Skip to main content
NCBI home page
Immunology logoLink to Immunology
. 1999 Jul;97(3):481–489. doi: 10.1046/j.1365-2567.1999.00790.x

Interferon-γ regulates the interaction of RBL-2H3 cells with fibronectin through production of nitric oxide

F L Wills 1, M Gilchrist 1, A D Befus 1
PMCID: PMC2326863  PMID: 10447771

Abstract

Interferon-γ (IFN-γ) is an important regulatory cytokine in cell proliferation, differentiation, adhesion, mediator release, and gene induction. This diversity of effector roles is achieved by a variety of incompletely understood mechanisms. In the mast cell (MC), IFN-γ downregulates mediator synthesis and secretion. The present study demonstrates and characterizes for the first time IFN-γ inhibition of adhesion of the MC analogue RBL-2H3 to the extracellular matrix protein fibronectin (FN). Inhibition requires preincubation of the cells with IFN-γ for 20 hr, and is statistically significant at 100 U/ml IFN-γ. Flow cytometry indicates that cell surface expression of very late antigen-4 (VLA-4), VLA-5, and the vitronectin receptor (VNR) remain constant following IFN-γ treatment, indicating the inhibitory effect of IFN-γ on adhesion to FN is not achieved through a reduction in integrin receptors for FN. Fluorescent labelling with Texas red phalloidin demonstrated rearrangement of the actin cytoskeleton in response to IFN-γ was not significant. The tyrosine phosphatase inhibitor vanadate, and the nitric oxide (NO) synthase inhibitor L-NAME, reduced the IFN-γ effect on adhesion to FN by 62 and 70%, respectively, demonstrating that the IFN-γ effect is dependent upon the production of NO, potentially though a tyrosine phosphatase dependent mechanism. The NO donors sodium nitroprusside and S-nitrosoglutathione mimicked the effect of IFN-γ. Thus, following stimulation with IFN-γ, NO plays an autocrine role in the MC, and is able to modulate integrin function. This adds to the pathways NO is able to inhibit in the mast cell, shows that endogenous NO is able to inhibit these pathways, and suggests NO is impinging upon an element common to many signalling mechanisms in the MC.

INTRODUCTION

Control of mast cell (MC) function is imperative to the management of many inflammatory diseases since MC mediators contribute to tissue damage and propagate the inflammatory responses. Understanding the signalling systems involved in mediator secretion and cytokine production is a major research focus, including agents which up or downregulate MC function. Several cytokines are known to enhance immunoglobulin E (IgE)-mediated MC secretion and cytokine production, including stem-cell factor (SCF), interleukin-3 (IL-3), and IL-4·1–4 However, the mechanisms of this potentiation are not understood. Factors which downregulate MC function have also been identified. Interferon-γ (IFN-γ) can decrease release of mediators, including histamine, tumour necrosis factor-α (TNF-γ),5 serotonin,6 and arachidonic acid metabolites.2 In addition, IFN-α/β,7 transforming growth factor-β1,8 as well as anti allergic drugs such as nedocromil sodium and sodium cromoglycate9 reduce mediator release. Using IFN-γ as a model for cytokine mediated down regulation of MC function, we wished to extend this observation to determine if adhesion to the extracellular matrix, which synergistically activates MC along with FcεR1 activation, could also be inhibited. Because the mechanism of IFN-γ inhibition is unknown, this would indicate if IFN-γ is inhibiting a portion of the signalling pathway common to both FcεR1 signalling and adhesion, or conversely, if it is selective in its inhibition of these functions. An understanding of how these agents regulate MC function may be exploited for the development of novel therapeutic agents.

