|Home | About | Journals | Submit | Contact Us | Français|
Mast cells, perhaps best known by their ability to trigger allergic reactions after stimulation through the FcεRI, express the unusual phosphatidylinositol-3-kinase (PI3K)-dependent, Rac-binding protein SWAP-70. Here we show that the IgE-mediated passive cutaneous and the systemic anaphylactic responses are strongly reduced in SWAP-70−/− mice. Cultured SWAP-70−/− immature bone marrow mast cells (BMMC) are also impaired in FcεRI-mediated degranulation, which can be restored by expression of exogenous wild-type SWAP-70, but less so if a phosphatidylinositol trisphosphate (PIP3) binding mutant is expressed. SWAP-70 itself supports inositol-3-phosphate and PIP3 production, the latter indicating a potential feedback from SWAP-70 towards PI3K. FcεRI-stimulated transcription and release of cytokines is controlled by SWAP-70. Key FcεRI signal transduction events like activation of LAT by phosphorylation, activation of Akt/PKB and of p38 MAPKinase are reduced in SWAP-70−/− BMMC, but ERK is strongly hyperactivated. Some requirements for SWAP-70 were apparent only under limited-strength signaling conditions. We suggest that SWAP-70 defines a new element of efficient mast cell activation upon FcεRI signaling, important for the control of mast cell dependent anaphylaxis.
Murine SWAP-70, originally isolated from activated, mature B lymphocytes , is strongly expressed in mast cells . Molecularly, SWAP-70 loosely resembles proteins of the Dbl-family of guanine nucleotide exchange factors (GEFs) for small RhoGTPases [3, 4]. SWAP-70 interacts specifically with the RhoGTPase Rac, which is a central molecular switch in a number of signal transduction pathways including FcεRI and c-kit signaling [5, 6], and supports accumulation of its activated form, to which it preferentially binds . In mast cells, SWAP-70 is found in the cytoplasm and at the cytoplasmic membrane, but unlike in activated B cells is not observed in the nucleus [2, 8, 9]. A pleckstrin homology (PH) domain  exists in the center region of SWAP-70, binds PIP3, the second messenger product generated by PI3K, and mediates membrane localization of the protein. As shown with a mutant (R230C) impaired in PIP3 binding, PIP3 binding is necessary for SWAP-70 to localize to membrane F-actin structures called membrane ruffles [7, 11], which represent sites of membrane F-actin remodeling, at which Rac activation was also reported to happen . The Dbl homology domain of SWAP-70 displays only limited amino acid identity (ca. 20 %), and overlaps with a tripartite coiled-coil region . In SWAP-70, the DH homology is positioned not N-terminal but C-terminal to its PH domain, contrasting nearly all proteins of the Dbl family. SWAP-70’s closest homolog, a protein called Def-6, SLAT, or IBP [13–15] is the only known protein that also carries the PH and DH domains in the N-C arrangement. Thus, SWAP-70 and IBP may form a rather unique RhoGTPase-interacting protein family. IBP was reported to act as a GEF for Rac and Cdc42, and functions in T cell signaling, particularly of the Th2 subset .
Earlier, we determined a function for SWAP-70 in mast cells in signaling from the c-kit receptor . SWAP-70−/− BMMC are impaired in a number of c-kit triggered processes indicating that SWAP-70 is an important regulator of specific c-kit effector pathways including mast cell activation, migration, and cell adhesion.
Mast cells contribute important and specific functions to the immune system [17–20]. One well-known function is in IgE-triggered allergic reactions. After binding IgE, the cells can be stimulated by crosslinking the FcεRI-IgE complex by either antigen or by anti IgE to release a variety of biologically active, preformed mediators such as histamine, proteoglycans, proteases, serotonin, and others. Crosslinking also induces transcription, synthesis and subsequent release of cytokines. Some reports suggest that binding of only IgE to the FcεRI without or with minimal crosslinking initiates limited signaling, which induces the production of certain cytokines [21–23].
