|Home | About | Journals | Submit | Contact Us | Français|
Mast cells are primitive immune cells that appear early in evolution and have since evolved into multifunctional cells in vertebrates.1 They have been long recognized as initiators of IgE-dependent allergic diseases but it is now realized that they also play a fundamental role in innate and adaptive immune responses to infection as well as inflammatory autoimmune diseases.2–4 In addition, there is evidence that mast cells participate in inflammatory responses to incipient tumors which may either facilitate or retard tumor growth depending on the type of cancer.5–7 Other suspected non-immunological roles for mast cells include promotion of angiogenesis, tissue remodeling, and wound healing.8–11 As such, mast cells occupy a precarious position in that their responses to endogenous and exogenous stimuli can be detrimental as well as beneficial to the host.
The scope of the physiologic and pathologic roles noted above illustrate the flexible nature of mast cells which is enabled in part by their ability to release a broad array of bioactive mediators, either preformed and stored in granules such as histamine and proteases, or newly generated such as eicosanoids, cytokines, and chemokines.12–14 Mast cells also exhibit phenotypic plasticity which is manifested by differential expression of receptors and granule constituents. The localization of mast cells in tissues as well as their replication and differentiation into distinct phenotypes are typically determined by the chemical environment of their final tissue destination.1, 13, 15 KIT ligand (stem cell factor or SCF) is critical for these processes although other growth and chemotactic factors are necessary. Flexibility in function is also conferred through expression of multiple types of receptors.14 These include Fc, cytokine, tyrosine kinase, and trimeric G protein-coupled receptors.12, 16 In addition, mast cells can respond to microbial products through pattern recognition receptors (PRR) which include the Toll-like receptors (TLRs) and the recently recognized nucleotide-binding oligomerization domain (NOD)-like receptor (NLRs).17 These two families of receptors recognize pathogen-associated molecular patterns (PAMPs) of microbial origin and include both cytosolic and surface-expressed receptors. Many of these receptors can act in concert with the IgE receptor, FcεRI, to substantially enhance mast cell responses to antigens or alter the pattern of response such that production of cytokines, for example, predominates over degranulation.18, 19 Moreover, mast cells can be negatively regulated through immunoreceptor tyrosine-based inhibitory motif (ITIM)-bearing receptors such as FcγRIIb, gp49B1, signal regulatory protein-α (SIRPα), the transforming growth factor-β (TGFβ) receptor, along with the non-ITIM-bearing receptor such as CD200.20, 21
All of the above features likely underlie the multifunctional nature of mast cells but little specific information is available as to how mast cells adapt in different immunological/pathological settings even though it is now evident that the relative composition and amount of mediators released from activated mast cells, and the degree to which they home to sites of inflammation, can be profoundly influenced by the extent to which surface receptors interact to engage alternative and complementary intracellular signaling pathways. Therefore, mast cell activation in health and diseases states should be considered in the context of co-stimulatory and inhibitory factors present in the surrounding tissue milieu and not just in terms of specific allergen-mediated processes. In the following sections we review studies that demonstrate potentiation or suppression of FcεRI-induced responses by activating or inhibitory receptors and the regulation of expression of these receptors by endogenous factors. Although most of these studies were conducted with mast cells in culture, they begin to provide insights as to how mast cell function might be modulated in situ. Our intent is to foster further interest in the interplay among mast cell receptors to create a sufficient body of evidence to provide mechanistic explanations for each of the immunological and non-immunological functions attributed to mast cells. Because of the impact on clinical practice, we also discuss receptor-independent mechanisms of mast cell activation by mast cell “secretagogues”, a category that includes opiates and other drugs in addition to antidotes to the biological effects of mast cell products. Finally, we assess the pathophysiologic roles of mast cells and their products in health and disease.
The acute reactions that occur as a result of mast cell activation are initiated as a consequence of degranulation and the generation of lipid-derived mediators, whereas more chronic mast cell-mediated symptomology is an outcome of the delayed generation of chemokines, cytokines and growth factors which follow enhanced gene expression.22, 23 The process of degranulation occurs within seconds of mast cell activation and the initial rapid phase is essentially complete within 5–10 minutes.24 Although mast cell proteases such as tryptase, chymase and carboxypeptidase, constitute the major components of mast cell granules, 25–27 histamine is the predominant granule mediator of acute reactions to mast cell activation.
Histamine is localized primarily, but not exclusively, in mast cells and basophils although its link to anaphylactic and inflammatory reactions was suspected long before its recognition as a major constituent of mast cells (reviewed in12). Histamine is sequestered in mast cell granules by proteoglycans such as heparin and chondroitin E. On release, it readily diffuses through tissues and the circulatory system but does not penetrate the CNS. It is rapidly inactivated by histamine N-methyl transferase and diamine oxidase. Histamine acts through at least four G protein-coupled histamine (H1 through H4) receptors (reviewed in28, 29). H1 receptors reside primarily on bronchial smooth muscle, endothelial cells, and certain neurons, and are largely responsible for bronchoconstriction, increased vascular permeability leading to hives and allergic rhinitis through separation of endothelial cells, itching, and pain. H2 receptors are located on vascular smooth muscle and gastric parietal cells and thus mediate vascular dilatation and gastric secretion. Collectively, the H1 and H2 receptors contribute to the wheal and flare reaction in skin. H3 receptors are present primarily in the CNS whereas H4 receptors are expressed on immunocompetent cells including basophils, mast cells, 30 and eosinophils31 and mediate chemotaxis. In total, the primary effects of histamine release from mast cells and basophils are increased vascular permeability, vasodilatation and bronchial constriction, which are readily reversed by antihistamines. The sedative side effect of classical antihistamines (H1 blockers) is now attributed to reversal of the normal CNS function of histamine in provoking wakefulness and was obviated by the introduction of antihistamines that do not penetrate the CNS. Antagonists of H3 and H4 receptors are under preclinical and clinical investigation as potential therapeutic agents for various CNS disorders but also for the treatment of allergic rhinitis, asthma, pruritis in atopic dermatitis, and inflammatory pain.31
Rodent mast cells also contain significant quantities of 5-hydroxytryptamine (5HT, serotonin) within the granules, which can induce inflammatory skin reactions in mice 32 and bronchoconstriction in rats.33 It appears to be a minor component in human mast cells.34 Whether the amounts present are sufficient to impact mast cell function in humans is unclear, but there are reports of effective therapy with the anti-serotonerigic agent, cyproheptadine, in patients with some forms of urticaria, 35, bullous lesions36 and irritable bowel syndrome, the latter being attributed to immunologically-induced release of serotonin from mast cells.37
Mast cells proteases constitute between 30–50% of the total protein content of mast cells; thus in terms of mass, they represent the major group of mediator released by exocytosis.25, 26 Although β-tryptase is the major protease expressed in mast cells, several other proteases and their respective homologs are also present in human and rodent mast cells.26 However, their expression in mast cells is heterogeneous depending largely on the on the phenotype and distribution of mast cells within the tissues. In the human for example, mast cells present in the mucosal tissues contain primarily tryptase, whereas those present in the skin and submucosa additionally contain chymase and carboxyeptidase.38 Within the granules, the proteases are complexed to proteoglycans which are thought to stabilize the proteases and to regulate their function.26 Although mast cell proteases, released from activated mast cells, have been implicated in allergic inflammation and tissue remodeling, 26 recent evidence suggests that they may also have a protective role for example against parasitic infection39, 40 and snake and insect venoms41, and indeed proteases may also help protect against allergic inflammation.
The bioactive eicosanoids, leukotriene C4 (LTC4), LTB4, prostaglandin D2 (PGD2) and, under specific circumstances, PGE2, are generated and released almost simultaneously with the granule-associated mediators. The initiating process in the generations of these molecules is receptor-mediated activation of phospholipase(PL) A2, with consequential hydrolyses of arachidonyl-containing phospholipids, primarily phosphatidylcholine, yielding free arachidonic acid.42 This is subsequently metabolized through the actions of 5-lipoxygenase and cyco-oxygenase to respectively generate LTs and PGs. These eicosanoids individually or together can profoundly affect processes associated with the allergic response for example, bronchoconstriction, increased vascular permeability, cellular infiltration and immune suppression.43 Reactive oxygen species, generated as byproducts of eicosanoid generation, have been implicated in mast-cell dependent inflammatory reactions.44
Mast cells generate a wide variety of both cytokines for example: IL-3, IL-4, IL-5, IL-6, IL-10, IL-13, IL-33, GM-CSF and TNFα; and chemokines including CCL2, CCL3, CCL5 and CXCL8.22, 23, 45–48 However, in contrast to degranulation and eicosanoid release, this is a delayed process taking several hours before significant levels of cytokines and/or chemokines are detected to be secreted from activated mast cells. Although it has been proposed that mast cells may store specific cytokines, for example TNFα, in their granules, at least in the human, this appears to be minimal compared to the levels generated de novo following mast cell activation.46 As will be discussed elsewhere in this review, under specific circumstances, for example, as part of an innate immune reaction, both cytokines and chemokines can be generated and released in the absence of degranulation. This suggests that cytokines and granule components have evolved to play different roles in the body’s defense mechanisms against, parasites, toxins and other harmful bioactive agents and organisms. Nevertheless, along with granule components, secreted cytokines manifest pathophysiology associated with allergic inflammation22, 23, 49–51 and may also contribute to angiogenesis and cellular hyperplasia associated with tumorigenesis.10
TSLP is now recognized as an important mediator in inflammatory responses to allergens, pathogens and trauma by directing the immune system towards Th2 responses.52, 53 Myeloid dendritic cells are thought to be a key target for TSLP although the possibility exists that mast cells play a similar role. The major sources of TSLP are epithelial cells at barrier surfaces, keratinocytes, dendritic, and stromal cells while its receptor (TSLPR) is expressed on a variety of immune cells including human mast cells, 54 eosinophils, macrophages, dendritic cells, B cells, and T cells.53 However, functional binding of TSLP to TSLPR requires cooperation of IL-7Rα, and in some respects TSLP shares functional similarities with IL-7 although both cytokines target different cells in human and mouse.52 Nevertheless, TSLP is produced in significant quantities in lungs of patients with asthma, 55 human primary small airway epithelial cells in response to TLR ligands, and skin explants from patients with atopic dermatitis.54 TSLP from all these sources potently stimulate human mast cells to produce inflammatory cytokines without inducing degranulation or eicosanoid production.54 Conversely, IL-4-primed human mast cells in culture produce and store TSLP on stimulation via FcεRI. Moreover, bronchial mast cells in atopic asthmatic patients accumulate TSLP and appear to be a significant source of TSLP in bronchial tissue in asthmatics.56 TSLP, so produced, is released spontaneously and following FcεRI aggregation but is rapidly degraded, in part by mast cell proteases, presumably limiting its actions to nearby cells.
Other evidence for a critical role for TSLP in atopic asthma is that the accumulation of TSLP in bronchial mast cells correlates with airway hypersensitivity in asthmatics56 and that TSLPR-deficient mice do not develop allergic airway disease.55, 57 An interconnecting link between TSLP and mast cells is the fact that, like dendritic cells, 58 mast cells can both respond and produce TSLP and serve as an additional source of TSLP under pathogenic conditions.