It has previously been established that MC, and in particular the MC line RBL-2H3, binds to fibronectin (FN), vitronectin, collagen type I, collagen type IV, and laminin. Adhesion of MC to extracellular proteins is dependent upon the activation state of the cell. Naive bone-marrow-derived cultured MC or peritoneal MC require prior activation to express the integrins necessary to adhere to extracellular matrix proteins. SCF, phorbol myristate acetate (PMA), ionophore, FcεRI aggregation and FcγRIII aggregation can activate MC binding to extracellular matrix proteins.11–14 RBL-2H3 cells constitutively express the integrins very late antigen-4 (VLA-4), VLA-5 and vibronectin (VNR), which mediate the interaction of these cells with FN. Thus, while the RBL-2H3 cell line may vary functionally from an in-vivo derived MC, it is an attractive, well-defined, model of a constitutively ‘activated’ MC, as might be found in an inflammatory state. Therefore, it can be used to study the downregulation of their function.11

The critical modulatory role of adhesion in MC function can be demonstrated by the fact that, in addition to enhancement by cytokines, antigen-induced histamine secretion can be potentiated when RBL-2H3 cells are bound to FN. This increase in antigen stimulated histamine release upon adhesion was blocked by antibodies to VLA-4, 5, and VNR integrins, as well as by peptide competitors for the integrin binding sites on FN.11 Using the MC line MCP5/l, it has also been shown that antigenic stimulation, using antigen concentrations which were insufficient to cause significant mediator release, markedly enhanced FN binding.15 Thus, the interaction with extracellular matrix components is a determinant of cellular phenotype, and it is important to understand its regulation and its effects upon the MC.

These results suggest that activation of MC in vivo affects their ability to bind to the extracellular matrix, and conversely, binding to the extracellular matrix influences MC secretion of mediators. The ability of FN to potentiate histamine secretion, while IFN-γ downregulates histamine secretion, suggests that FN and IFN-γ impinge on the same signalling pathways in an antagonistic/opposite fashion. Thus, we postulated that IFN-γ also downregulates the interaction of MC with FN. To test this hypothesis, we have used an adhesion assay to monitor the ability of RBL-2H3 cells to bind to FN following IFN-γ treatment. Understanding the ability of IFN-γ to modulate multiple MC activation pathways will to contribute to the knowledge of MC regulation and offer mechanisms to modify their function in inflammatory disease.

MATERIALS AND METHODS

Cell culture

RBL-2H3 (provided by Dr Mark Swieter, NIH, Rockville, MD) cells were cultured in Eagle’s minimum essential medium (Flow Laboratories Inc., McLean, VA) containing l-glutamine pH 7·2, supplemented with 15% heat-inactivated fetal bovine serum (FBS), and 40 U/ml penicillin/streptomycin (Gibco BRL, Grand Island, NY). All cell cultures were confluent prior to being harvested for use in experiments. To prepare cells for assays, they were plated in 12-well flat-bottomed tissue culture plates (Corning Costar Corp., Kennebunk, ME) at 5 × 105 cells in 1 ml, to be treated with recombinant rat IFN-γ (Gibco BRL, 4 × 106 U/mg), Nω-nitro-l-arginine methyl ester (L-NAME) (Sigma, St Louis, MO), vanadate (Sigma), genistein (Calbiochem, La Jolla, CA), sodium nitroprusside (Sigma), S-nitrosoglutathione (Molecular Probes, Eugene, OR), or 1H-(1,2,4)oxadiazolo[4,3-a]quinoxalin-1-one (ODQ) (Tocris Cookson Inc., Ballwin MO) as indicated. The treatment time was 24 hr at 37°, unless otherwise indicated. Cells were washed (0·8% NaCl, 0·04% KCl, 0·035% NaHCO3, 0·02% ethylenediamine tetra-acetic acid (EDTA)), and then adherent cells were removed from tissue culture flasks using trypsin EDTA (0·05% trypsin, 0·53 nm EDTA, Gibco BRL). The cells were then washed with media to remove excess trypsin prior to use.