In exocytosis assays that monitor release of hexoseaminidase as an indicator for degranulation, SWAP-70−/− BMMC, stimulated by crosslinking the IgE/FcεRI complex with anti IgE, yield a signal of about one-fifth of that of wild-type (wt) BMMC. If cross linked by antigen, the reduction is about 50 % . Triton lysis and ionomycin-triggered degranulation, however, generated the wild type-like signal, indicating that SWAP-70−/− BMMC contain functional granules, which upon ionomycin-triggered Ca++-influx release their content. Thus, not degranulation per se, but the IgE/FcεRI-signaling pathway leading to exocytosis requires SWAP-70 . Events downstream of the FcεRI required for mast cell degranulation include, among others, activation of LAT, IP3 and PIP3 production, activation of kinases such as Akt, p38, and ERK, but whether SWAP-70 acts in any of these processes is unknown. Generally, the precise signaling pathways that lead to degranulation are not yet described in sufficient details, and more molecules important in this pathway such as regulators or effectors need to be determined and characterized [24–29].
Our studies were done in wt or SWAP-70−/− mice or with cultures of IL-3 dependent BMMC derived from them . These cultures, homogenous after about 4–5 weeks, reflect mostly the mucosal type of mast cells [30, 31]. FACS analysis of surface-bound IgE showed that more than 98 % of wt or SWAP-70−/− cells bind IgE. The equal intensity of the signals also indicates similar numbers of receptors on their surface. Wt or SWAP-70−/− mast cells respond similarly to IL-3 with regard to proliferation, cell cycle, and apoptosis under starvation conditions .
Together, indications exist that SWAP-70 plays important role(s) in mast cell signaling. This study aimed at identifying specific contributions of SWAP-70 to FcεRI signaling in vivo and in vitro.
In order to assess the importance of SWAP-70 for mast cell function in the organism and to study mast cell FcεRI mediated reactions in vivo we induced passive cutaneous anaphylaxis (PCA) in mice by intravenous injections of the antigen DNP-BSA with 0.5 % Evans blue into wt or SWAP-70 deficient mice that were sensitized by intradermal ear injections the day before with IgE in HBBS (left ear) or HBSS alone (right ear). Extravasation of the dye is caused by histamine and serotonin release from FcεRI-activated mast cells  and the dye accumulates in the tissue. 30 min after antigen stimulation the extravasated dye in the ears of mice was measured spectrophotometrically. SWAP-70−/− mice showed 1.7-fold lower total levels of extravasated dye as compared to wt (Fig. 1A). However, the signal observed with SWAP-70−/− mice (0.255 ± 0.019) is at background levels as derived from the control ear (0.253 ± 0.033), while a clear signal of 0.41 ± 0.048 was obtained from wild-type mice (control at 0.207 ± 0.041).
In addition, we carried out systemic anaphylaxis experiments by challenging IgE-primed mice with 500 μg DNP-KLH and measuring changes in body temperature caused by anaphylactic shock  (Fig. 1B). A decrease in basal body temperature from 38 °C to about 30 °C was observed within 30 min in wt animals. Temperature returned to normal after about 2 h. No change in body temperature was observed in mice that were not IgE-primed prior to DNP-KLH stimulation. SWAP-70−/− mice primed with IgE and challenged with DNP-KLH, showed only a small decrease in body temperature to about 35 °C and a quick recovery, indicating a strong reduction in the systemic anaphylactic response induced by FcεRI.
Since mature peritoneal mast cells do not express SWAP-70 , we asked which types of mature tissue mast cells express SWAP-70. Analyses of fixed, paraffin embedded sections from back skin and sections from frozen stomach mucosa by immunohistochemistry using FITC-conjugated avidin to stain mast cells  and polyclonal rabbit anti-SWAP-70 followed by secondary anti-rabbit Cy3, showed that mast cells from back skin as well as mucosal mast cells strongly express SWAP-70 (not shown). Back skin sections from SWAP-70−/− mice showed fewer number of mast cells in agreement with an earlier report on two-fold reduced skin mast cells in these mice , while mucosal stomach mast cell numbers are unaltered compared to wild type. Gastrointestinal tract mast cells were stained with toluidine blue and no difference in numbers between wt and SWAP-70−/− mice were found.