In keeping with their functional versatility, activated mast cells are also a source of angiogenic peptides such as angiopoietin-1, FGF, VEGF (see also Section VI D), and renin59 which, by promoting localized angiotensin formation in cardiac, lung, and kidney, can induce ischemia/reperfusion arrhythmia, 60 bronchoconstriction, 61 and renal disease.62 In addition, reactive oxygen and reactive nitrogen oxide species have been implicated in mast cell-related inflammatory conditions.44 Angiopoietin-1, FGF, and VEGF are all expressed in mast cells as is renin which in cardiac tissue is expressed exclusively in mast cells. In addition to renin, angiotensin-II is produced as a result of release of mast cell chymase and this mechanism may be more important than the renin-angiotensin system in the generation of angiotensin following mast cell activation.63
The role of the high affinity IgE receptor in mast cell activation, and the mechanisms by which this receptor regulates mast cell biology, has been extensively reviewed over the past few years. For this reason, we present a summary of these topics to provide a point of reference for later discussions. Readers are referred to the following selected reviews for more detailed information.16, 64–67
The FcεRI, which belongs to the immunoreceptor superfamily, comprises a single chain IgE-binding α subunit, a signal transducing/amplifying tetramembrane-spanning β subunit, and a signal-transducing γ chain homodimer subunit, 64 which is also responsible for relaying transmembrane signaling for the FcγRI and FcγRIII IgG receptors.68 The FcεRI allows mast cells to be activated in an antigen-specific manner following Th2 cytokine-driven production of antigen-specific IgE by B lymphocytes and subsequent binding of the IgE to the FcεRI. Cell activation is initiated upon binding of antigen to its specific recognition sites on the IgE resulting in receptor aggregation and thence downstream signaling.64 These signaling events are sufficient to ensure the generation and/or release of all three major classes of mast cells mediators16 and mast cell chemotaxis (Figure 1).69 However, FcεRI may also influence mast cell biology independently of a direct response to antigen.70, 71 In this regard, monomeric (non-aggregated) IgE has been described to promote mast cell survival, 72, 73 induce mast cell migration74 as well as the generation of the cytokines.73, 75 Unlike aggregated IgE, monomeric IgE failed to activate the necessary signaling pathways required to induce degranulation and cysteinyl leukotriene production in cultured human mast cells75 although modest degranulation and LTC4 production has been noted in freshly isolated human lung mast cells.76
The ability of antigen to induce the release of all major categories of inflammatory mediators from mast cells and to promote mast cell chemotaxis, requires the coordinated activation of sequential, parallel, and interacting signaling pathways that generate the divergent processes required for these multiple responses (Figure 2). The more receptor-proximal signaling events generally share common signaling elements whereas the more distal events show significant divergence. It is the divergence in these signaling pathways that may allow chemotaxis or release one category of mediators in the absence of others. Although the pathways regulating mast cell activation are complex they can be condensed into the following major signaling nodes or axes: 1. FcεRI β and γ chain phosphorylation leading to Syk recruitment/activation; 2. the LAT/ PLCγ1-calcium/protein kinase C (PKC) axis; 3. the phosphoinositide 3-Kinase (PI3K)-axis; 4. MAP kinase pathways and transcriptional regulation. These axes are discussed in subsequent sections below. Overall, such signaling can be additionally thought of in terms of a primary activation signaling pathway, which can either be upregulated by a secondary amplification/maintenance pathway, or downregulated by inhibitory signaling pathways.16 This concept could facilitate the understanding of how co-receptors that modify mast cell function can differentially exert their influence on FcεRI-mediated responses.
The progression of the signaling events following FcεRI aggregation requires initial phosphorylation of specific tyrosine residues in the immunoreceptor tyrosine-based activation motifs (ITAMs) contained within the β and γ chains of the FcεRI which respectively leads to recruitment of the Src family kinases Lyn77–79 and Fyn80 to the β chain and the ZAP 70 family member Syk to the γ chain domodimer.81, 82 Although Lyn has been suggested to be the tyrosine kinase responsible for this initial phosphorylation, there remains compelling reasons to suspect that other kinases may additionally or alternatively regulate this obligatory response. Namely, in human mast cells and, depending on the genetic background, mouse mast cells Lyn appears to be dispensable and even inhibitory for subsequent mast cell activation.83, 84 One explanation could be that Src kinase family members have overlapping functions in mast cell activation. Certainly, in addition to Lyn, other family members including Fyn, 80 Fgr, 85, 86 and Hck87 have been implicated in mast cell activation albeit by the regulation of different processes. Nevertheless, the recruitment of Syk to the phosphorylated γ chains and its subsequent activation is accepted to be an obligatory step for mast cell activation.
Recruitment of regulators and components of the subsequent signaling nodes, as defined above, into the receptor-signaling molecule signaling complex is coordinated via a series of constitutive and inducible protein-protein and lipid-protein interactions through a framework of transmembrane (e.g. CBP/PAG, LAT and LAT2) and cytosolic (e.g. SLP 76, Grb2, Gads Gab2, Vav, Shc) adaptor proteins.88, 89 It is thought that such interactions occur within discrete localized regions enriched in glycolipids, termed lipid rafts.90–93 Consequently, early signaling events associated with the FcεRI may be controlled as much by distribution and redistribution of the specific signaling elements within the membrane micro-domains as by posttranslational modification.
The preferential engagement of specific adaptor molecules by individual surface receptors and their capacity to selectively recruit particular signaling elements may underlie the ability of different receptors to independently regulate the release of the individual classes of mediators. The transmembrane adaptor molecule LAT is a major substrate for Syk and a central coordinator of downstream events required for mast cell activation following FcεRI aggregation. LAT contains multiple tyrosines within the cytosolic tail that, once phosphorylated by Syk and potentially other kinases, recruits PLCγ1 and the cytosolic adaptor molecules SLP76 and indirectly Gads, which stabilize this complex. The importance of these interactions is illustrated by the markedly attenuated capacity of antigen to elicit a calcium response, degranulation, and the generation of cytokines in mast cells derived from the bone marrow (BMMCs) of lat−/−, 94–96 slp76−/−, 97, 98 and gads/−99 mice.
PLCγ exists as two isoforms. PLCγ1 and PLCγ2, both of which are expressed and activated upon FcεRI aggregation in mast cells.100–102 However, at least in human mast cells, PLCγ1 may be the predominant form.103 PLCγ activation leads to the hydrolysis of the membrane-associated phospholipid, phosphoinositide 4, 5 bisphosphate (PIP2) thus liberating inositol trisphosphate (IP3) and diacylglyerol (DAG).104 These products respectively liberate calcium from the endoplasmic reticulum, and activate specific PKC isoforms, critical signals for mast cell activation.105, 106 The IP3-dependent release of calcium from the intracellular stores results in depletion of these stores which is detected by the calcium sensor STIM1 which is critically located in the ER membrane.107–109 The subsequent conformational change in STIM1 and its interaction with the calcium transporter ORAI1110–112 on the cell membrane allows calcium influx from extracellular stores. Such influx in mast cells may also, in part, be regulated by members of the TRPC family of calcium channels.113, 114 The exact contribution of this family of proteins to the calcium signal, however, is still a matter of debate. The calcium signal is essential for a variety of downstream signaling process regulating not only degranulation, but also for generation of other mediators and chemotaxis.115, 116 For example, calcium is required for the activation of the calcium-dependent forms of PKC, a family of serine/throineine kinases critical for the generation of all classes of mediators.117–121 Calcium is also necessary for the activation of PLA2 required for the generation of arachidonic acid leading to eicosanoid generation and for activation of the calcium-regulated transcription factor NFAT which contributes to cytokine generation.115
Although the activation of PLCγ is primarily regulated via the aforementioned process, the maintenance and amplification of this signal is regulated by the PI3K/Btk axis.16, 24, 122 PI3K is recruited to the signaling complex following the Fyn-dependent binding of PI3K to Gab2.123 It then promotes membrane localization of specific signaling molecules through the generation of phosphatidylinositol 3, 4, 5 trisphosphate which binds to pleckstrin homology (PH)-domains contained within these signaling molecules.124 In this manner, the Tec kinase, Btk is recruited to the cell membrane where it is activated; a response which further increases PLCγ1 activity and its capacity to reinforce the calcium signal.24, 69, 124, 125 Although the initial phase of the calcium signal mediated by PLCγ1 is unaltered by inhibition of PI3K, the maintenance of this signal is significantly reduced following inhibition of PI3K either by the expression of an inactive form of the enzyme122 or through the use of PI3K inhibitors103 and in btk−/− BMMCs.24, 69, 126, 127 The maintenance /amplification of the PLCγ1 signal following FcεRI aggregation may be coordinated additionally by the LAT-related transmembrane adaptor molecule LAT2 (NTAL, LAB)128, 129 although LAT2 may also regulate an inhibitory pathway for FcεRI-mediated mast cell degranulation, at least in mouse BMMCs.95, 96 Certainly LAT2 knock down in human mast cells130 and the RBL 2H3 rat mast cell line128 attenuates FcεRI-mediated degranulation through reduction of the calcium signal. Our studies suggested that LAT2 may be contributing to the PLCγ1-mediated calcium signal by indirectly binding PLCγ1.129 Subsequent studies suggested that this interaction may be occurring through SLP76.131 As further discussed later, this maintenance/amplification pathway for antigen-mediated responses in mast cells may be the portal by which signaling pathways initiated by other mast cell receptors allow synergistic potentiation of FcεRI-mediated degranulation and cytokine production.
The experiments utilizing mouse mast cells expressing catalytically inactive PI3Kδ and those using PI3K inhibitors further revealed that PI3K regulates not only mast cell degranulation but also generation of reactive oxygen species (ROS), eicosanoid and cytokine production, cell growth and survival, cell adhesion and chemotaxis.122, 124, 126 Central to the ability of PI3K to regulate such diverse processes is its pivotal role in the regulation of multiple signal transduction pathways. Although several of the responses mediated by PI3K, for example degranulation and the generation of ROS and eicosanoids, may to a certain extent be explained by downstream activation of Btk24, 126 other PI3K-regulated pathways certainly contribute to mast cell homeostasis and function. For example, the ability of PI3K to regulate cell growth, survival, and transcriptional regulation may be partly explained by its regulation of the activities of the mammalian target of rapamycin complex (mTORC) 1 and 2132, 133 and GSK3β134 through the activation of AKT. Other pathways controlling transcriptional regulation in a PI3K-dependent manner undoubtedly also contribute to cytokine and chemokine generation. In this respect, the antigen-dependent activation of the transcription factors NFAT and NFκB are both dependent on the PI3K/Btk axis as demonstrated in the btk−/− BMMCs. Similarly, the ability of co-activating receptors to enhance transcription activity leading to synergistically enhanced antigen-mediated cytokine generation, requires engagement of the PI3K/Btk axis.24 For more in depth discussions on the role(s) of PI3K in mast cell function readers are referred to a recent review article on this topic.124
Receptor-mediated generation of cytokines, chemokines, growth factors and other proteins in mast cells requires activation of the mitogen-activated protein (MAP) kinase pathways and enhanced transcription through the activation of specific transcription factors that interact with the promoter regions of the encoding genes. All three major MAP kinase pathways (ERK, JNK, and p38 MAP kinase) are activated via the Ras-Raf pathway in antigen-stimulated cells as discussed in detail in a previous review.66 The transcription factors activated include fos and c-jun which are components of the dimeric API complex, 19, 125, 135 NFAT, 18, 19, 136–141 NFκB, 19, 142, 143 ATF2, 135 and ELK1.137 Both the Ras-Raf-MAP kinase pathway and the PI3K-regulated signaling network contribute to the activation of these factors. These signaling cascades are critical to the generation of, for example, cytokines and chemokines.16 In addition, both PKC and calcium, as regulated by the LAT/ PLCγ1-calcium/ PKC axis, can contribute to transcriptional regulation, independently of MAP kinases and PI3K.16 It is clear however, that the signaling processes leading to transcriptional regulation can function in isolation from those required for degranulation.18, 19 For example, as discussed below, certain mast cell stimulants induce cytokine generation in the absence of a detectable degranulation response. Furthermore, the signal transduction events induced by these stimuli can interact with those initiated through FcεRI to markedly enhance cytokine production without enhancing degranulation.18, 19
In addition to the high affinity for IgE receptor, FcεRI, multiple other receptors for IgG are expressed on mast cells.45, 46 Such expression, however, may be dependent on the cytokine content of the surrounding milieu. Under the appropriate condition, mast cells express the FcγRI, FcγRIb and FcγRIII IgG receptors.45, 46, 68, 144 Both the FcγRI and FcγRIII consist of IgG-binding α subunits and the γ chain homodimer which is identical to the FcεRI γ subunits. The FcγRIα subunit binds IgG with high affinity (KA~10−8 – 10−9 M) whereas the FcγRIIIα subunit binds IgG with relatively low affinity (Ka~ 10−7 M).68 By virtue of the ITAM motifs contained within the γ chain cytosolic tails, FcγRI and FcγRIII have the capacity to activate mast cells is a similar manner to that elicited by the FcεRI.