Adhesion assay

Linbro (Flow Laboratories, Inc. McLean, VA) 96-well flat-bottomed plates were coated with rat plasma FN (Sigma or Gibco BRL FN behaved identically), or FN fragments (Gibco, α-chymotryptic 40 000 MW fragment, α-chymotryptic 120 000 MW fragment) at 10 μg/ml in phosi-buffered saline (PBS) pH 7·2, 37° for 2 hr prior to the experiment. The FN concentration was optimized such that the plate was saturated, as indicated by observations that blocking with 3% bovine serum albumin (BSA; Sigma, fraction V) did not affect binding of the RBL-2H3 cells to FN. Thus, no blocking step with BSA was required. The plate was washed three times with binding buffer (RPMI media (Gibco BRL)) supplemented with 0·03% BSA using a Nunc-Immuno Wash 8 (Canadian Life Technologies Inc. Burlington, Ontario, Canada). For the assay, the RBL-2H3 cells were removed from the 12-well plate, washed once with binding buffer, and then labelled with the acetoxymethyl ester derivative of 2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxy-fluorescein (BCECF-AM) (Molecular Probes) in binding buffer at 50 μg/6 × 106 cells/1·2 ml for 30 min at 37°. The cells were then washed twice with binding buffer, and resuspended at 5 × 105 cells/ml in binding buffer. BCECF is fluorescent only once it enters the cell and is hydrolysed by cell esterases. The fluorescence measured is proportional to the number of cells in each well. The cells were plated on the FN at 5 × 104 cells/well, incubated for 1 hr at 37°, and then washed twice with binding buffer using the Nunc-Immuno Wash. Control wells that were not washed were used to determine the fluorescence of the total cells added. Percentage binding was calculated as:

graphic file with name imm0097-0481-m1.jpg

Percent inhibition is calculated as:

graphic file with name imm0097-0481-m2.jpg

Total fluorescence for differently treated cells was compared with cell counts to ensure that the various treatments did not affect the extent of BCECF labelling or the fluorescence properties of the BCECF. Fluorescence was determined using a Millipore Cytofluor 2350 plate reader.

Flow cytometry

RBL-2H3 cells were incubated with 10% normal mouse serum for 1 hr at 4° to block Fc-mediated binding to their surface, then labelled with fluorescein isothiocyanate (FITC) conjugated antibodies (Pharmingen, San Diego, CA) against rat VLA-4 α-chain CD49d (MRα4) at 50 μg/ml, rat VLA-5 α-chain CD49e (HMα5-1) at 50 μg/ml, or the rat VNR α-chain CD61 (F11) at 30 μg/ml, for 1 hr at 4°. The cells were fixed with 1% formalin, and then analysed on a FACScan (Becton Dickinson, Mountain View, CA) using Lysis II software. Non-specific binding was determined using concentration matched FITC-labelled isotype controls immunoglobulin G2bκ (IgG2bκ), hamster IgG (Pharmingen), and IgG1 (Southern Biotechnology Associates Inc., Birmingham, AL), respectively. The fluorescence of the cells treated with the isotype control was subtracted from the fluorescence of the cells treated with the integrin antibody to determine the percentage of cells expressing each integrin epitope.

Fluorescent labelling of the actin cytoskeleton

RBL-2H3 cells were treated for 24 hr at 37° with IFN-γ, while adhering to FN-coated coverslips, or pretreated for 24 hr with IFN-γ, then allowed to bind to FN for 2 hr. These cells were then fixed in 10% formaldehyde, and labelled with Texas red phalloidin to stain the actin cytoskeleton. A Reichert polyvar 2 fluorescent microscope was used with a 100 × apochromatic lens, NA 1.32, and photographs taken using Kodak Plus-X pan 125 film.

Statistics

Analysis of variance was used, followed by Student’s t-test, using paired means to determine the statistical significance of the results. When P < 0·05, * is indicated over the data point, while when P < 0·01 ** is used. The mean ± SE of the mean for at least three experiments is shown unless otherwise indicated.

