathogen signals, and chemokines, which act as intercellular communication signals, have alphabet sizes of >100,000 100 and ∼45 ligands, 103 respectively. In addition to these two large alphabets, mammalian neutrophils respond to two small alphabets: LTB4 and PAF, which encode general-use inflammation signals, and complement pathway products (e.g., C5a), which encode pathogen danger signals. 90 Considering the Hartley principle as above, in which symbols in large alphabets encode more information than those in small alphabets, formylated peptides are the most “information-rich” chemoattractant class for neutrophils, followed by chemokines, then LTB4, PAF, and complement, which are on par with E. coli and D. discoideum chemoattractant systems. This theoretical information hierarchy agrees with what is known biologically, as diverse bacterial pathogens encode highly specific, “information-rich” chemical signals as compared to broad-utility chemoattractant cues like leukotrienes and complement products. Chemokines, which have an intermediate information capacity according to Hartley's principle, are used broadly as a family, although individual ligand-receptor systems can be employed in highly specific contexts.
Using communication theory to compare chemoattractant systems among E. coli, D. discoideum, and neutrophils highlights the following question: Why do neutrophils need larger, more diverse alphabets than their eukaryotic D. discoideum counterparts, despite sharing much of the same receptor and intracellular machinery? The answer to this question may be understood by comparing the context-specificity afforded by small versus large chemoattractant alphabets. Large chemoattractant alphabets allow neutrophils to undergo chemotaxis to serve functionally diverse roles as a major arm of the innate immune system, whereas D. discoideum undergo chemotaxis in only two contexts. Thus, while both organisms utilize similar machinery, differences in message alphabet size between neutrophils and D. discoideum emphasize the difference in complexity between the two eukaryotic organisms, illustrating the need for a more complete understanding of organism- or cell type-specific (e.g., mammalian rather than broadly eukaryotic, neutrophil rather than macrophage) chemotaxis.
3.3 Understanding chemotactic information by message complexity
The amount of information encoded in a chemoattractant ligand can also be evaluated in terms of chemical complexity. To demonstrate this principle, we borrow a concept developed by Hann and colleagues, in which ligands and receptors are depicted by strings of “+” and “–“ signs (e.g., receptor: – – + – + – – + – ligand: + + –) (Fig. 4B). 125 In this model, binding events only occur when ligand and receptor contain strings of complementary signs, which represent the combination of interactions (e.g., shape, electrostatics, van der Waals, H bonding, etc.) that create binding specificity. Intuitively, the larger the ligand, the more requirements there are for it to make complementary interactions across a more distributed receptor surface. While the authors use this analysis to derive strategies for optimizing drug discovery efforts, it is useful to consider the effect of increasing ligand complexity in instances where nature has already provided specific ligand–receptor pairs.
The chemical complexity of chemoattractant messages for the mammalian immune system vary drastically by both composition (small molecules, modified peptides, lipids, and proteins) and size (∼330 Da for cAMP vs. 9,000–12,000 Da for chemokines). Comparatively, the composition and size of bacterial chemoattractant (or chemorepellant) ligands is more limited, with the best studied bacterial ligands (e.g., serine, aspartate, dipeptides, O2) < 500 Da (although it should be noted that some chemoattractants bind periplasmic binding proteins which then bind MCP receptors, see section 2.2). Though D. discoideum is a eukaryotic system, the canonical D. discoideum chemoattractants are much closer in size to prokaryotic (i.e., E. coli) chemoattractants than to those of mammalian neutrophils, with cAMP at ∼330 Da and folate at ∼440 Da. Thus, based on chemical complexity, mammalian neutrophil chemoattractants have the potential to encode much more information and conceivably more complex outcomes than either E.coli or D. discoideum. In effect, whereas E.coli and D. discoideum ligands would be represented by short strings of +’s and –‘s, neutrophil ligands would be represented, on average, by much longer strings of +’s and –‘s.
What are the functional implications of utilizing “information-rich,” complex ligands to organize chemotaxis? The comparatively complex structure of chemokines, for one, allow many of them to encode signaling specificity (i.e., biased signaling) at a given receptor, demonstrating how a more chemically complex message enables complex cellular outcomes. 106 Interestingly, despite their structural complexity, chemokines possess a highly conserved tertiary structure, which constrains their chemical diversity to some extent. Moreover, the high degree of chemokine structural similarity suggests that among chemokines, complexity is encoded by substituting residue identities on a structurally conserved scaffold. It should be mentioned that beyond receptor-interactions, the ligand complexity of chemokines allows them to encode numerous other functions, some of which are associated with chemotaxis, like GAG binding and homo-/hetero-dimerization, as well as non-chemotactic functions, such as antimicrobial activity. 126-128
Formylated peptides, which have lengths between 3 and 10 amino acids on average, demonstrate much less individual complexity than chemokines, but with >100,000 approximated members, they have the potential to encode significantly more information by way of combinatorial complexity. 100 Nevertheless, molecular modeling studies suggest that formylated peptides, like chemokines, share a conserved structure, despite divergent amino acid sequences. Bufe and colleagues demonstrated the formation of a common “clawlike” fold, stabilized by residues at only 3 peptide positions, which allows for considerable sequence diversity at other peptide residues. 100 Interestingly, substitution of peptide residues outside of 3 conserved positions had little effect on chemotaxis. 100 In the context of the Hann model, formylated peptide structural conservation and sequence constraints may functionally limit their complexity such that instead of being represented by >100,000 unique strings of +/– symbols, apparently different formylated peptides may in fact be redundant, such that a common string of +/– symbols may more appropriately represent their chemical complexity.