Although multiple FcγRI transcripts have been recorded, only one full length protein (FcγRIα1) is expressed that binds IgG and that can elicit signaling.45 Under resting conditions, the FcγRI is not expressed on rodent or human mast cells. However, in CD34+ peripheral blood-derived human mast cells, exposure to IFN-γ results in upregulation of FcγRI mRNA and the receptor is observed on the cell surface.45, 46, 144 Constitutive FcγRI expression is observed, however, in the LAD2 human mast cell line.145 Furthermore, FcγRI is present in mast cells in psoriatic skin where IFN-γ levels are elevated, 146 implying that mast cell FcγRI expression is associated with specific disease states. The FcγRI thus expressed on human mast cells has been shown to be functional in that FcγRI aggregation resulted in degranulation and cytokine production in a similar manner to that observed following FcεRI aggregation.46 The notable difference being the markedly higher production of TNFα observed following FcγRI aggregation compared to the relative lack of TNFα produced upon FcεRI aggregation.46 The reason for this disparity is not entirely clear but may be due to differential requirements of FcεRI, and FcγR for PI3K in their activation, 147 and/or the influence of the β chain which is part of the FcεRI but not the FcγRI.
As would be expected with the FcεRI and FcγRI sharing common γ subunits, there appears to be little additive or synergistic enhancement of mast cell activation when both receptors are simultaneously, but independently, aggregated. Nevertheless signals produced upon FcγRI aggregation synergistically interact with those induced by the C3a GPCR to enhance mast cell degranulation.148
The FcγRIII, although expressed in mouse BMMCs following SCF treatment, does not appear to be expressed to the cell surface in human mast cells.45 In the mouse however, as with the FcγRI, it can induce degranulation, and generation of eicosanoids following aggregation.149–152 Furthermore, FcγRIII can induce mast cell adhesion to fibronectin.153
In contrast to the FcγRI and FcγRIII, the FcγRIIβ receptor is a single chain receptor which is not associated with the common signaling γ chain homodimer. Therefore the FcγRIIβ does not possess the capacity to induce mast cell activation. However, due to the Immunoreceptor tyrosine-based inhibitory motif (ITIM) contained within the cytosolic tail, the FcγRIIb, when co-ligated with the FcεRI, it results in the down-regulation of antigen-induced degranulation.154 The inhibitory properties of FcγRIIb is a consequence of recruitment of the Src homology 2 domain-containing inositol phosphatase (SHIP) following phosphorylation of the tyrosine residue in the FcγRIIb ITIM upon ligation and reversal of PI3K-mediated signaling.154 This inhibitory function of the FcγRIIb, and other ITIM-containing inhibitory receptors expressed on mast cells, has led to the consideration of approaches to down-regulate mast cell activation through the engagement of these receptors. Studies demonstrating that a fusion protein, which coligates the FcγRI and FcγRIIb, produces inhibition of antigen-mediated human mast cell and basophil activation, provides proof of principle for this approach.155
KIT is a member of the growth factor receptors with inherent tyrosine kinase activity family. It is a single chain receptor with 5 extracellular immunoglobulin-like domains and a split tyrosine kinase catalytic domain in the cytosolic tail. Also contained within the cytosolic tail are a number of tyrosine residues which are auto/transphosphorylation sites156–158 or may additionally be target substrates for Src family member tyrosine kinases.159 KIT activation is essential for mast cell growth, differentiation and survival156, 157 and which can furthermore induce mast cell migration/homing through chemotaxis.158 KIT catalytic activity and downstream signaling is initiated upon dimerization induced by binding of its specific ligand, stem cell factor (SCF).160, 161 Thus, mast cell numbers, homeostasis and function in resident tissues are to a large extent dictated by localized concentrations of SCF.162 SCF is presented in two major forms, a soluble form and a membrane-bound form, and is primarily expressed by fibroblasts but is also generated by mast cells and other tissues.163–166 There is some evidence that soluble and membrane bound forms of SCF may differentially influence KIT-mediated signaling in mast cells.157, 167 Whether this has functional consequences in a physiological setting however is unclear.
In addition to its effects on mast cell homeostasis and chemotaxis, under experimental conditions, SCF can substantially modulate mast cell degranulation and cytokine production.168–172 Nevertheless, the consensus of studies suggests that the signals generated by activated KIT alone are insufficient to induce degranulation but they can induce cytokine production in the absence of other stimulants (Figure 1).130, 172 The synergistic enhancement of antigen-mediated degranulation and cytokine production is seen in both mouse and human mast cells, and at least in the human mast cells, can be observed with concentrations of antigen which produce minimal degranulation in the absence of SCF130 Such synergy is not limited to antigen-dependent mediator release, however, as SCF has been shown to synergize with adenosine A2b receptor agonists to synergistically enhance IL-8 production in the HMC-1 human mast cell line173 and LPS (Toll-like receptor (TLR4) agonist) to synergistically enhance PGD2 and cytokine production in BMMCs.174 Furthermore, mast cell degranulation can be promoted by SCF in the presence of monomeric IgE, neither of which induces degranulation on their own.76 Additionally, we have demonstrated that SCF and antigen, synergistically interact to markedly enhance mast cell chemotaxis.69
The signaling pathways elicited by KIT share many common features with those induced by the FcεRI even though KIT lacks the ability to activate critical signals required for the ability to induce mast cell degranulation (Figure 2).172 Although KIT activates Src kinases, PLCγ1, PI3K and the MAPK cascade, 24, 130, 172 it does not appear to be able to recruit and activate Syk, phosphorylate LAT130 or activate PKC172 at least to the same extent as that produced by the FcεRI. In addition, the calcium signal is much reduced and delayed compared to that observed upon FcεRI aggregation.130, 172 These absent or weak signals would account for the inability of KIT to induce a degranulation response. However, when these signals are supplied by aggregation of FcεRI, KIT then has the capacity to enhance degranulation and cytokine production through reinforcement or potentiation of FcεRI signals.
In mouse BMMCs24 and especially in human mast cells, 172 simulataneous addition of SCF and antigen induces a marked potentiation and prolongation of the calcium signal produced by antigen. This could be directly explained by a similar enhancement and extended duration of Btk and PLCγ1 activations under these conditions.24 Although a similar enhancement in PI3K activity was not observed, 24 both PI3K and Btk were determined to be essential for the synergistic enhancement of antigen-mediated degranulation and cytokine production by SCF as demonstrated by use of BMMCs expressing a catalytically inactive p110δ subunit of PI3K, PI3K inhibitors, 122 and btk−/− BMMCs.24 Similar approaches have revealed that the PI3K/Btk/PLCγ1 axis also regulates the enhanced cytoskeletal reorganization responsible for the synergistic enhancement of chemotaxis in response to antigen and SCF.69 Whether this axis can sufficiently explain the enhanced cytokine release is less clear. However, synergy is observed in the activation of the transcription factors NFAT and NFΚB which could account for the enhanced cytokine expression is markedly reduced in btk−/− BMMCs24. In addition, the calcium-dependent activation of PKC-β appears to be necessary for the synergistic production of IL-6 in response to SCF and antigen in BMMCs.175
The amplification by SCF of the calcium signal and degranulation in antigen-stimulated human mast cells appears to require coordination through LAT2. Accordingly, siRNA-targeting of LAT2 inhibits such amplification.130 We observed that KIT as well as Lyn and Syk, when activated via FcεRI, can phosphorylate LAT2 but at entirely different residues. Although the residue(s) phosphorylated by Syk were shown to be responsible for the indirect binding of PLCγ1, 129 it remains to be determined how the residues phosphorylated by KIT enhance the recruitment of PLCγ1 and the PI3K/Btk/PLCγ1-mediated calcium signal.
Whether the amplification noted with SCF occurs in vivo, especially in disease states, is an obvious question and one that has yet to be addressed. Given the role of SCF in homeostatic regulation of mast cells, it is to be expected that mast cells in vivo would be in constant contact with either the soluble or membrane-associated SCF. However, this chronic exposure has not always been considered in studies examining interactions between SCF and other mast cell stimulants. The studies described above have largely employed short term (30 minutes in the case of degranulation) exposure of SCF in cultures which have been starved for 4–16 h to down-regulate constitutive KIT activation. It is thus highly likely that the manifestations of chronic SCF exposure on mast cell activation would be markedly different from those observed and described to date. Nevertheless there are conflicting data as what may be the outcome for such exposure in an in vivo setting. The myeloproliferative disorder, systemic mastocytosis, is associated with an activating (D816V) mutation in KIT, and an increase incidence of anaphylaxis has been reported in individuals experiencing this disorder.176 This would argue that KIT may indeed potentiate FcεRI-mediated mast cell degranulation in vivo. In contrast, it has been reported that repeated subcutaneous injections of SCF in mice reduces the incidence of fatal antigen-induced anaphylaxis which would argue the contrary.177 Further studies are therefore required to determine the precise influence that SCF plays on antigen-mediated mast cell responses both in health and disease states.
The discovery that the mammalian homologs of Drosophilia TLRs are pathogen-recognizing receptors and that they are expressed on mast cells indicated that mast cells could directly participate in host defense. The primary response to TLR ligands is the production of inflammatory cytokines rather than degranulation (Figure 3A). The initial reports from 2001 to 2003 focused on TLR2 and TLR4178–183 but, since then, other functional TLRs have also been discovered on mast cells (reviewed in184). Each of these receptors recognizes a distinct category of microbial products such as peptidoglycans (PGN) by TLR2, dsRNA by TLR3, lipopolysaccharides (LPS) by TLR4, and bacterial DNA and CpG-containing DNA by TLR9. Some TLRs rely on co-receptors that have no signaling capabilities by themselves. Examples include the interaction of TLR4 with CD14 which is essential for binding of lipopolysaccharide to a TLR4/CD14 complex and TLR2 with dectin-1 which acts as a co-receptor for the detection of fungal zymosan.185
Expression of TLRs may vary among different subsets of mast cells and culture conditions.184 For example, mast cells derived from human cord blood lack TLR4 under certain culture conditions but then express functional TLR4 on exposure to IFN-γ or IL-4.182, 183 The inducibility of TLRs and other pathogen-recognition receptors in mast cells is a topic, however, in need of a more systematic investigation.