RESULTS

RBL-2H3 cells spontaneously bind to FN.10 To determine the optimum coating concentration of FN on tissue culture plates under our conditions, dilutions of FN were used and the effect of a subsequent blocking step with 3% BSA was tested (data not shown). At a coating concentration of 10 μg/ml, binding of RBL-2H3 cells to FN plateaus, and a blocking step was not required to prevent non-specific binding. Therefore, 10 μg/ml of FN was used for all subsequent experiments. This resulted in 58·3±7·3% of the cells binding to the FN in the absence of any treatment.

IFN-γ inhibits RBL-2H3 binding to fibronectin in a dose-and time-dependent fashion

To determine if IFN-γ could modulate adhesion of RBL-2H3 cells to FN, cells were pretreated for 24 hr with IFN-γ at 20–1000 U/ml (Fig. 1). Treatment with IFN-γ decreased the percentage of cells bound to FN by up to 34·6±3·8% at 500 U/ml IFN-γ. This inhibition of binding occurs in a dose-dependent fashion, and is significant at 100 U/ml, while for maximum effect 500–1000 U/ml is used. Therefore, 1000 U/ml IFN-γ was used in subsequent experiments. The effect of treatment time upon the ability of IFN-γ to inhibit binding of RBL-2H3 cells to FN was determined following various periods of culture with IFN-γ (data not shown). While evidence of inhibition existed by 12 hr of treatment, the decrease in binding became statistically significant (P < 0·05) at 20 hr (18·6±4·1%), and maximal effect was observed at 48 hr of treatment (27·2±5·2%). This extended time period required for incubation suggests multiple events and new protein synthesis may be involved to change the phenotype of the cell in its adhesion to FN. In addition, the amount of inhibition increased with time of treatment.

Figure 1.

Figure 1

Open in a new tab

Dose response for IFN-γ effects on adhesion of RBL-2H3 cells to fibronectin. RBL-2H3 cells were treated with IFN-γ for 24 hr, at 37°, then tested for their ability to bind to fibronectin over 1 hr. Results are expressed as the percentage inhibition of binding of treated cells compared to untreated cells at each dose. The statistics were calculated using a paired mean t-test based on percentage binding so they could be compared to untreated control, although percentage inhibition is shown in the graph. n = 4, *P < 0·05, and **P < 0·01.

Surface expression of FN adherent integrins is not altered by IFN-γ

One possible mechanism for the inhibition by IFN-γ of adhesion of RBL-2H3 cells to FN could be through downregulation of the number of integrins on the cell surface. To study this, antibodies to three integrins known to interact with FN including VLA-4, VLA-5, and VNR, were used in flow cytometry to investigate cells treated with or without IFN-γ for 24 hr (Fig. 2). The expression of these integrins on the surface of the cell did not differ between the two treatments. As a control, the same group of cells were tested for their ability to bind to FN, and they demonstrated inhibition of binding following IFN-γ treatment despite no change in their expression of VLA-4, VLA-5, or VNR. This indicates that IFN-γ does not change the number of integrins on the surface, and thus, IFN-γ must alter the ability of these integrins to bind FN or signal the cell to modulate adhesive behavior.

Figure 2.

Figure 2

Open in a new tab

Integrin expression on the surface of RBL-2H3 cells is not altered by IFN-γ. Cells treated for 24 hr with 1000 U/ml IFN-γ were compared to sham-treated cells for their ability to bind antibodies for specific integrins on the surface of the cell. Following incubation with antibody, cells were fixed, and flow cytometric analysis performed to determine the level of expression of (a) VLA-4, (b) VLA-5, and (c) VNR on the surface of the cells. Black solid line: no IFN-γ treatment, Grey solid line: IFN-γ-treated cells, and dotted line: isotype control; results of a single experiment representative of three separate experiments.