Earlier, we reported that SWAP-70−/− BMMC are defective in vitro in FcεRI mediated degranulation of preformed mediators such as hexoseaminidase . To confirm a direct role of SWAP-70 and to get insights into the molecular mechanisms of its function we transfected mutant BMMC with pEGFP constructs to transiently express SWAP-70, or the site-specific mutant R230C, which is reduced in PIP3 binding , or with vector alone for control. Expression of SWAP-70 in transfected SWAP-70−/− and control wt BMMC was assessed by performing western blot analysis (Fig. 2A). The transfected cells were starved for 2 h, primed with anti DNP-KLH IgE for 1h, followed by crosslinking using 100 ng of the antigen DNP-KLH. The results are as shown in Fig. 2B as the average of 3 experiments. Compared to wt cells SWAP-70−/− BMMC show app. 1.4-fold reduced degranulation after crosslinking with antigen, similar to the two-fold reduction reported earlier . Infection of wt or SWAP-70−/− cells with control vector had only marginal effects. Wt BMMC that were transfected with SWAP-70 showed an increase of about 1.8-fold in degranulation upon stimulation with antigen. If background levels are deducted, the stimulation is 2.5-fold. Reconstitution of SWAP-70−/− BMMC with SWAP-70 restored degranulation to levels similar to that seen in non-transfected wt BMMC. Expression of the SWAP-70 R230C mutant, which is compromised in PIP3 binding , in wt or SWAP-70−/− cells stimulated degranulation to a lesser degree than expression of the unaltered SWAP-70. The failure of R230C to efficiently restore degranulation despite its quite strong expression indicates a requirement for PIP3 binding of SWAP-70 in signaling to degranulation.
Activation of mast cells induces synthesis and secretion of cytokines rendering mast cells important contributors to the cytokine repertoire of an organism. Earlier, we obtained indications for reduced secretion of IL6 or TNFα by FcεRI stimulated SWAP-70−/− BMMC . To assay not only for degranulation of preformed mediators as above, but also for induction of new synthesis of mediators, we asked whether SWAP-70 is involved in regulating transcription of cytokine genes. Using RT-PCR with RNA prepared at various time points after either crosslinking of the IgE-loaded FcεRI or after IgE loading only without crosslinking, we show here that transcription of several interleukins is altered in SWAP-70−/− BMMC (Fig. 3A&B). Upon crosslinking, induction of transcription of IL1, IL2 or IL10 largely fails in SWAP-70−/− cells, and transcription of IL6 and TNFα is mildly reduced if compared to wt cells. IL5 transcription is upregulated as in wt cells, IL4 expression is mildly increased, IL3 transcription may be slightly reduced. Transcription of β-microglobulin is unchanged as compared to the HPRT control.
To test whether additional requirements for SWAP-70 would become apparent under limiting signaling conditions, i.e. by sensitization with IgE alone, we tested cytokine gene expression at different time points after IgE binding (Fig. 3B). Expression of IL6, IL10 or TNFα under these conditions is readily observed in wt but very weak in SWAP-70−/− cells, where IFNγ is also clearly reduced. IL5 expression is similarly induced in both genotypes, and IL3 (not shown) is not detectable in either. Remarkably, expression of IL4 is strongly upregulated in mutant BMMC, while it is not induced in wt cells. Since IL4 released by mast cells upon activation through the IgE receptor has been previously shown to provide an initial pulse to trigger T cells for sustained release of IL4  and to contribute to B cell activation , and thus is biologically relevant, we investigated basal levels of IL4 in the serum of SWAP-70−/− mice, compared to wt (Fig. 4A). Analysis by ELISA revealed two-fold higher levels of IL4 in SWAP-70 deficient mice independent of the strain background, i.e. in both, the 129SvEMS and C57BL/6 wt or SWAP-70−/− strains. To further evaluate the effect of IgE on IL4 levels in SWAP-70−/− mice compared to wt we injected different concentrations of IgE and 24 h later analyzed the serum by ELISA for IL4 (Fig. 4B). Both in control mice and in mice injected with 50 or 100 μg IgE, SWAP-70−/− mice showed higher serum IL4 levels compared to wt, even though at 100 μg IgE an inhibitory effect was observed in all strains.
Degranulation as well as cytokine gene expression triggered by FcεRI stimulation is mediated by activation of kinases such as Akt, ERK and p38. Therefore, we analyzed activation of these kinases in SWAP-70 deficient BMMC. Western blotting using antibodies specific for phosphorylated Akt (P-Ser 473) revealed differences between wt and SWAP-70−/− BMMC, both upon IgE sensitization only as well as after FcεRI crosslinking. In total cell lysates of wt cells, Akt phosphorylation is higher than in lysates from SWAP-70−/− cells (Fig. 5A&B). Phosphorylation was further mildly increased for a few minutes after crosslinking. SWAP-70−/− cells showed only weak Ser 473-phosphorylation of Akt, visible often only upon prolonged exposure of the film. Total amounts of Akt protein were the same in both cell types.