The chemical complexity framework highlights the following paradox: how do small, low-complexity chemoattractants, such as LTB4 maintain receptor specificity, while larger, high-complexity ligands, such as chemokines, demonstrate widespread promiscuity? Regarding the chemokine system, we speculate that chemokines sharing common receptors are more evolutionarily related to one another than those binding different receptors. Consequently, even though all chemokines would be represented by long strings of +/– symbols that should, in principle, be highly specific for an individual receptor, chemokines sharing a common receptor would have closely related +/- strings. While numerous structural studies of unique chemokine-receptor interactions have highlighted how chemokines recognize their receptors for specific ligand–receptor pairs, structures of different chemokines bound to the same receptor (and vice versa) and new bioinformatics approaches will be needed to understand how chemokines are able to be so promiscuous despite highly complex interactions with their receptor counterparts.
3.4 Understanding chemotactic information by transmitter capacity
In our discussion of information content of chemotactic systems, we must also consider how the chemoattractant message is encoded by the transmitter (i.e., receptor) (Fig. 4C). Transmitters may vary (1) in the ways in which they encode different messages, and (2) the extent to which they can encode the full complexity of the message. In other words, while complex (i.e., “information-rich”) ligands can, in principle, relay more information to the cytosol than can simple ligands (i.e., “information poor”), ligand complexity may be lost when the signal is passed through the receptor such that the output signal contains less information than the original ligand message. Thus, while message alphabet size and message complexity are significant when considering the information content of chemotactic systems, the transmitter of the chemotactic message (i.e., the chemoattractant receptor) plays a significant role as well. In this section, we summarize this significance using the concept of “transmitter capacity.”
Shannon discusses encoding as a process of “operating on a message” to change its format in a way that is “suitable for transmission over the channel,” for instance encoding words into Morse code or encoding speech into a digital signal to send over a telephone wire. 34 In chemotactic signal transduction, the receptor (e.g., MCP receptor or GPCR) encodes the message of the chemoattractant by undergoing conformational and dynamic changes in the ligand-bound state, and these changes are interpreted by cytosolic signaling molecules (e.g., CheA/W proteins in bacteria or G proteins/β-arrestins in eukaryotes) which relay the converted message to the interior of the cell. Transmitter capacity, or the ability of chemoattractant receptors to adequately decode the complexity of chemoattractant messages for transmission to the intracellular environment, is functionally constrained (1) at the structural level, in the extent to which a given receptor can interpret the differences of its numerous chemoattractant binding partners (i.e., receptor promiscuity), and (2) at the signaling level, in the extent to which the same receptor can initiate complex outcomes through any number of cytosolic effectors (i.e., biased signaling).
Considering first transmitter capacity at the structural level, bacteria and eukaryotes have evolved vastly different receptor systems to transmit information encoded in chemoattractant ligands to cytosolic effectors that can regulate chemotaxis. Comparing the structures used by prokaryotic and eukaryotic chemoattractant receptors, the E. coli MCP receptors are >3000 kDa (i.e., as trimer-of-dimers) and span ∼350 Å, whereas GPCRs are typically 40–50 kDa and span ∼ 50 Å. 129, 130 It is interesting to consider the size discrepancy of the two systems, relative to the sizes of their respective ligands. MCP receptors are on the order of 10 4 times larger than their ligands by mass, whereas GPCRs range from ∼1–10 times the mass of their ligands. Despite this size difference, receptor promiscuity, or the ability of a chemoattractant receptor to respond to more than one chemoattractant ligand, is common to both bacterial MCP receptors and eukaryotic GPCRs. However, the number of chemoattractant messages that a single receptor can transmit (i.e., the degree of receptor promiscuity) differs for each organism. The ∼10 chemotactic messages for E. coli are transmitted through only 5 MCP receptors, with each MCP receptor transmitting 1–2 messages (or 10–20% of the total), on average. In contrast, the 6 chemoattractant GPCRs (4 cAMP, 2 folate) for D. discoideum transmit only 2 chemotactic messages, with a given GPCR only transmitting 1 of the 2 messages (or 50% of the total). Mammalian neutrophils are considerably more complex than either E. coli or D. discoideum, with ∼30 chemoattractant GPCRs to interact with >100,000 potential messages (see Section 3.2). For mammalian neutrophils, however, it is perhaps more valuable to consider the transmitter fidelity for each chemoattractant class (e.g., chemokines vs formylated peptides) rather than the group together. For example, we say that mammalian neutrophils have >100,000 potential chemoattractant messages, but most of these are of the formylated peptide class which are transmitted through only 2 (or possibly 3) receptors (Fig. 3).