Actin cytoskeleton structure is not altered by IFN-γ

Having determined that the phenotype of the cell had changed following IFN-γ treatment, resulting in decreased adhesion, we investigated the actin cytoskeleton of the cell to determine if its arrangement or actions had been altered to account for the inhibition. RBL-2H3 were treated with 1000 U/ml IFN-γ for 24 hr on FN, or pretreated for 24 hr with IFN-γ, and then allowed to bind to FN for 2 hr. They were then stained with Texas red phalloidin. These time periods were chosen to allow maximum binding of the cells. The cells in this assay would contain the population with decreased adhesion caused by IFN-γ, because, in the adhesion assay, only cells adherent to the tissue culture plate were harvested for study, and this population demonstrated a decrease in adhesion because of IFN-γ treatment. Therefore, the cell population used in the phalloidin staining is representative of the cells observed in the adhesion assay. Fluorescent labelling of actin, however, shows no obvious differences in cell actin architecture following IFN-γ treatment (Fig. 3). Stress fibres were not observed, and interference reflection microscopy was unable to detect actin localized into focal adhesions except at what appears to be the leading edge of the cell (data not shown). No difference was observed between IFN-γ-treated and sham-treated cells. These photographs demonstrate that RBL-2H3 cells are highly variable in morphology and size, which is commonly observed in our cultures.

Figure 3.

Figure 3

Open in a new tab

Actin cytoskeletal structure is not altered by treatment with IFN-γ. RBL-2H3 cells treated with 1000 U/ml IFN-γ for 24 hr at 37°, were compared to untreated cells, for changes in actin cytoskeleton structure upon adherence to fibronectin. Cells were fixed, labelled with Texas red phalloidin, and observed by fluorescence microscopy. (a) no IFN-γ, 2 hr adhesion on fibronectin; (b) 1000 U/ml IFN-γ, 2 hr adhesion; (c) no IFN-γ, 24 hr adhesion; (d) 1000 U/ml IFN-γ, 24 hr adhesion. Actin filament organization in RBL-2H3 cells did not visibly alter following IFN-γ treatment (24 hr) when the cells adhered to fibronectin for 2 hr or 24 hr. Scale bars represent 20 μm.

IFN-γ downregulates multiple fibronectin interactions

Because flow cytometry indicates that RBL-2H3 cells express more than one integrin capable of binding to FN, we studied if the IFN-γ-induced inhibition of adhesion could be assigned to a specific integrin. The binding sites for the integrins VLA-4 and VLA-5 on FN have been identified, and can be distinguished using chymotryptic fragments of FN. Previous literature has demonstrated that the 40 kDa chymotryptic fragment contains the CS-1 domain that binds VLA-4,16 while the 120 000 MW chymotryptic fragment contains the RGD site, which binds to VLA-5·17 This RGD site is also where VNR (αvβ3) binds.18 We used these fragments to dissect the inhibitory effect of IFN-γ upon VLA-4 versus VLA-5 function. IFN-γ is able to decrease binding of RBL-2H3 cells to intact FN by 31·8±9·4%, to the 120 000 MW fragment by 38·5±0·6%, and to the 40 000 MW fragment by 32·0±3·8%. There is no significant difference (P < 0·05) between inhibition of binding to the intact protein compared to either fragment (data not shown). This indicates that the binding to both of these sites on FN is affected by IFN-γ treatment of RBL-2H3 cells, and thus, adhesion mediated by both VLA-4 and VLA-5 is modulated by the IFN-γ treatment.

Tyrosine phosphatase inhibitor prevents IFN-γ-mediated inhibition of adhesion

To determine the requirement of tyrosine phosphorylation in IFN-γ inhibition, RBL-2H3 cells were coincubated with IFN-γ and the broad spectrum tyrosine kinase inhibitor genistein (Fig. 4). Genistein potentiated the effect of IFN-γ by 57±9% at 1 μg/ml genistein. This effect could be a direct effect upon the IFN-γ pathway, or on a separate pathway in which tyrosine phosphorylation inhibits adhesion. To further explore this, we used a tyrosine phosphatase inhibitor, vanadate, to determine its effect on the action of IFN-γ. Vanadate (100 μm) reduced the inhibitory effect of IFN-γ by 62% from a value of 28·0±6·7% in the absence of vanadate, to 10·6±5·3% (Fig. 4). This indicates that tyrosine phosphatase activity is a necessary step in the ability of IFN-γ to effect its modulation of adhesion. Dose responses were performed for both genistein (0–10 μg/ml) and vanadate (0–1 mm), and the doses resulting in the optimal effect on IFN-γ inhibition without any deleterious effects on cell viability, as determined by either trypan blue exclusion or the ability to hydrolyse BCECF-AM, are presented here.