TLRs belong to a superfamily of receptors consisting of two subgroups; one, the TLRs; and the other, receptors for IL-1, IL-18, and an orphan receptor ST2 which is now known to selectively recognize IL-33 (see next section). Both subgroups contain a common intracellular Toll/interleukin-1 receptor (TIR) domain but they differ in their extracellular ligand-binding domains; the TLRs possessing leucine-rich helical domains that recognize pathogen molecular motifs and the cytokine-receptor subgroup having Ig-like domains. Many of the TLRs, when bound by the appropriate ligands, interact with the adaptor protein, myeloid differentiation factor 88 (My88) and IL-1 receptor–associated kinases (IRAKs). Formation of this complex leads to recruitment of TNF receptor-associated factor 6 (TRAF6) and, in turn, TGFβ-activated kinase1 (TAK1). TAK1 activates the nuclear factor κB (NFκB), JNK, and p38 mitogen-activated protein (MAP) kinase pathways (Figure 3B). The latter pathways are also activated via FcεRI but by different mechanisms. As discussed in the following section, TLR ligands, especially TLR2 and TLR4 ligands, interact synergistically with antigen to substantially enhance cytokine production with minimal effect on degranulation (Figure 3A).
IL-33 is known to mediate Th2 immune responses and to activate mast cells.186–190 IL-33 accumulates in the affected tissues of patients with Crohn’s disease, rheumatoid arthritis, atopic dermatitis, psoriasis, and anaphylactic reactions and is thought to play a role in these diseases which are associated with heightened mast cell activity.191–196 In animal models, IL-33 is reported to exacerbate autoantibody-induced arthritis with enhanced autoantibody-mediated mast cell degranulation in synovial tissues197 and to be essential for the late phase inflammatory reaction during passive cutaneous anaphylaxis.198
The IL-33 receptor, ST2 is expressed on mast cells and is a member of the Toll/interleukin-1 receptor (TIR) family of receptors199 but acts in conjunction with the IL-1 receptor accessory protein (IL-1AcP).200–202 ST2, in common with IL-1/IL-18 receptors and TLRs, contains a TIR domain which enables recruitment of MyD88 to the ST2/IL-1AcP receptor complex and, in turn, IRAK1, IRAK4, and subsequently TRAF6.203, 204 As with TLR2 and TLR4 ligands, the engagement of TRAF6 leads to the activation of TAK1, the MAP kinases and key transcription factors including NF-κB through dissociation of the inhibitor of the NF-κB complex (IKK) from this complex (Figure 3B).19, 186, 188 IL-33-mediated signaling189 and cytokine production19 are MyD88-dependent as demonstrated in MyD88-deficient mouse BMMCs. Although some details of ST2-mediated signaling have not been unequivocally established in mast cells, they appear to share the same general features as those for other TIR family members including TLRs205–207 with key roles for the NF-κB, p38 MAP kinase, and JNK pathways (Figure 3B).
IL-33 is reported to stimulate cytokine production, but not degranulation or eicosanoid production in mast cells187–190 and basophils, 208–211 and it enhances FcεRI and G protein-coupled receptor mediated cytokine production in mast cells and basophils.188, 212 However, there are discrepancies in the literature in regard to IL-33 and degranulation. Our finding that IL-33 is unable to stimulate degranulation19 is consistent with the findings of other workers187–190 with the exception of Melendez and coworkers who reported that IL-33 induces an increase in [Ca2+]i and degranulation in mast cells.192 We find, however, that RBL-2H3 cells, BMMC, and cultured human mast cells exhibit no such responses to IL-33 under the exact conditions described by these workers.19
IL-3319, like TLR2 and TLR4 ligands18, in combination with Ag can markedly enhance production of the inflammatory cytokines in mast cells even at low subthreshold concentrations. Synergy in signaling was apparent at the level of TAK1 and gene transcription. The synergistic pulse appears to propagate from TAK1 through the JNK/c-Jun and p38 MAP kinase/ATF-2 pathways as depicted in Figure 3B. IL-33 and the TLR ligands by themselves are unable to initiate signals that are essential for mobilizing Ca2+ and activation of NFAT. However, the activation of NFAT by Ag or thapsigargin provides a strong signal for enhancing cytokine production through co-operative interactions of NFAT with other transcription factors such as AP-1.19 The resulting amplification of cytokine production could lead to increased severity of mast cell-related diseases.
The NOD-like receptors (NLRs) are of primary importance in responding to bacterial infections such as Heliobacter pylori and Listeria monocytogenes and are associated with Crohn’s disease, asthma, and other inflammatory diseases (reviewed in213). They are expressed largely in epithelial cells and antigen presenting cells (APCs) but have recently reported to exist in mast cells.214 In contrast to TLRs, NLRs are cytosolic receptors but, like TLRs, they contain leucine-rich repeat domains that recognize microbial molecular motifs probably in the form of breakdown products rather than native microbial components because of the cytosolic location of NLRs. NLRs can be segregated into several subfamilies on the basis of their signaling domains which engage different signaling molecules to produce signaling patterns that differ from those produced through activation of TLRs.
The major categories of NLRs, namely the NOD, NALP, IPAF, and NAIP proteins, were identified as pathogen pattern recognition receptors (PPRs) when it was apparent that TLRs would not account for protection against all microbes. NOD1 and NOD2 recognize different subunits of PGN and their activation results in activation of NFκB and JNK with production of inflammatory cytokines215 As the downstream signaling pathways recruited by TLRs and NLRs and the spectrum of microbial products sensed by these two types of receptors differ, it was inferred that they acted independently of one another. Nevertheless, recent studies in dendritic cells and mast cells suggest that these two types of receptors can interact in a positive or negative manner.
In the case of human cord blood-derived mast cells, these cells express message for NOD2 and NALP3 both of which recognize the muramyl dipeptide fragment of PGN and its synthetic analog, murabutide.214 Neither compound stimulated production of any of a wide range cytokines tested whereas the TLR2 agonists, intact PGN from Staphylococcus aureus and the synthetic Pam3Cysteinyl lipopeptide, stimulated production of IL-1β, IL-6, IL-10, and CXCL8. However, muramyl dipeptide and murabutide potentiated the production of IL-6 by Pam3Cysteiny lipopeptide without affecting production of any of the other cytokines that are produced in response to the lipopeptide (Figure 3A). The selective potentiation of IL-6 production was attributed to a unique requirement for the cAMP response element-binding protein (CREB) in addition to NFκB216, both of which appear to be activated via TLR2.214 Although IL-6 production by the TLR agonists, alone or in combination with TLR agonists, was suppressed in a dose dependent manner by an adenylyl cyclase inhibitor, the signaling interactions that lead to the potentiation were not defined. Neither TLR nor NLR ligands stimulated degranulation, although TLR ligands stimulated LTC4 production which was unaffected by co-stimulation with NLR agonists. This one study adds to the expanding list of mast cell regulatory receptors and provides yet another example how mast cell responses can be selectively regulated by a wide range of endogenous and microbial factors.
Adenosine receptors (AR) are of four subtypes (A1, A2A, A2B, and A3) that couple to different G proteins. A1R and A3R act through Gi or Go and, to a lesser extent, Gq to inhibit adenylate cyclase but additionally can activate PLC and downstream signals. In contrast, A2AR and A2BR couple to Gs thus activating adenylate cyclase although A2BR, but they can also engage Gq to activate the PLC pathway. However, signaling properties may be modified by formation of AR homo- or hetero-dimers. Selective agonists and antagonists have been developed for all four subtypes which are either in clinical use or undergoing preclinical or clinical evaluation (reviewed in 217–219). In addition to this array of ligands, evaluation of the biological roles of adenosine, and thus suitability of ARs as therapeutic targets, is further aided by the development of single and double knockout mouse models for all AR subtypes as well as adenosine deaminase, an enzyme responsible for rapid inactivation of adenosine.220 Virtually all cells, including mast cells, express one or more AR subtypes and they are generally cytoprotective in tissues under stress, although adenosine has manifold effects on vascular, muscular, and immune systems depending on the AR subtype involved.218 Intracellular adenosine levels increase as a result of metabolic stress or can be generated extracellularly as a result of the action of a plasma membrane 5’-nucleotidase on AMP in addition to export of intracellular adenosine through a bi-directional nucleoside transporter. Increases typically occur in tissues during ischemia, tissue injury, and inflammation associated with allergic and autoimmune disorders.220 In these conditions, adenosine is poised between being either anti- or pro-inflammatory.
The role of adenosine in allergic/autoimmune diseases, especially asthma and rheumatoid arthritis, has been of long-standing interest because of elevated production of adenosine in afflicted tissues and increased levels in related fluids such as bronchoalveolar lavage fluid in asthmatics (see reviews221–223). Adenosine is believed to exacerbate the underlying disease as, for example in inflammatory lung disease. Indications of this, among many, are that diagnostic administration of AMP, as the source of adenosine, induces bronchoconstriction in asthmatics, but not normal subjects, and that genetically-induced adenosine-deficiency in mice elicits an inflammatory phenotype reminiscent of asthma, with elevated IgE, extensive mast cell degranulation, eosinophilia, and impaired airway dynamics (reviewed in224), Of relevance to this review, adenosine markedly influences antigen-induced mediator release from human and rodent mast cells in vitro and in vivo.
Human and rodent mast cells express A2A, A2B and A3 receptors. The presumption is that selective stimulation of each type of receptor will elicit responses in accordance with the G protein to which the subtype can couple. Therefore, actions of adenosine itself may vary with concentration because of the different affinities of these receptors for adenosine. Early reports indicated that adenosine potentiates mediator release from human lung mast cells225 and mast cells in brochioalveolar lavage fluid from allergen-challenged asthmatic patients.226 However, inhibitory and potentiatory effects of adenosine on IgE-dependent degranulation were also noted in mechanically-dispersed human lung mast cells.227 Subsequent studies revealed either synergistic or inhibitory effects of adenosine or its analogs on mediator release in cultured or mechanicallydispersed human mast cells and the HMC-1 mast cell line. The heterogeneity of mast cells from different sources and previous lack of specific AR ligands were impediments to defining the exact roles of individual ARs. Nevertheless, the preponderance of evidence pointed to adenosine acting through high affinity A2A receptors at physiologically low concentrations thus activating adenylate cyclase and dampening mast cell activation. At high, perhaps pathological, concentrations, adenosine was thought to activate the low affinity A2BR which can couple PLC pathway via Gq (for more detailed discussion see221).