Further, the activation mechanisms of GPCRs and MCP receptors share both “piston” type movements (e.g., at TM2 in MCP receptors and at TM5 in GPCRs) as well as large scale dynamic alterations, although the structural differences between the two receptors make the dynamic changes very different. 129, 131, 132 In both cases, long clusters of alpha helices allow the receptor systems to amplify the small changes occurring at the ligand binding pocket into much larger changes intracellularly, although despite their increased length, conformational changes in MCP receptors are subtle and localized to the distal CheA and CheW contact sites. 133 Despite their distinctive size advantage, bacterial MCP receptors are much more limited than D. discoideum or mammalian neutrophil GPCRs in the amount of information they can transmit to the cytosol, and eventually, to the chemotaxis machinery. As discussed above, bacterial MCP receptors alter chemotaxis by modulating the rate of autophosphorlation of the cytosolic effector CheA, which directly interacts with the flagellar motor complex. While the rate of CheA autophosphorylation can be modulated by additional proteins (e.g., methyltransferase CheR and methylesterase CheB), the information encoded in individual chemoattractant ligands is limited to altering the function of a single effector in CheA.
Contrastingly, GPCRs are capable of eliciting activation of numerous cytosolic effectors, including but not limited to G proteins, G protein-coupled receptor kinases (GRKs), and β-arrestins. Moreover, GPCRs exhibit widespread biased signaling, in which different ligands preferentially activate a subset of the available effectors to elicit ligand-specific outcomes. 134 Biased signaling is widespread in the chemokine system, with different chemokine ligands eliciting unique signaling profiles at the same receptor. 106 While much less established, biased signaling may also play a role in the formylated peptide receptor system. 98, 99, 135-137 Although biased signaling has not been explicitly described in D. discoideum, D. discoideum nevertheless has numerous cytosolic effectors available to initiate complex signaling outcomes that are adaptable in different environmental circumstances. 138 In effect, eukaryotic chemoattractant receptor systems are better equipped to initiate diverse outcomes. In other words, the large, diverse, and complex chemoattractant messages used by eukaryotes are in many cases transformed into complex downstream outcomes. Comparatively, the diverse E. coli chemoattrantant messages are funneled into activation of a single cytosolic effector, thus limiting the extent to which diverse messages can encode diverse processes.
A close inspection of GPCR-chemoattractant interactions sheds some light on the ways in which eukaryotic signals are capable of eliciting such complex intracellular pathways. Recent structural studies of chemokine–receptor interactions demonstrate an enormous ligand–receptor interface, spanning as much as 1700 Å 2 . 139 Importantly, other recent studies suggest that differences in chemokine-receptor interactions far from the traditional receptor binding pocket influence how a single receptor can mediate alternative functional responses. 140, 141 In one example, CXCL12 provokes and arrests chemotaxis in a monomeric and dimeric form, respectively, and structures of CXCL12 bound to its receptor CXCR4 in both forms demonstrate mutually exclusive interactions that may influence CXCL12's opposing effects on chemotaxis. 141 Comparison of three chemokine-receptor crystal structures demonstrates diverse binding orientations, suggesting that in addition to making unique ligand–receptor contacts, ligand orientation may also contribute to signal transmission. 139, 142, 143 These and other examples demonstrate the ways in which eukaryotic chemoattractant receptors, which transmit messages encoded in generally large and diverse chemoattractants, are high “capacity” transmitters insofar as they convey the full extent of the encoded messages by activating diverse functional responses.
3.5 Understanding chemotactic information by combinatorial complexity
Bacterial and eukaryotic cells navigate through complex environments in which they integrate multiple chemoattractant messages to devise a resultant chemotactic route. In an analogous way, Shannon demonstrates how one can construct complex sentences resembling English by semi-randomly choosing letters from the alphabet with the help of transition probability tables, which define the frequency at which each letter follows all other letters. 34 In both Shannon's stochastic English sentence algorithm and in chemotactic organisms, individual symbols or messages are “strung together” to create more complex outcomes. We will discuss this principle in the context of chemoattractant-receptor systems as “combinatorial complexity” (Fig. 4D).