Figure 4.

Figure 4

Open in a new tab

Tyrosine kinase inhibition enhances the inhibitory effect of IFN-γ on binding of RBL-2H3 cells to fibronectin, while tyrosine phosphatase inhibition abrogates the inhibitory effect of IFN-γ. RBL-2H3 cells were treated for 24 hr at 37° with 1000 U/ml IFN-γ alone, or with IFN-γ and either 1 μg/ml (3·7 μm) genistein (n = 3) or 100 μm vanadate (n = 3). They were then tested for their adherence following 1 hr incubation on fibronectin. Results are expressed as the percentage inhibition of binding. *P < 0·05, **P < 0·01 compared to inhibition observed with IFN-γ alone.

Inhibition of nitric oxide synthase abrogates the inhibitory effect of IFN-γ on adhesion

Because IFN-γ induces the expression of nitric oxide synthase (NOS) in other cell types, and NO can modulate adhesion in other systems we investigated if NO production was involved in the IFN-γ modulation of adhesion. If NOS production of NO was required for the IFN-γ effect, a NOS inhibitor coincubated with IFN-γ should prevent the inhibition. L-NAME (500 μm) reduced by 70% IFN-γ-mediated decreased adhesion of RBL-2H3 cells to FN, from an inhibition of 37·4±3·8% in the absence of L-NAME, to 11·3±4·4% in the presence of L-NAME (Fig. 5). Thus, IFN-γ requires NOS activity for its function. L-NAME, in the absence of IFN-γ had no significant effect on binding of RBL-2H3 cells to FN. Because L-NAME is a competitive inhibitor of arginine, it is possible that some NO is still being produced explaining the incomplete reversal of the IFN-γ effect.

Figure 5.

Figure 5

Open in a new tab

Nitric oxide synthase inhibitor interferes with the inhibitory effect of IFN-γ on RBL-2H3 adhesion to fibronectin. RBL-2H3 cells were treated for 24 hr at 37° with different concentrations of L-NAME with or without 1000 U/ml IFN-γ. They were then tested for their ability to adhere following a 1-hr incubation on fibronectin. Results are expressed as the IFN-γ inhibition at each dose of L-NAME. n = 4, *P < 0·05.

NO donors mimic the inhibitory effect of IFN-γ on adhesion

To confirm that NO is an effector molecule in IFN-γ inhibition, the effect of the NO donors sodium nitroprusside and S-nitrosoglutathione were investigated. If IFN-γ is acting through NO production, an NO donor should be able to produce the same effect. Incubation of the RBL-2H3 cells with sodium nitroprusside or S-nitrosoglutathione for 24 hr decreased adhesion to FN by 23·2%, and 28·6%, respectively, at 500 μm (Fig. 6). Thus, NO is able to inhibit adhesion of RBL-2H3 cells to FN, supporting the idea that the mechanism of IFN-γ inhibition is through stimulation of NO production. Dose responses were performed for both sodium nitroprusside (0–1 mm) and S-nitrosoglutathione (0–1 mm), and the doses resulting in the optimal effect binding without any deleterious effects on cell viability, as determined by either trypan blue exclusion or the ability to hydrolyse BCECF-AM, are presented here.

Figure 6.

Figure 6

Open in a new tab

Nitric oxide donors are able to mimic the inhibitor effect of IFN-γ on RBL-2H3 adhesion to fibronectin. Sodium nitroprusside (n = 5), and S-nitrosoglutathione (n = 3) (0 and 500 μm) were used to treat RBL-2H3 cells for 24 hr at 37°. They cells were then challenged for their ability to bind to fibronectin during a 1-hr incubation. Both NO donors inhibited binding of the cells to fibronectin in a fashion similar to that produced by IFN-γ. *P < 0·05, and **P < 0·01.