Interpretation of studies in rodent systems has been equally vexing. Overall, studies indicate that adenosine can initiate or enhance of mediator release from rodent primary and tumor mast cell lines (Figure 1).228–230 In one report, potentiation was attributed to A2BR rather than A2AR while other reports seemed to indicate that the A3R is a prominent locus of adenosine action on mast cells.228, 231, 232 A preeminent role for the A3R on mouse lung mast cells in vitro and in vivo was directly verified by use of selective AR ligands and A3R-deficient mice.230 Collectively, these studies point to the A3R as the predominant receptor although A2BR may also contribute to the activation of rodent mast cells. It should be noted, however, that human lung mast cells do not appear to express significant amounts of A3R whereas human eosinophils do so and may be a key site of action of adensosine in asthmatic subjects.233 In contrast, in the mouse, A3Rs are expressed on lung mast cells and eosinophils and both cell types may be equally important in the contribution to pathogenesis of the asthmatic phenotype in mice.234 Studies such as these illustrate the pitfall in extrapolating results from one species to another or even from cell culture where the pattern of expression of AR subtypes are altered by exposure to the Th2 cytokines, IL-4 and IL-13, 235 or glucocorticoids.229, 236, 237
AR agonists by themselves are generally weak stimulants of mast cell degranulation but they can substantially upregulate expression of cytokines primarily IL-1β, IL-3, IL-4, IL-8, and IL-13 in HMC-1 cells (Figure 1), which appeared to promote IgE production when co-cultured with B cells in an A2B-dependent manner.238 Adenosine and related agonists elicit little, if any, degranulation of RBL-2H3 cells and mouse BMMCs which correlates with the relatively weak and transient calcium signal after such stimulation.228, 239 The most pronounced effect of adenosine is to enhance mast cell activation in conjunction with other stimulants including antigen, ionophore, and phorbol ester.229, 240
Prostaglandin E2 (PGE2) is an eicosanoid, generated by multiple cell types including fibroblasts, macrophages, dendritic cells, epithelial cells and mast cells located in inflamed tissues.241–243 Thus, PGE2 has the capacity to modulate the responses of mast cells in such tissues. Multiple GPCRs for PGE2 are described and at least two of these receptor subtypes, the EP2 receptor and the EP3 receptor, are expressed in mast cells.244–246 The EP2 receptor is coupled to Gs which is linked to adenylate cyclase and which provides the capacity to down regulate antigen-mediated mast cell degranulation and cytokine production.245, 247 In contrast, the EP3 receptor, which in mast cells enhances mast cell-mediated responses (Figure 1) is linked to Gi, Gq and Gs. Nevertheless, responses elicited by the EP3 receptor in mast cells are predominantly elicited by Gi as indicated by their reversal by the ADP-ribosylating agent pertussis toxin.248 Depending on the species and circumstance, both EP2 and EP3 can modulate mast cell responses. Therefore, the influence of PGE2 on antigen-mediated responses in mast cells represents a balance between inhibitory signals produced by the EP2 receptor and activation signals produced by the EP3 receptor. The ability of PGE2 to modulate mast cell activation is remarkably similar to that observed with SCF, in that, at least in mouse mast cells, PGE2 alone fails to induce degranulation but markedly enhances antigen-mediated degranulation, 244, 246, 248–250 cytokine production, 244, 246, 248, 249 and chemotaxis.133 However the ways in which the signals are integrated to produce these synergistic responses are distinctly different. This is likely a consequence of the different modes by which PGE2 and KIT initiate their respective signaling cascades: Gi-GPCR vs tyrosine kinase activation. There is some disagreement, however, about how Gi-linked receptors may enhance antigen-mediated degranulation and cytokine production. By investigating the ability of adenosine to enhance antigen-mediated responses in PI3Kγ knockout mice and mast cells derived from the bone marrow of these mice, the conclusion was made that GPGR-enhanced antigen-mediated degranulation required PI3K activation.239 However, in our subsequent studies, 248 we determined that although PI3K certainly contributed to the FcεRI-mediated component of the response to co-stimulation of cells with antigen and PGE2, it was not the primary contributor to the synergistic interaction between these two stimulants. Instead, the synergy between the signaling processes, and hence degranulation, was determined to be a consequence of cross synergy between PLCγ1 activated downstream of the FcεRI and PLCβ activated downstream of the EP3 receptor resulting in a marked enhancement of the calcium signals Figure 2).
Mast cells migrate into sites of inflammation and, based on the documented presence of PGE2 and other mediators at these sites, we recently investigated whether GPCR agonists and SCF synergistically acted with antigen to induce enhanced mast cell chemotaxis.251 In sensitized BMMCs, antigen, adenosine, PGE2 and SCF, when added alone, all induced mast cell chemotaxis although the antigen response was noticeably lower than those produced by the other agents. Nevertheless, the chemotactic response was markedly augmented when antigen was added in combination with PGE2, adenosine or SCF. This synergy was found to be mediated by a PI3K-Btk dependent pathway leading to enhanced Rac-mediated F-actin rearrangement (Figure 4). This mechanism, at least for the PGE2, differed from that required for the synergistic enhancement of degranulation which was both PI3K and Btk independent.
In contrast to the synergistic responses elicited by PGE2 and antigen in mouse mast cells, the synergistic effects in human mast cells are not apparent in all subjects. In this respect, the ability of PGE2 to enhance antigen-mediated degranulation and cytokine production was observed to be highly donor dependent with synergy being observed in mast cells generated from the peripheral blood of only 50% of the donors examined.246 This was not attributable to differential expression of the inhibitory EP2 receptor, which is known to down-regulate antigen-mediated responses in human mast cells, or to differential expression of the EP3 receptor. Nor was the ability of the EP2 receptor to generate cAMP or the EP3 receptor to activate critical signaling processes impaired in the respective PGE2-responding and non-responding cells. However, in the nonresponding cells, PGE2 lacked the ability to enhance the antigen-mediated membrane translocation of PLCγ1 and subsequent calcium response for reasons that are still unclear. Human mast cells also differ from rodent mast cells in other respects. We have yet to observe chemotaxis of cultured human mast cells in response to PGE2 or indeed adenosine and antigen, either added alone or concurrently. Whether such differences occur in vivo has not been determined.
Sphingosine1-phosphate (S1P) is a potent lipid-derived mediator formed by the phosphorylation of sphingosine as a result of activation of sphingosine kinases (SphK) 1 and 2.252 SphK1 and 2 are both activated in mast cells253, 254 by a Fyn-regulated pathway upon FcεRI aggregation.254 However, the respective roles of the individual SPKs in mast cell function is still a matter of debate.255 Nevertheless, It is probable that S1P is generated and released from mast cells as a consequence of FcεRI-dependent SPK activation.253, 256 Although S1P may function as an intracellular signaling molecule in mast cells, 257, 258 SIP also appears to regulate mast cell activation through binding to cell surface GPCRs.253, 256
Five GP-coupled SIP receptors are described (S1P1–5), however mast cells only appear to express two of these receptors SIP1 and SIP2.256, 259 SIP1 and S1P2 receptors have quite diverse roles in mast cell function. Studies from gene-deletion and/or knock out approaches in mouse BMMCs, RBL 2H3 cells and human mast cells have suggested that whereas the SIP1 receptor induces mast cell chemotaxis, 253 the SIP2 receptor induces mast cell degranulation, and the release of chemokine and cytokines.253 The respective sensitivities of these responses to pertussis toxin suggest that SIP1-induced chemotaxis may be mediated via Gi whereas S1P2-dependent mediator release may be mediated via Gq or G12.260 Higher concentrations of S1P are required to elicit mediator release than those required for the chemotaxis response and these higher concentrations also down-regulate the chemotaxis response.253 From this it was inferred that localized SIP concentration gradients in vivo would initially promote mast cell migration to tissues where S1P concentrations are high. Once there, the high SIP concentrations would inhibit further mast cell chemotaxis but would promote degranulation.253
The precise role of the S1P2 receptor in mast cell activation is still not entirely clear. Although data from SIP2-deficent BMMCs suggests that S1P released from FcεRI-activated mast cells acts through S1P2 to induce degranulation, 253 direct addition of S1P to both BMMCs and RBL 2H3 cells has a limited effect on degranulation.253, 254 However, other reports suggest that SIP added together with antigen produces at least an additive degranulation response in BMMCs.260 Furthermore, in human mast cells it has been reported that S1P does induce substantial degranulation260 and that degranulation, as well as cytokine production in these cells, could be inhibited by the addition of an SIP2 antagonist or by siRNA knock down of SIP2 expression.255 Antigen-mediated passive systemic anaphylaxis reactions are reported to be similarly reduced in mice injected with an SP2 antagonist and in SP2-knockout mice when compared to wild type controls.261
The above reports suggest a role for SIP in the regulation of mast cell-driven reactions in a physiological setting. Nevertheless, the contribution of S1P to disease states remains to be established. There are certainly reports of elevated localized levels of SIP in asthma. However, the S1P may originate from multiple sources in addition to mast cells.262 Evidence has suggested that whereas SIP released from mast cells is largely responsible for mast cell chemotaxis via S1P1, SIP from other sources may be responsible for promotion of degranulation mediated by S1P2.262 SIP can also act on other tissues that contribute to the anaphylactic response. Indeed, recent reports suggest that SIP may actually promote recovery from anaphylactic reactions by acting on the vasculature.263 Further studies are thus necessary to define the role of S1P in mast cell-driven disorders especially in the human disease.
In addition to PGE2 and SIP, agonists of other GPCRs are reported to possess potent mast cell chemotactic properties. These agents include serotonin (5-hydoxytryptamine, 5-HT), the anaphylatoxin peptide (C3a), fractalkine (CX3CL1), MIP-1α (CCL2), eotaxin (CCL11), MCP-1 (CCL3) and RANTES (CCL5) (as reviewed in264). Although 5-HT, 265 CX3CL1, 266 CCL11, 267 and CCL2268 do not appear to influence degranulation, C3a, 269–271 CCL3, 239 and CCL5239 can regulate mast cell degranulation and cytokine production. C3a elicits its specific cellular responses via the Gi-linked C3a receptor and, in both human and rodent mast cells, C3a induces both degranulation and cytokine production and can act in an additive manner with antigen to further enhance degranulation.68, 148 Similarly, CCL3, 239 probably via the CCR1 receptor, and CCL5, probably via the CCR1, CCR3, or CCR5 receptors, are reported to potentiate antigen-mediated mast cell degranulation.264 As is true for adenosine, this potentiation appears to require activation of PI3Kγ.239
There is evidence for biological interactions between neurons and mast cells. Mast cells are found in close proximity to neurons and can be activated by neuropeptides.272 The best studied of these neuropeptides is substance P but vasoactive intestinal peptide (VIP), neuropeptide Y, and the calcitonin gene-related peptide (CGRP) have also been shown to activate at least some types of mast cells.273 To what extent these peptides activate the GαI subunit of the trimeric Gα protein directly274 or indirectly through GPCRs is not entirely clear. The activation of GαI, whether direct or indirect, is indicated by the suppression mast cell activation by pertussis toxin in the instances where this has been tested.
Substance P is released from nerve terminals of sensory nerves in the CNS and peripheral tissues and is associated with inflammation and pain (neurogenic inflammation) in response to infection and injury. The endogenous receptor for substance P, the GPCR neurokinin-1 receptor (NK1R), is expressed in human CD34+-derived cultured mast cells and LAD2 cells275 and is thought to be expressed in rat peritoneal mast cells on the basis of studies with NK1R ligands.276–278 Human mast cells and mast cell lines, and rat peritoneal mast cells, degranulate276, 279, 280 as well as produce cytokines275, 281 in response to substance P. However, studies with chemically modified versions of substance P suggest that activation of rat peritoneal mast cells was not dependent on NK1R but was an inherent property of the polybasic N-terminal domain of substance P, 277 possibly permitting direct activation of G proteins.274 Similar studies suggest that substance P can also act independently of NK1R in human mast cell lines while acting in part through NK1R.275 As with other polybasic mast cell-secretagogues that are thought to directly activate Gi proteins274 (see also Section 7 below), the secretory activity of substance P is suppressed in pertussis toxin-treated cells.280 Nevertheless, mast cells appear to express functional NK1Rs in certain circumstances. Peritoneal mast cells from a strain of Wistar rats are reported to secrete histamine when exposed to nanomolar concentrations of substance P and this response is blocked by a NK1R antagonist.278 NK1Rs are inducible in BMMCs by co-culture with SCF and IL-4 and, in this circumstance, respond to physiologically relevant concentrations of substance P with secretion of histamine which is inhibited a NK1R antagonist.282
Other histamine-releasing neuropeptides include VIP which can activate isolated human skin mast cells283 as well as human CD34+-derived and LAD2275 mast cell lines, neuropeptide Y which activates rat peritoneal mast cells, 284 and CGRP in human skin mast cells283 as well as mouse BMMC284. In the latter study, degranulation of BMMC in vitro in response to CGRP, substance P or compound 48/80 (see also section 7 below) was associated with a transient increase in [Ca2+]I and, in the presence of CGRPcontaining neurons, activation of these neurons. Bidirectional communication between mast cells and CGRP-containing neurons was proposed on the basis of these and previous findings of increases in densities of mucosal-type mast cells and primary CGRP-containing afferent nerve fibers, often in close apposition, in the ileum of schistosomiasis-infected mice285 and references cited therein). Bidirectional interactions have also been noted in co-cultures between RBL-2H3 cells and neurons286 via substance P and NK1R287 as indicated by increases in [Ca2+]I in both cell types.