Since phosphorylation of Akt is affected in SWAP-70−/− BMMC, we performed Akt kinase assays using GSK3-α fusion protein as substrate in the presence of ATP with BMMC lysates from SWAP-70−/− or wt BMMC (Fig. 5C). About 10-fold less phosphorylated GSK-3α was noted if SWAP-70−/− BMMC lysates were used as compared to wt lysates. This confirms the requirement of SWAP-70 for proper activation of Akt in FcεRI-stimulated BMMC.
The extracellular signal-regulated kinases (ERK) constitute a subfamily of mitogen-activated protein kinases (MAP kinases). ERKs are involved in the control of cell proliferation, cell shape, cell mobility, and cell-cell interactions [38–40]. Since signaling through the FcεRI stimulates ERK, [21, 41, 42] we asked whether ERK activation may be affected by SWAP-70 deficiency as well. We tested phosphorylation of ERK1 and ERK2 by a phospho-specific antibody. After triggering signal transduction by IgE sensitization alone (Fig. 6A) or by crosslinking with anti IgE (Fig. 6B) we found phosphorylation of both ERKs – though under these conditions primarily ERK2 – upregulated in wt cells. However, phosphorylation in SWAP-70−/− cells is much stronger than in wt cells. This increased phosphorylation is independent of the mode of stimulation, although the difference between wt and SWAP-70−/− cells was highest if the FcεRI was crosslinked. The total amount of ERK2 is similar in cells of both genotypes. In SWAP-70−/− BMMC, an additional band appears that reacts with the anti ERK1/2 antibody and migrates at a position between ERK1 and ERK2. A comparable polypeptide has been observed earlier  and seems to replace the ERK1 band in non-stimulated or briefly stimulated cells. After 1 h of IgE stimulation, the typical ERK1 signal appears. The identity of the SWAP-70−/− -specific band is subject to future studies. Differentially regulated isoforms of ERK have been reported in various systems .
Kinase assays using an ELK-1 fusion protein as a substrate for ERK immunoprecipitated from BMMC lysates, which were prepared after stimulation with only IgE, revealed much higher levels of activated ERK in SWAP-70−/− BMMC lysates as compared to wt lysates (Fig. 6C) (app. 12-fold higher at the 10 min time point). This supports our notion of hyperactivation of ERK in the absence of SWAP-70. Similarly increased ERK activity was found upon crosslinking of the FcεRI (data not shown).
As another key component, kinase p38 has been implicated in mast cell signaling to control migration, adhesion, cytokine production, and survival [21, 41, 44, 45]. We analyzed p38 phosphorylation and activity. Upon FcεRI stimulation through IgE binding only, we observed that SWAP-70−/− BMMC show similar or only mildly lower phosphorylation of p38 compared to wt BMMC (Fig. 7A). The difference is much more pronounced if the FcεRI/IgE complex is crosslinked. (Fig. 7B; there is no detectable signal after 1 h pretreatment with IgE only). There appears also to be more p38 expressed in wt BMMC before and after stimulation, although slightly more lysate may have been loaded in Fig. 7B. To assess p38 kinase activity, we subjected lysates from wt or SWAP-70−/− cells with or without FcεRI crosslinking to immunokinase assays with the ATF2-GST fusion protein as substrate (Fig. 7C). Very little p38 activity was observed in SWAP-70 deficient BMMC, indicating a reduction of ca. 10- to 25-fold (5 min time point) compared wt.
Phospholipids are produced during FcεRI and serve as key second messengers. PI3K is activated and catalyzes the synthesis of PIP3 . This is considered a key reaction, for several signaling proteins like serine threonine kinases, PTKS, GEFs, and GTPases depend on PIP3, which they bind, for their translocation from the cytosol to the membrane . SWAP-70, which also binds PIP3, is considered to act downstream of PI3K, but we wondered whether its absence nevertheless affects PIP3 production, since IL4 production is increased and known to down-regulate PI3K by recruiting SHP-1 . We assayed PIP3 production by PI3K, which was immunoprecipitated from wt or SWAP-70−/− cells (Fig. 8A). Upon IgE binding PIP3 production by PI3K was found app. 1.5-fold lower in SWAP-70−/− cells than in wt. By further crosslinking of the IgE-loaded FcεRI, PIP3 production remained consistently lower (average wt 48 +/− 5 pmol; SWAP-70−/− 40 +/− 4.5 pmol). Immunoprecipitation experiments from RIPA lysates of IgE/anti-IgE activated BMMC with anti phosphotyrosine antibody (4G10) and subsequent probing of the precipitate in immunoblotting with anti p85 indicated reduced phosphorylation of the p85 regulatory subunit of PI3K (not shown).