DISCUSSION

MC play an integral role in inflammation, responding to, and contributing to the extracellular milieu. The release of inflammatory mediators from MC is able to damage surrounding tissue in inflammatory disease such as asthma. IFN-γ is a cytokine that is mainly produced by NK and T cells (reviewed in 21), which affects a variety of MC functions including: histamine release and TNF-α mediated cytotoxicity,5 serotonin release,6 differentiation/proliferation, and major histocompatibility complex (MHC) class II expression.24 In cultured human MC, IFN-γ has been shown to promote survival and histamine release, which is in contrast to the results observed in rodent systems, and emphasizes the need to understand how this cytokine is functioning.25

Integrin interaction, which initially was thought to play a mainly adhesive role, is now known to modulate such diverse activities as cell-cycle regulation, apoptosis, differentiation, mediator production and secretion, in addition to cell spreading, migration, and anchorage within specific tissues. This is accomplished through signalling into the cell involving src family tyrosine kinases and focal adhesion kinase, serine/threonine kinases including protein kinase C, phosphatases, intracellular pH, Ca2+ concentration, and activation of inositol lipid and arachidonic acid metabolism (reviewed in 26 and 27) Much attention is currently focused upon unravelling the pathways involved in this signalling process.

This is the first report of a cytokine that decreases MC adhesion to extracellular matrix proteins and demonstrates that IFN-γ downregulates rat MC adhesion to FN in a dose- and time-dependent manner. The inhibition of adhesion does not involve a decrease in the expression of integrins on the cell surface, as shown by flow cytometry, indicating that IFN-γ is instead altering the signalling pathways that the integrins use for strong adhesion. IFN-γ does not directly decrease adhesion, but rather requires 24 hr to establish a new cell phenotype, presumably through either new protein synthesis, or degradation of existing proteins. That this change is mediated through more than one FN binding site, involving multiple integrins, was shown using chymotryptic fragments of FN. Thus, the mechanism of inhibition is not by affecting the function of a single integrin but rather through a mechanism that is able to affect general adhesion competency. We demonstrate that the mode of action of IFN-γ includes phosphatase activity and production of NO, both of which may impinge on both secretion and adhesion pathways. Many effects of NO are known to be produced by its stimulation of cGMP production, and ODQ is able inhibit this activity.28 In our system, ODQ was able to reduce IFN-γ-mediated inhibition of RBL-2H3 binding to FN from 31·4±4·9% in the absence of ODQ, to 13·0±3·7% in the presence of 50 μm ODQ (data not shown). This suggests NO inhibits adhesion through a cyclic guanosine monophospahte (cGMP)-dependent mechanism. Basal binding to FN was also reduced by ODQ, however, indicating further studies are required to understand the mechanism of action of ODQ in this system. Vanadate, L-NAME, and ODQ interfere with IFN-γ’s inhibitory effect by similar amounts (62, 70, and 58·5%, respectively) indicating they may be acting upon the same pathway. Future experiments will delineate whether these inhibitory activities overlap or are complementary, and further explore the ability of IFN-γ to utilize NO/cGMP to downregulate MC function.

There are multiple points at which the integrin, antigen activation, and growth factor signalling pathways use common signalling molecules and cross talk may occur. Tyrosine kinases, the rat sarcoma family of GI Pases, mitogen-activated protein kinase (ras/MAPK) signalling pathway, and regulation of inositol lipid metabolism, can all be activated following integrin activation, Fc receptor stimulation, or growth factor signalling.29–32 Also, Hamawy et al.33 demonstrated the phosphorylation of paxillin, a cytoskeletal protein associated with focal adhesions, in response to FcεRI aggregation in RBL-2H3 cells. It is evident from the potentiation of histamine release by FN, and adhesion by FcεRI stimulation, that the integrin signalling pathway and the FcεRI activation pathway can potentiate one another in MC. This is supported by the enhancement of integrin stimulated tyrosine phosphorylation of focal adhesion kinase, pp105-115, and paxillin following costimulation with FcεRI. In addition, localization of vinculin and talin into discrete focal adhesion-like structures following stimulation with a variety of activators of secretion in RBL-2H3 cells indicates a strong link between secretion and adhesion.36 That IFN-γ can regulate both of these pathways suggests its point of regulation may involve one of these cross-talk points.