The notion of a functional relationship between mast cells and neuropeptides is further supported by reports of elevations of VIP, substance P and somatostatin in patients with urticaria pigmentosa and systemic mastocytosis288–290 and a correlation between plasma levels of substance P, VIP, CGRP, and somatostatin and mast cell load and mast cell expression of NK1R in patients with cutaneous and systemic mastocytosis.291 Apart from neuropeptides themselves contributing to the multiple symptoms associated with mastocytosis, the neuropeptides may promote mast cell recruitment and mast cell-driven inflammation and thus have impact at multiple levels.
Neuropeptides such as substance P, VIP, and CGRP exhibit variable degrees of antimicrobial activity towards a disparate range of microorganisms 292 a topic that is discussed in more detail in the next section. In some instances, the antimicrobial activities were enhanced by heparin which might fit a scenario whereby release of neuropeptides induces release of mediators, including heparin, and thereby cause inflammation while suppressing further microbial infiltration. Such a scenario, however, remains theoretical and requires experimental validation.
Antimicrobial peptides evolved as part of an innate host-defense mechanism in primitive invertebrate species and have retained this function throughout evolution. The major categories include short amphipathic α-helical peptides (e.g., maganins and cathelicidins such as LL-37), cationic-rich peptides (e.g., indoleicidins and histatins), anionic (e.g., dermicidins), and cysteine-stabilized β-pleated sheets consisting of anionic with cationic peptides (e.g., α- and β-defensins). These peptides are expressed largely in epithelial cells but also in keratinocytes, dendritic cells and in immune cells such as neutrophils, intestinal Paneth cells, NK cells, and B cells. Human and murine mast cells express cathelicidins293 which appear to protect against skin infection.294 Mast cells are also reported to form extracellular DNA-chromatin webs, with cathelicidin LL-37 attached, that trap and kill microbes.295 In addition to their direct microbicidal activity, because of their ability to form membrane pores, they induce cytolysis, and in some instances suppress essential microbial intracellular cellular functions (reviewed in296), the antimicrobial peptides also act indirectly by augmenting innate and adaptive immune responses by regulating function of the host’s immune cells. An important component of these actions is the induction of inflammatory reactions by activating mast cells and thus recruitment of other immune cells273 as described elsewhere in this article.
Many of the antimicrobial peptides are thought to act through GPCRs because their actions on host immune cells are blocked on treatment with pertussis toxin but the identity of these receptors in mast cells is largely unknown.273 However, it is postulated from studies with signaling inhibitors that the human α and β defensins stimulate mast cell chemotaxis through separate GPCRs with each family acting through a common GPCR which is subject to PKC-independent homologous desensitization.297, 298 In addition to chemotaxis, the defensins and cathelicidin LL-37 induce degranulation as well as production of PGD2, 299–302 PGE2, and LTC4.303 These peptides also stimulate production of Th1, Th2, puritogenic, and inflammatory cytokines including IL-2, IL-4, IL-5, IL1β, TNFγ,302 IL-6, IL-31, and GM-CSF.303 The β-defensins and LL-37 induce phosphorylation of the MAP kinases p38 and ERK.301, 304 As with the identity of receptors, little else is known about the signaling processes involved except from studies with inhibitors that point to dependence on Gi/PLC, PI3-kinase, and PKC pathways for the actions of β-defensins and LL-37.300, 301, 303, 305, 306 The antimicrobial neuropeptide, catestatin, was shown recently to activate mast cells in a similar manner as β-defensins and LL-37 to induce chemotaxis and degranulation as well as production of eicosanoids and inflammatory cytokines.307
A diverse group of cationic compounds, sometimes referred to as “mast cell secretagogues”, were known to cause degranulation of mast cells long before the discovery of IgE-mediated activation of mast cells. Classic examples include the polymeric compound 48/80, the polyamines spermine and spermidine, various kinins and cationic peptide hormones, hymenoptera venom constituents such as mastoparan, mellitin, and mast cell degranulating peptide.274, 308 They all induced degranulation within seconds in only some categories of mast cells (e.g. serosal but not mucosal mast cells in rodents) and act through pertussis toxin-sensitive heterotrimeric Gi proteins. In addition to degranulation, compound 48/80 in common with substance P and VIP peptides induces production of IL-3 IL-8, TNFα, and GMCSF but not IL-4, IFNγ, and eotaxin in the human LAD2 mast cell line.275 Interestingly compound 48/80 and these peptides also induced substantial production of several chemokines and thus differed from IgE-mediated stimulation which did not.
It was originally proposed that the “secretagogues have the ability to penetrate the plasma membrane, via either α-helices of peptides or aromatic rings of compound 48/80, and directly activate Gi proteins in a receptor independent manner.308 Such a mechanism, however, would not be applicable to all known basic “secretagogues” and the concept has been since extended to include additional mechanisms for their membrane translocation and access to Gi such as low affinity non-selective membrane receptors or transporters on serosal mast cells.309. Also, some, but by no means all, basic peptides have since been shown to act through their own GPCRs309 and recently the action of compound 48/80 was reported to be dependent on the presence of neuronal GPCRs known as Mas-related genes (Mrgs), specifically MrgX1 and MrgX2.271 More studies are clearly needed to clarify mechanisms whether they involve direct activation of Gi proteins, GPCRs, or both.
Of clinical concern is that certain categories of cationic drugs also possess “secretagogue” activity and can directly activate human man mast cells in vitro and in vivo. These include opiates, neuromuscular blocking agents (e.g. tubocurarine), antimicrobial agents (e.g. pentamidine and noemycin), local anesthetics (e.g. lidocaine), and antimalarials (e,g, quinidine).310 Allergic- or anaphylactic –like (often referred to as “anaphylactoid”) symptoms are manifested within hours or minutes on first exposure to the drug, presumably through non-immunological mechanisms. Although relatively rare, such reactions can be mild or life threatening depending on route of administration.
Current information on the signaling systems activated by compound 48/80 and other classic “secretagogues” is incomplete and somewhat incoherent. A seminal observation was that compound 48/80 induced rapid stimulation of a pertussis toxin sensitive membrane GTPase activity (identified as Ni and later renamed Gi), transient production of inositol polyphosphates, and Ca2+-influx in rat peritoneal mast cells.311 This influx coincides with release of histamine, arachidonic acid, 311, 312 and PGD2.313 Subsequent studies indicated direct interaction of compound 48/80 with the COOH-terminal end of the α-subunit of Gi3 at the plasma membrane314 and initiation of additional signals via PKC and PI3K.315 The latter signals appeared to play no role in degranulation but were essential for production of arachidonic acid.315 The proposed model was that 48/80 activates Gi3, which suffices to induce degranulation, and a PKC/PI3K/tyrosine pathway that leads to activation of activation of PLA2 with the associated production of arachidonic acid metabolites such as PGD2. The activation of PKC and PI3K is also evident in 48/80-stimulated NIH3T3 cells except that in this model PI3K is thought to be primary effector with downstream activation of PKCα as well as PKB and calcium mobilization.316 In addition, compound 48/80317 and mastoparan318 can activate PLD, either directly or via Gi3 (or Gi2), which theoretically could result in activation of conventional PKC isoforms in conjunction with elevated cytosolic Ca2+ levels.319 Apart from the question of whether compound 48/80 activates multiple independent pathways or not, it is important to note that compound 48/80 and similarly acting “secretagogues” act on other cell types (e.g. NIH3T3 cells). Therefore, their actions in vivo do not necessarily implicate mast cells exclusively.
In addition to the specific mast cell receptors that upregulate antigen-induced responses, such responses can be down-regulated by a variety of other receptors expressed on the surface of mast cells.20, 21, 68 These include the FcγRIIb IgG receptor, 154 Sialic acid-binding, Ig-like lectins (Siglecs), 320 Mast cell function-associated antigen (MAFA), 321 Platelet endothelial cell adhesion molecule (PECAM-1), 322 Gp49B1, 323 paired Ig-like receptor-B (PIR-B), 324 CD200 receptor, 325 CD300a, 326 and CD72.327 Although these receptors are quite diverse in nature, with the exception of CD200 receptor, they all contain at least one, and up to four, immunoreceptor tyrosine-based inhibitory motifs (ITIM).21 ITIMs are consensus motifs (I/V/L/SxYxxL/V, where×is any amino acid) that are targets for Src kinases activated upon receptor ligation.21 Following phosphorylation of the tyrosine residues contained in these sequences, they recruit inhibitory phosphatases and other signaling proteins which can then impact responses elicited by both FcεRI and KIT. Indeed a fusion protein colligating both the FcγRIIb and FcεRI has proven efficacious in preventing activation of human mast cells and basophils.155 Both the tyrosine phosphatases, SHP1 and 2, and the inositol phosphatase SHIP are recruited to the phosphorylated ITAM.21, 68 However, there appears to be a degree of selectivity in the recruitment of these phosphatases to specific receptors. For example, the FcγRIIb recruits SHIP whereas other ITIM-containing inhibitory receptors such as Siglecs, Gp49B1, LAIR-1 and PECAM-1 primarily recruit SHPs. By virtue of their ability to dephosphorylate tyrosine residues of signaling proteins, SHP1 and 2 terminate signaling processes regulated by tyrosine kinases including Lyn, Fyn, Btk and Syk. SHIP on the other hand dephosphorylates inositol 3, 4, 5-trisphosphate at the D5 position thereby terminating signaling processes regulated by PI3K.124 These molecules, therefore, effectively reverse both of the major pathways regulating mast cell activation downstream from FcεRI and KIT.
Until recently, the natural ligands for many of these inhibitory receptors were unknown. However, recent studies have determined that some of these ligands are widely distributed in blood and tissues, for example IgG/immune complexes for the FcγRIIb and MHC-1 molecules for PIR-B.21 This may provide a clue as to why so many inhibitory receptors have evolved to downregulate the activation of mast cells and other cell types. Mast cell degranulation must be highly regulated to prevent inappropriate consequences for the host organism. This would be especially true during migration of precursors and immature mast cells from the bone marrow and during the maturation process before the mature mast cells take up residence in their target tissues. Engagement of the inhibitory receptors during the early steps of mast cell development and migration through encounters with commonly expressed molecules would thus ensure mast cells are maintained in an inactivated but primed state until required for immune defense in the appropriate tissues. Disruption of these inhibitory processes through inactivating mutations would likely have the potential to significantly impact disease.21, 271
Mast cells exhibit phenotypic variability depending upon the type of tissue in which they reside.13 In rodents, mast cells have been broadly categorized into connective tissue-type (also known as serosal) and mucosal-type mast cells on the basis of their distinctive histochemical staining characteristics which reflect differences in the glycosoaminoglycan component of the proteoglycan core of mast cell granules. The granules of serosal mast cells exhibit metachromatic staining with toluidine blue (i.e. a shift from navy blue to purple) and stain red with safranin due to the presence of heparin, a highly sulfated glycosoaminoglycan, In contrast, mucosal mast cell granules contain a less highly sulfated chondroitin E as the predominant core component and stain blue after sequential staining with alcian blue and safranin.328–330 Serosal mast cells are found largely in skin, connective tissue, and tissue spaces such as the peritoneum and for this reason were referred to as connective tissue mast cells. Mucosal mast cells, as implied, are found in mucosal tissue of intestine and airways.