Phospholipase Cγ (PLCγ) converts phosphatidylinositol-4,5-biphosphate (PIP2) to inositol-1,4,5-triphosphate (IP3), which through binding to IP3 receptors on the endoplasmic reticulum allows Ca++ channels to be opened . This process may be at least partially dependent on PIP3 production and is required for FcεRI-triggered mast cell degranulation [49–52]. Ca++ flux is mildly impaired upon FcεRI in SWAP-70−/− BMMC (data not shown), similar to our observation of reduced Ca++ flux upon c-kit activation of these cells (Sivalenka and Jessberger, 2004). PLCγ1 and PLCγ2 phosphorylation in FcεRI-triggered SWAP-70−/− BMMC are similar to wt, but their distribution between NP-40 soluble and insoluble fractions differs, since we find much lower amounts of PLCγ in the soluble fraction, but more in the cytoskeletal fraction, prepared from SWAP-70−/− cells as compared to wt fractions (not shown). We analyzed levels of IP3 in FcεRI-stimulated SWAP-70−/− or wt BMMC. Kinetic analysis in 30 sec intervals of IP3 levels in defined numbers of cells revealed that the spike of IP3 production in wt cells, seen within 1–3 min after FcεRI crosslinking, is missing in mutant cells, where only a mild increase was observed (Fig. 8B). In wt cells, levels of IP3 increase from an average of 8 pmol to 24 pmol per 107 cells. Analyzing IP3 levels at 15 sec intervals confirmed the absence of the spike in IP3 production in SWAP-70 deficient cells (not shown).
The transmembrane adaptor molecule LAT is another key element in early steps of FcεRI signaling . Upon its phosphorylation and activation by tyrosine kinases, it binds to adaptors and other signaling proteins to drive several transduction pathways, including PLCγ, Grb2, Gads, SLP-76, and forms membrane-bound signaling complexes including many of these proteins . To determine whether SWAP-70 is involved in a step as early as LAT phosphorylation, we analyzed SDS-solubilized total extracts (RIPA extracts) from SWAP-70−/− and wt BMMC for LAT phosphorylation using phospho-specific antibodies to LAT (pTyr 171). SWAP-70−/− BMMC showed reduced phosphorylation of LAT as compared with wt BMMC, both upon and prior to stimulation with IgE/anti-IgE, while total LAT expression remained unaltered in both (Fig. 9A). We further performed immuno precipitation experiments using anti-SWAP-70 antibody with BMMC lysates prepared in less stringent RIPA buffer that only solubilized a fraction of LAT. Still, we observed some IgE-dependent co-precipitation of SWAP-70 and phosphorylated LAT, which disappeared after anti IgE crosslinking (Fig. 9B). However, the reverse precipitation, i.e. using anti LAT antibody, did not precipitate LAT with any antibody tried. Together, these experiments indicate that SWAP-70 also affects LAT and thus an early step in FcεRI signaling.
Initial data from a previous study suggested a role for SWAP-70 in mast cell FcεRI signaling  and we set out to identify specific requirements for SWAP-70 in the FcεRI signaling pathway. Our data reveal roles for SWAP-70 in the activation of LAT, Akt, ERK, and p38, in the production of PIP3 and IP3, and in the regulation of cytokine gene expression as illustrated in a working model (Fig. 10). In addition, IgE-dependent local and systemic anaphylaxis are impaired in SWAP-70−/− mice, correlating with impaired FcεRI signaling and degranulation seen in vitro.