Human mast cells were initially characterized according to differences in the expression of tryptase and chymase in mast cell granules. One type, called MCT, contains tryptase only and is the predominant subtype in respiratory and intestinal mucosa often in association with T cells. Another, MCTC, contain both tryptase and chymase in addition to other proteases such as carboxypeptidase A and cathepsin G and is predominant in skin, synovium, conjunctiva, lymph nodes, myocardium as well as submucosa of stomach and intestine. A third type, MCC, expresses chymase and is found in mucosal tissue of stomach, small intestine, and colon. MCTC is, in some respects, the counterpart of the serosal mast cell in rodents whereas MCT most closely resembles rodent mucosal mast cells.
It is thought that mast cell heterogeneity in different tissues arises from differences in the tissue microenvironment given that mast cell maturation occurs within peripheral tissues.13 In addition, it is unlikely that mast cells are irreversibly cast into distinct subtypes as described above but retain phenotypic flexibility to respond to physiologic/pathologic events and the accompanying alterations in the milieu of cytokines and other conditioning factors. Phenotypic changes can be teased in culture as, for example, co-culture of tissue-derived mast cells331–333 or RBL-2H3 cells334 with fibroblasts or by use of different culture supplements335 as well as in tissues by Trichinella spiralis infection336. The distinctive features and locations of the mast cell subtypes suggest functional specialization, a concept that that is further supported by differences among mast cells from different tissues in the pattern of cytokines produced and in their sensitivity to secretagogues such as compound 48/80 and morphine.337
As should be evident from this review, a variety of receptors positively or negatively influence FcεRI-induced mast cell activation. Expression of these regulatory receptors varies from one species to another338 but remarkably little is known about the differences in expression of receptors among mast cell subtypes in a given species although there are indications that this may be so. Early studies had indicated that rodent serosal, but not mucosal, mast cells responded to the G protein stimulant, compound 48/80.339 Similar differences in responses to this compound had been noted between mast cells from human skin and those from lung, colon, tonsils and adenoids 340 and between MCTC and MCT cells from human skin and lung.341 The latter study also indicated that both types of cells responded to FcεRI ligation but only MCTC express C5aR (CD88) and respond to C5a. Expression of the neurokinin receptor 1 (NKR1) and responsiveness to its agonist, substance P, occur after co-culture of murine BMMC with SCF and IL-4282 in which these cells acquire the characteristics of connective tissue-type cells.335 Clearly these studies indicate that growth factors or culture conditions normally associated with expression of particular subtypes of mast cells can induce expression of receptors not normally expressed on mast cells
Mast cells and basophils are derived from pluripotent hematopoietic stem cells but their developmental pathways probably differ342 although there are remaining issues.343 Human mast cells are thought to originate from CD34+ stem cells in bone marrow, circulate in the blood as progenitors, and acquire their mature phenotype within tissues as a result of exposure to tissue factors such as, SCF, IL-3, IL-4, IL-9, nerve growth factor (NGF), and likely others.344 SCF is produced mainly by stromal cells where it expressed on the cell surface or released in soluble form.345 The SCF receptor, Kit (CD117), is expressed on hematopoietic stem cells and is retained on mast cells throughout their development and differentiation but is down-regulated during differentiation of basophils and other bone marrow-derived cells. In contrast to mast cells, basophils reach their mature state in bone marrow before release into blood.
The early observations of two histochemically distinct populations of mast cells in rodents328, 329 and that one subset, the mucosal mast cell, was resistant to compound 48/80346 led to the concept of mast cell “heterogeneity”. This concept was reinforced by significant differences in expression of proteoglycans and proteases in serosal and mucosal mast cells.347, 348 Nevertheless, these two subsets exhibit “plasticity”. When BMMC from normal mice are injected into mast cell-deficient mice, the cells retain their mucosal cell characteristics at tissue sites that normally bear this phenotype while adopting a serosal mast cell phenotype at sites that contain such cells 349. A similar change in phenotype occurs on co-culture of BMMC with fibroblasts.333 IL-3 and SCF were found to be two factors regulating differentiation to, respectively, either the mucosal (i.e. chondroitin sulfate E positive)350 or serosal (i.e. heparin positive)351 mast cell phenotypes. Subsequent studies also imply that mast cells are conditioned by their cytokine environment under normal or pathological situations 162. The “heterogeneity” and “plasticity” of mast cells likely contribute to their multifunctional capability in immunological and pathological settings.
Evidence has been presented in the proceeding sections that the multifunctional adaptability of mast cells is conferred by the flexibility in receptor expression, and that their interactions, might allow selective release of mediators in given situations. The following sections examines such flexibility in the immunological and non-immunological conditions that have been attributed to mast cells. Some of the consequences of mediator release from mast cells, regardless of stimuli, are illustrated in Figure 5.
The mechanisms and consequences of mast cell activation in IgE-dependent allergic reactions, classified as type I or immediate hypersensitivity reactions, are described in detail in the previous sections. They include allergic asthma, rhinitis, conjunctivitis, atopic dermatitis, food allergies, and anaphylactic reactions. As these disorders have been extensively reviewed, 13, 338, 352–354 they need no further discussion here. The detrimental effects of mast cell activation, however, extend to other types of hypersensitivity reactions that include various autoimmune diseases.355. The most widely studied are bullous pemphigoid rheumatoid arthritis ( type II hypersensitivity), rheumatoid arthritis (categorized as a type III hypersensitivity by some or type IV by others), and multiple sclerosis (type IV or delayed hypersensitivity).355
The participation of mast cells in bullous pemphigoid was suspected from reports of high concentrations of IgE, degranulated mast cells and mast cell-derived chemoattractants in the subepidermal skin blisters of patients with bullous pemphigoid. These lesions contain high levels of histamine, eosinophils, and IgE class autoantibodies356 which elicit lesions reminiscent of bullous phemphigoid when injected into human skin grafted on athymic mice.357 These lesions likely result from formation of IgE and IgG autoantibodies against two skin hemidesmosomal proteins, BP180 and BP230. The disease can be closely mimicked in normal but not mast cell-deficient neonatal mice by intradermal injection of IgG antibodies against murine BP180 or BP230.358 Using this and genetic models, the mast cell C5a receptor and MCP-4 protease have been implicated in mast cell activation and subsequent tissue damage (See359 and citations therein).
The synovial joints are the primary, but not only, sites affected in rheumatoid arthritis. The underlying etiologic factors involved are uncertain although rheumatoid factor, an autoantibody the Fc portion of IgG, and anti-citulinated cyclic peptide antibodies are characteristic of this disease in many but not all patients. Uncertainty extends to animal models as no two are alike and none exactly reproduces human disease.360 The involvement of mast cells, among many ongoing processes, is indicated by their increased numbers, some in a degranulated state, in synovial tissues and the appearance of mast cell-derived mediators in synovial fluid and tissue.361 Mast cell-deficient mice fail to develop the disease when injected with serum from arthritic K/BxN mice, 362, 363 one of the several mouse models of this disease. Reconstitution studies in genetically modified mice have demonstrated that mast cell FcγRIII and mast cell-derived IL-1 are essential for progression of disease.363, 364 Mast cells have also been implicated in collagen as well as various forms of antibody-induced arthritis in mice355, 365 in which activation of mast cells by IL-33 through the IL-33 receptor (ST2) is said to play a critical role.197 IL-33, which is expressed in synovial membranes from patients with rheumatoid arthritis, markedly enhances production of TNFα and IL-1β in cultured synovial fibroblasts from these patients. It is proposed that IL-33, along with IgG, stimulates synovial mast cells to generate additional cytokines thereby amplifying the inflammatory response. Experimentally induced Arthus reaction, like rheumatoid arthritis, is dependent on formation of antigen-antibody complexes. The inflammatory reaction is dependent on activation of mast cells through FcγRIII and is not apparent in mast cell-deficient mice. 366
Interest in the role of mast cells in multiple sclerosis (MS) and its experimental counterpart, experimental allergic encephalomyelitis (EAE) in mice355 arose from early reports of correlations between progression of disease (inflammatory demyelination) and localization of mast cells in a degranulated state in human lesions367 in addition to the presence of mast cell proteases in cerebrospinal fluid.368 EAE is induced by immunization of genetically susceptible mice with myelin peptides although the causative agent in humans is unknown. Along with Th1, Th9, and Th17 cells, mast cells have been implicated in EAE, in part because of the effects of various pharmacologic agents.369 More direct evidence is the diminished severity of EAE in mast cell-deficient mice, unless deficiency is rectified by adoptive transfer of wt BMMC, 370 by mechanisms that likely require activation of mast cells via Fcγ receptors, 371 production of TNFα, and mast cell-associated neutrophil infiltration.372 Other autoimmune disorders thought to be mast cell related, based largely on correlative evidence, include type 1 diabetes, Guillain-Barré syndrome, scleroderma, ulcerative colitis, Crohn’s disease, Sjögren’s syndrome, Graves’ eye disease, and chronic idiopathic urticaria.355, 365 One large family of chronic urticarial diseases, cryopyrin-associated periodic syndrome (or CAPS), has been linked recently to constitutive production of IL-1β by skin mast cells which bear an active NOD-like receptor (NLRP3) caused by an activating missense gene mutation common to CAPS patients.373 The production of IL-1β results in the characteristic vascular leakage and neutrophil infiltration in CAPS urticarial lesions.
As discussed in previous sections, the widespread distribution of mast cells in skin and mucosal surfaces, and their ability to detect a broad variety of pathogens, places mast cells in a prime position for detection of invading organisms. The first evidence to support this hypothesis came from studies of models of intestinal helminth infection in which mast cells appear to accumulate and to degranulate at sites of parasite infestation, whereas mast cell-deficient W/Wv mice had diminished capacity for worm expulsion.374, 375 Subsequent studies have revealed that mast cells protect against diverse types of parasitic infections in animal models, including malarial parasitaemia, by a variety of mechanisms that involve recruitment and proliferation of mast cells, recruitment of other key immune cells which include eosinophils, release of SCF, IL-3, IL-4, mast cell proteases, and expulsion of parasites (reviewed in4).
Mast cells have also been shown to play a critical protective role, at least in animal models, against systemic or localized bacterial infections at sites as varied as peritoneum following caecal ligation, 376, 377, lung, 376, 378 bladder, 379 intestine, 380 and skin379, 381 by facilitating clearance of enterobacteria from these tissues376, 378, 379 or by limiting skin lesions caused by Pseudomonas aeruginosa.381 In some of these studies, protection is dependent on mast cell production of TNFα, 376, 377, 379 IL-6, 378 and leukotrienes382 as well as recruitment of neutrophils376 or dendritic cells.379 Data on viral infection is limited although it is known that mast cell numbers increase during pulmonary viral infections in rats and humans383, 384 and that cultured mast cells can be activated by viral challenge385, 386 or viral products.386, 387 One immunological consequence of viral challenge (with Newcastle disease virus), which is impaired in mast cell-deficient mice, is the recruitment of CD8+ T cells.386 Fungi are also known to exacerbate airway inflammatory disease by IgE-dependent and -independent mechanisms. The ubiquitous Aspergillus fumigatus, for example, is reported to induce degranulation of RBL-2H3 cells in an IgE-independent manner388 in addition to its well established propensity to cause IgE-dependent allergic disease in humans. Fungal spores of another common fungus, Trichoderma viride, potentiate IgE-dependent histamine release from human cells obtained by bronchial lavage.389 Yeast zymosan is an activator of mast cells, resulting in release of IL-6, 390 GM-CSF, IL-1β, and leukotrienes but not degranulation181, as well as mast cell-mediated peritoneal inflammation in mice.391, 392 The release of GM-CSF, IL-1β, and leukotrienes in the absence of degranulation in human cord blood-derived mast cells181 illustrates the capacity of these cells to respond selectively to a particular stimulant.