The key contribution of mast cells to anaphylactic reactions has been widely demonstrated. For example, anti-IgE induced mast cell degranulation is involved in anaphylactic reactions that result in cardiopulmonary changes and mortality in mice . Mast cell deficient W/Wv as well as Sl/Sld mice were shown to exhibit negligible anaphylactic responses upon anti-IgE stimulation as compared to wt mice. Upon mast cell reconstitution cardiopulmonary alterations and mortality were restored to wt levels indicating the essential role for mast cell FcεRI signaling in anaphylaxis. Further, functional elimination of certain effectors of FcεRI signaling like LAT , SLP-76 , Btk , or vav , causes diminished responses in mast cell mediated anaphylaxis. In contrast, mice lacking the signal transducer PIPKIα, which negatively regulates mast cell degranulation, exhibit enhanced anaphylactic responses upon FcεRI stimulation . However, only a limited number of FcεRI signaling components are known, and only few have been shown to be involved in the anaphylactic reaction. Here we demonstrate SWAP-70 as an important factor in passive cutaneous and active systemic anaphylaxis. The reduction in anaphylactic responses is likely a direct consequence of impaired FcεRI signaling as observed in SWAP-70−/− BMMC. However, not all types of mature mast cells express SWAP-70, which is absent in mature peritoneal mast cells . We have now determined that skin mast cells and stomach mucosal mast cells express SWAP-70. As reported earlier , there are somewhat fewer skin mast cells present in SWAP-70−/− animals, but the number of mucosal mast cells, at least in the stomach, is like in wt mice. The strong reduction of systemic anaphylaxis indicates that SWAP-70 is particularly important in the mucosal cells, since they express SWAP-70, show an unaltered frequency in SWAP-70−/− mice, and are similar to BMMC, which are impaired in degranulation in the absence of SWAP-70. The cutaneous anaphylactic response cannot be explained by the two-fold reduced number of skin mast cells in SWAP-70−/− mice, since if background signal is deducted, SWAP-70−/− mice show essentially no cutaneous anaphylaxis signal at all. Thus, SWAP-70 is of key importance to the anaphylactic reaction. In any case, the two explanations for the reduced anaphylactic response in the skin – impaired FcεRI signaling in mature mast cells or developmentally decreased numbers of mature mast cells – are not mutually exclusive.
To unravel the specific contribution of SWAP-70 to FcεRI signaling, we tested the SWAP-70 dependence of a number of events associated with degranulation and FcεRI signaling and demonstrate a series of hitherto unknown functions of SWAP-70. Some functions, such as regulation of cytokine expression, are apparently downstream of PI3K. The supportive role of SWAP-70 in PI3K activation and PIP3 production, however, is seemingly upstream of PI3K. Rather than classifying SWAP-70 as an upstream or downstream factor, we hypothesize that SWAP-70 plays a complex role in regulating and perhaps coordinating different signaling events within a signaling pathway such as the FcεRI pathway. Once activated by binding PIP3, and likely by other events such as phosphorylation (Pearce and Jessberger, manuscript in preparation), SWAP-70 may – as our data indicate – in a feedback-loop directly or indirectly (see below) further increase or maintain the generation of PIP3 itself. Thus, SWAP-70 may have a signal multiplier role. Although SWAP-70 may serve as a feedback multiplier for PI3K activation, e.g. by down-regulating IL4 production, which has been shown to inhibit PI3K activity in an autocrine fashion , the loss of this feedback loop still allows production of some PIP3.
The impairment in PIP3 production seen in SWAP-70−/− BMMC may be caused by increased IL4 expression in these mutants, and therefore be an indirect effect. Interestingly, in c-fos deficient mast cells expression of many cytokines is reduced but IL4 expression is increased , similar to SWAP-70−/− BMMC. Since Lee et al. also showed that c-fos−/− mast cells express much less SWAP-70, our results suggest that the effect on cytokine expression seen in c-fos−/− mast cells may be a consequence of reduced SWAP-70 expression. Mast cells contribute IL4 especially in the early phase of an immune response, where their IL4 supports expression of IL4 in T cells [19, 60].
The cytokines that are only weakly expressed in SWAP-70−/− BMMC belong to various classes, among them pro-inflammatory cytokines such as IL1, IL2, or TNFα and the anti inflammatory IL10, rendering the biological consequences complex. While IL4 may downregulate expression of the FcεRI , IL6 is thought to upregulate it . There is induction of IL6 expression upon FcεRI crosslinking to almost wt levels in SWAP-70−/− cells, but if only IgE is bound, SWAP-70−/− BMMC only weakly express IL6. The parallel upregulation of IL4 may suggest that expression of the FcεRI is no longer as much supported in the SWAP-70−/− cells as it is in wt cells if treated with IgE only. However, RT-PCR and FACS indicated equal levels of FcεRI expression in wt and SWAP-70−/− BMMC before and after stimulation .
Cytokine gene expression depends on upstream signaling events that include molecules like ERK (IL4) [63–65] or Akt (IL2) . Impaired Akt activation and ERK hyperactivation in SWAP-70−/− BMMC may contribute to the observed changes in cytokine expression, which, however, may also be generated through other pathways, e.g. altered activation of transcription factors or translocation of factors into the nucleus through altered Rac activation and F-actin rearrangements. Nevertheless, the increase in IL4 production correlates with increased ERK activation, and positive stimulatory crosstalk between IL4 and ERK was reported .
Akt kinase acts in multiple processes including apoptosis and cell survival, proliferation, cell motility, gene expression and angiogenesis . Akt is activated in signaling from several receptors including the FcεRI and c-kit and its activation depends among others on PIP3 [68, 69]. Akt binds PIP3 in the cytoplasmic membrane via its PH domain, becomes phosphorylated, and then dissociates from the membrane to act on various substrates. Phenotypic consequences of impaired Akt activation in BMMC may include reduced exocytosis, gene expression, or compromised cell survival . Indeed, SWAP-70−/− BMMC were found to be more sensitive to γ-irradiation , impaired in degranulation and to display altered cytokine gene expression patterns. Since activated Akt may inhibit ERK , hyperactivation of ERK observed in stimulated SWAP-70−/− BMMC may be a result of impaired Akt activation. It remains to be shown, however, through which mechanism ERK becomes hyperactivated. ERK is known to be activated via the Ras/Raf/MEK pathway independently of PI3K/Akt [71, 72]. The upregulation of ERK activity may constitute a compensatory mechanism for the loss of Akt activation to support cell survival. However, we cannot exclude that SWAP-70 acts through an Akt-independent pathway as a negative regulator of ERK. This pathway is also likely not dependent on the LAT/PLCγ/IP3 pathway, for LAT and IP3 are reduced in the absence of SWAP-70. Experiments to pharmacologically inhibit PI3K activity by the broad spectrum PI3K inhibitor LY294002 abolished Akt and ERK activation in wt BMMC with and without FcεRI crosslinking, but did not abolish most of the activated ERK in SWAP-70−/− BMMC, pointing to an PI3K independent ERK activation pathway (not shown). IL4 gene expression was not affected by the LY294002 treatment, indicating that regulation of cytokine gene expression occurs through a PI3K/Akt independent pathway.
There are several lines of evidence indicating that the importance of SWAP-70 depends on the strength of the signal originating at the FcεRI. Crosslinking IgE bound to FcεRI sometimes triggers similar responses in wt or mutant cells, e.g. in expression of IL4, IL5, IFNγ, IL6, or TNFα. Binding of IgE, which weakly crosslinks due to its partially oligomeric form, to the FcεRI without further crosslinking – thus triggering a weaker signal – reveals a requirement for SWAP-70 for proper transcription of IL4, IL6 or TNFα. Similarly, some phenotypes seen after FcεRI crosslinking are barely observable upon binding of IgE only, e.g. low p38 phosphorylation. Thus, for full amplification of a signal of limiting strength, SWAP-70 is often required. Upon crosslinking of the receptor, SWAP-70 deficiency can sometimes be overcome, probably by complementing factors or alternate converging pathways. The importance of the degree of activation of the FcεRI is for the plasticity of mast cell responses has been described [21, 73]. SWAP-70, with its roles in low-strength signaling and as a regulator and perhaps signal multiplier, likely constitutes an important factor in determining FcεRI dependent mast cell plasticity. Given the known role of SWAP-70 in rearrangement of cytoskeletal, cytoplasmic membrane-proximal F-actin structures, we hypothesize that SWAP-70 affects the FcεRI pathway and the specific components identified here through intracellular recruitment, assembly into proper signaling complexes at the right time and the right location. This hypothesis can now be tested.
Bone marrow derived murine mast cells (BMMC) cultures were derived from wt or SWAP-70−/− mice in the 129SvEMS genetic background. Cell culture procedures, reagents and assays for phosphorylation and activity of Akt, and ERK are described in Sivalenka and Jessbeger, 2004 . pEGFP vectors expressing SWAP-70 or the PIP3-binding defective SWAP-70 mutant R230C were described earlier , as were degranulation assays . The IL4 ELISA kit was obtained from Peprotech, NJ, USA. p38 phosphorylation and activity assays and RT-PCR for cytokine expression are described in Supplementary Material.