Apart from acting as first responders to microbial agents, mast cells can recruit other immune cells within lymph nodes to set the stage for adaptive immune responses.393 Early reports indicated that induction of contact dermatitis in a mouse model leads to migration of mast cells to regional lymph nodes394 and in such locations, mast cells produce MIP-1α to promote recruitment of T cells.394, 395 Mast cells were later shown to play a critical role in lymph node hypertrophy and the engagement of T cells with APCs upon bacterial infection, 396 Leishmania major parasite infection, 397 contact dermatitis, 398, 399 and IgE-dependent cutaneous anaphylaxis.400 Mast cell-derived TNFα seemed to play a predominant role in lymph node hypertrophy during bacterial infection, 396 contact dermatitis, 399 and IgE crosslinking;401 but only a partial role in the responses to the bacterial component, PGN.401 Other mast cell mediators implicated in cell recruitment include MIP-1α394, 395, IL-1402, and histamine.400 Mast cells also express ligands that activate co-stimulatory molecules on T cells. For example, OX40L, has been identified on mast cells as enhancing T cell activation following antigen stimulation403.More details need to be elucidated but it is apparent that mast cells promote migration of peripheral lymphocytes, 396 Langerhans cells, 400 and dendritic cells404 to regional lymph nodes thereby promoting interaction of APCs with T helper cells as a prelude to T cell/B cell collaboration405 or cytotoxic T cell (CTL) activation406 (Figure 5). The CTL connection was established by topical application of a TLR7 ligand (imiquimod) as adjuvant along with a MHC class 1 restricted cytotoxic T lymphocyte peptide in mice.402 Mast cell-derived IL-1 and TNFα mediated different phases of the response, namely, inflammation, migration of Langerhans cells, and lymph node hypertrophy. These responses were much diminished in mast cell-deficient mice but were restored by pre-immunization with peptide-loaded dendritic cells.
Evidence for the participation of mast cells immune suppression comes from studies of the regulatory T cell (Treg)-dependent allograft tolerance model in mice. In this model CD4+ Th1 and Th2 cells enable rejection of allogeneic tissue grafts whereas Treg cells suppress rejection. The possibility of interaction between Treg cells and mast cells was initially indicated by the finding that expression of specific markers of both cell types were increased in syngeneic skin grafts or tolerated allogeneic grafts.407 Production of IL-9 by Treg cells was thought to promote recruitment and activity of mast cells in tolerant tissue. It was reported later that inoculation of mast cell-deficient mice, which do not tolerate allogenic grafts, with BMMCs from normal mice restored tolerance in an IL-9-dependent manner.408 There is also evidence for direct interaction of Treg cells with mast cells through the OX40-OX40L connection.409 This interaction suppresses FcεRI-mediated mast cell activation in vitro and dampens the allergic response in vivo to suggest that mast cells have a pro-growth role in immune tolerance. These results could have therapeutic value once the mast cell factors involved and their roles have been identified.
An angiogenic role for mast cells was suspected because of their proximity to blood and lymphatic vessels and their accumulation in tissues associated with angiogenesis such as hemangiomas, polyps, and tumors. Moreover, mast cells are now known to release various angiogenic factors including fibroblast growth factor-2 (FGF2), vascular endothelial growth factor (VEGF), TGFβ, IL-8, histamine, heparin, and angiopoietin 1.11, 410, 411 Past evidence relied on the use of pharmacologic agents to block effects of mast cell mediators as well as the mast cell stabilizing compounds cromolyn and salbutamol. More direct evidence has come from studies involving reconstitution of mast cells in mast cell-deficient mice as, for example, the demonstration of mast cell-dependent angiogenesis in arthritic lesions in the K/BxN mouse model412 and Mycdriven cancer of pancreatic β cells in mice.413 Membrane tissues such as the chick chorioallantoic membrane and rat mesentery have been used also as models to identify histamine, 414 VEGF-A, 415 and FGF2416 as mast cell products capable of inducing angiogenesis in addition to the activation of mast cells themselves.414, 417 As noted in recent reviews, interest has focused on angiogenesis during tissue repair11, 13 and tumor development.6, 11, 418 Several mast cell mediators have been linked to angiogenesis in these models but a remaining challenge is dissecting their specific roles in the various phases of angiogenesis.
Mast cells are also thought to regulate tissue remodeling during wound healing, 9, 11 tumor progression, 419 and autoimmune arthritis.420 Apart from the original proposal that histamine was an essential mediator in wound healing and rapid tissue growth some 50 years ago, 421 it has been proposed that mast cells regulate all phases of healing from the initial inflammatory response to tissue remodeling and re-epithelialization.9 Mast cells accumulate along edges of healing wounds, scar tissue, keloids, and hypertrophic scars, and can produce fibroblast as well as keratinocyte growth factors in addition to matrix remodeling factors.9, 11, 422 Mast cell tryptase and chymase specifically stimulate α-1 collagen production by dermal fibroblasts and production of angiogenic factors by the extracellular matrix and thus implicated in tissue remodeling in squamous carcinomas.423 Studies of the role of mast cells in skin-incision healing models in both animals and humans have resulted in inconsistent conclusions. Mast cell reconstitution studies in W/Wv mice suggested that skin wound closure was dependent on mast cells in one study424 but not in another where collagen remodeling appeared to be defective during the final phase of wound healing.425 In a scald-injury model, extravasation and fibrous proliferation was impaired in W/Wv mice during wound healing although eventual wound closure and re-epithelialization was not.426 Apart from the previously noted caveats in the use of W/Wv mice, 427 a further handicap in the interpretation of such studies is the complexity of the healing process and the participation of other cells during the healing process. Although it is thought that mast cells play a role, perhaps substantial, in angiogenesis during tissue remodeling, 11 the individual roles of other participating cells and their interactions with mast cells need further clarification.
Equally complex is the role of mast cells in tumor formation and progression. Ehrlich first noted the abundance of mast cells in carcinomas where mast cells accumulate at the periphery rather than the interior of the tumor nodules428 Current studies reveal both positive and negative relationships between mast cell activity and tumor progression (see reviews5, 11, 13, 429, 430). Indicative of protumorigenic relationships are reports of abundant numbers of mast cells in proximity of blood vessels surrounding certain tumors and an apparent correlation with tumor-associated angiogenesis and adverse course of disease in patients.431–433 Positive relationships are also indicated in studies with mast cell-deficient mice. For example, neoplastic progression of squamous carcinoma423 and pancreatic islet tumors413 are substantially reduced in mast cell-deficient W/Wv and Wsh/Wsh mice, respectively. W/Wv mice are also less susceptible to development of chemically-induced intestinal epithelial tumors434 and metastases from subcutaneously administered murine B16 melanoma cells.435 A protumorigenic role is also implied in mouse studies of thyroid436 and colonic437 carcinomas. In the latter study, the evidence was buttressed by ablation of hematopoetic cells in a chimeric mouse model of polyposis, a predisposition to colon cancer. Reconstitution of bone marrow cells from Wsh/Wsh but not wt mice resulted in remission of existing polyps suggesting that mast cells were essential for polyps formation and by inference the environment conducive to malignancy. In these studies, tumor-induced angiogenesis and tumor penetration into the extracellular matrix appear to be dependent on release of cytokines, chemokines, and proteases from mast cells.415, 423, 436
One topic of particular interest is the involvement of mast cells in the inflammation associated with malignancy. Depending on the type of tumor, inflammation may exist before malignant changes or be induced by the malignant process itself but in either case inflammation may be conducive to tumor growth.438 Tumors can arise at sites of chronic inflammation and certain chronic inflammatory diseases are associated with increase susceptibility to cancer. The onset of malignancy itself may lead to recruitment of inflammatory cells including mast cells. The production of growth-promoting, chemotactic, and angiogenic factors by mast cells may then facilitate progression of tumors.5, 413, 439, 440
In contrast to the above reports, several clinical reports note favorable correlations between patient survival and numbers of tumor-associated mast cells in patients with breast441 and prostate442 cancers and B-cell lymphomas.443 Similarly, mast cells appeared to play a protective role in an intestinal adenoma mouse model.444 When these mice were bred with Wsh/Wsh mice, mast cell deficiency resulted in a marked reduction of tumor cell apoptosis without affecting tumor cell proliferation. An added twist to the story was the recent report445 that the reported tumorigenic activity of mast cells in the B16 melanoma mouse model435 was reversed after stimulation of mast cells by injection of a TLR2 agonist (Pam3CSK4), Moreover, the tumor suppressive action of Pam3CSK4 was not observed in Wsh/Wsh mice but was apparent after provision of BMMCs from wild type but not TLR2-deficient mice.445 Co-culture of BMMCs and melanoma cells indicated enhanced production of cytokines and chemokines by TL2-activated mast cells without provoking degranulation. Production of CCL3 and IL-6 were apparently critical factors in recruitment of NK and T cells as well as tumor growth.
The general impression from the reports so far is that the role of mast cells varies according to the type of tumor and phase of malignant growth. As noted in recent reviews, 5, 429, 446 research is handicapped by the correlative nature of clinical research, the diversity of experimental models, and the complexity of tumor biology. Even so, the cumulative evidence points to a regulatory role for mast cells in tumor growth and could be a potential target for adjuvant therapy429 in some but not all types of cancers.400
Although mast cells are well recognized initiators of acute allergic reactions associated with urticaria, rhinitis, atopy, anaphylaxis, and in combination with other immunological cells chronic allergic inflammation, it is now apparent that they also contribute to the body’s innate and adaptive defense mechanisms against invading micro-organisms. Central to these defense mechanisms, is the ability of mast cells to generate and release a wide array of bioactive molecules that produces a concerted attack on pathogens and which promote recruitment of other critical immune cells. However, it is these same mediators that can adversely affect surrounding tissues in the host resulting in autoimmune disease as well as allergic disorders. As discussed here, other less well established roles, adverse and beneficial, include immune suppression, tumor generation and progression, and wound healing. Therefore, complete assessment of the biologic role of mast cell remains elusive but much progress has been made in recent years. We have noted, as have others, that activation of mast cells is not necessarily global but, depending on the type of stimulus or combination of stimuli, can be more attuned for a given pathophysiologic situation. Some recent examples of selective release of mediators have been described that, as more examples emerge, point to possible refinement of therapy for particular disease types. An example is the NOD-like receptor-dependent release of IL-1β from mast cells and ensuing histamineindependent rash in CAPS. This disease is effectively treated with IL-1 receptor antagonists instead of antihistamines.373 This topic, in our view, is worthy of additional research.
Work in the authors’ laboratories has been supported by funding from the Intramural Research Program of the National Institute of Allergy and Infectious Disease and National Heart, Lung, and Blood Institute, National Institutes of Health. Due to space limitations, we do not cite all pertinent literature when adequately covered by most recent reviews. This does not imply that studies not quoted are of lesser merit.