To induce PCA SWAP-70−/− or wt control mice were injected intradermally with 20μg IgE (anti-DNP, clone SPE-7) in HBSS in their left ears and HBSS as control into the right ears. After 24 h the mice were challenged intravenously with 100 μg of DNP-BSA in HBSS containing 1% Evans blue. Control mice received HBSS with Evans blue alone without DNP-BSA. 30 min after antigen challenge quantification of extravasated Evans blue dye was performed. Briefly ear tissues were minced in formamide and incubated at 80° C for 2 h. The optical density of dye in the extract was measured at 610 nm using a spectrophotometer. Optical density of samples extracted from the right ears (IgE-primed) was normalized to the internal control samples from respective left ears (non-primed).
Wt or SWAP-70−/− mice were sensitized by intravenous injection of 50 μg anti DNP-IgE, and were challenged with 500 μg DNP-KLH per mouse intravenously after 48 h. Respective controls were performed without IgE sensitization. Changes in body temperature associated with anaphylaxis  were monitored by measuring rectal temperature before and up to 2.5 h after antigen challenge using a rectal probe (Yellow springs instrument Co., Yellow springs OH).
Cultured wt or SWAP-70−/− BMMC (1 × 106/sample) were nucleofected with 2.5 μg of pEGFP-C1vector alone, pEGFP containing SWAP-70, pEGFP containing the R230C-PIP3 binding defective mutant of SWAP-70, or were mock transfected without DNA for control. An AMAXA nucleofection kit (macrophage kit) from AMAXA Biosystems was used with the Y01 program per manufacturers instructions. After 48 hr. BMMC were harvested and transfection efficiency measured by FACS analysis. An average of about 20–30% BMMC were positive for GFP. In control nucleotransfections, viability, proliferation, of transfected BMMC and their responsiveness to IL-3 or SCF were found normal.
Where indicated statistics were calculated using Student’s t test.
PI3K activity was measured using an activity ELISA kit obtained from Echelon Biosciences Inc., USA, as per manufacturers instructions. Briefly, PI3K was immunoprecipitated (rabbit anti p85; Cell Signaling Technology Inc.) from 106 BMMC stimulated with 0.5 μg/ml IgE and 0.5 μg/ml anti IgE for various time intervals. Subsequently, immunoprecipitated PI3K was used in kinase assays, with 100 pmol (2.4 μg) diC8 PI(4, 5) P2 as substrate. Reactions were terminated by addition of EDTA to 5 mM, and the reaction mixtures were transferred to an incubation plate. Alternatively, standards were prepared in separate wells of the same incubation plate containing PIP3 at concentrations ranging from 0 to 100 pmol per 50 μl standard solution. Standards and samples were run in triplicate. Equal volume of PIP3 detector (1:200) was added to each of the wells and incubated at room temperature for 60 min in the dark. 100 μl of secondary detection reagent (1:40 dilution) was added and incubated for another 60 min. Solutions were then discarded and wells washed with 300 μl of Tris-buffered saline (TBS) or TBS containing 0.05 % Tween-20. 100 μl of TMB was added to each well, and reactions were stopped after color development (within 20–30 min) by addition of 100 μl 0.5 M H2SO4. Absorbance was measured at 450 nm.
IP3 production was measured using a commercially available assay kit BIOTRAK TRK1000 (Amersham Pharmacia Biotech, NJ). 107 BMMC were sensitized with anti-DNP IgE and stimulated in PBS containing protease and phosphatase inhibitors in the absence or presence of 0.5 μg/ml of anti IgE. Stimulation was terminated at the desired time intervals by adding 1/10 V/V ice-cold 100 % trichloroacetic acid. Samples were then neutralized and IP3 partitioned into the aqueous phase using water-saturated diethyl ether. The amount of IP3 in each sample was quantified by employing competitive binding assays of IP3 to an IP3 binding protein, competed for by a tracer of D-myo-H3 inositol 1,4,5–trisphosphate. Standard IP3 provided with the kit was assayed simultaneously in concentrations ranging between 0.19 – 25 pmol; pmol IP3 in a sample was derived from the standard curve.
We thank Dr. Svetlana Kupershtokh, Ms Astrid Berg, and Mr Michael Chopin for technical help, and Dr. Glen Pearce for discussions. This work was supported by a grant from the NIH (AI 049282).