There is increasing evidence that NO is produced in MC, and plays an autoregulatory role.4,37–41 IFN-γ induces inducible nitric oxide synthase (iNOS) production in macrophages, and NO released from macrophages can modulate MC mediator release.41 However, little is known about the regulation of NO production in the MC and its autocrine effects. Previous reports with connective tissue and activated MC subtypes document NO production detected in the cell supernatant as nitrite. We attempted to find nitrite in the supernatant of RBL-2H3 cells, both treated with IFN-γ and untreated, but were unable to detect it under our conditions by the Griess reaction or high pressure liquid chromatography (HPLC). This suggests that the NO being produced is in small amounts, and used within the cell for autoregulatory purposes such as nitrosylation of proteins. In addition, RBL-2H3 are a mucosal-type mast cell, and the lack of secreted nitrite compared to peritoneal MC may reflect functional differences between the subtypes of MC.

In order to increase NO production in RBL-2H3 cells, IFN-γ must regulate one of the isoforms of NOS. We have preliminary polymerase chain reaction (PCR) evidence of endothelial nitric oxide synthase (eNOS), but not iNOS message detected using mRNA from the RBL-2H3 cells (data not shown). Previous reports showed preliminary evidence of NOS isoforms in MC by immunohistochemistry. Also iNOS mRNA has been found in mouse mastocytoma where it mediated apoptosis,45 and in mouse bone marrow mast cells where NO was released following FcεRI crosslinking.42 These same authors have also shown downregulation of MC function by NO,46 making it interesting to speculate upon a negative feedback role for NO. This negative feedback would also explain reports in the literature of histamine upregulating NO production.47 Thus, endogenous NO may depress MC activation minimizing deleterious effects to surrounding tissue. NO may be induced by cytokines before activation to suppress mediator release upon stimulation, or produced following FcεR1 crosslinking to return the MC to a resting state.

Most exciting is the possibility of understanding the processes by which NO regulates MC function. Downregulation of adhesion of chondrocytes to FN by NO results from disruption of focal adhesions connected to the integrin clustering.48 Interestingly, this did not prevent clustering of the integrin itself, in agreement with our results that the number of integrins on the surface is not affected by this signalling pathway. T-cell proliferation has also shown to be down-regulated by NO, with a concomitant decrease in phosphorylation of the signalling proteins Janus kinase 3 (Jak3) and signal transducer and activator of transcription 5 (STAT5). It will be of interest to determine if these same processes are used by NO to inhibit MC adhesion. Thus, the next step is the identification of the mechanism by which NO exerts its effect: how it simultaneously affects growth, mediator secretion, and adhesion in MC.

Acknowledgments

We would like to thank Dr D. Begg for helping us with the fluorescent actin labelling ad fluorescence microscopy, M. Ridyard and Dr E. Sanders for help with interference reflection microscopy, and Drs M. Muscara and J. Wallace for HPLC analysis for nitrite. This work was supported by the Medical Research Council of Canada. F.L.W. is the recipient of an Alberta Heritage Foundation of Medical Research Fellowship.

Glossary

Abbreviations

FN

fibronectin

MC

mast cell

SCF

stem-cell factor

VNR

vitronectin receptor

L-NAME

Nω-nitro-l-arginine methyl ester

BCECF-AM

acetoxymethyl ester derivative of 2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxy-fluorescein

ODQ

1H-(1,2,4)oxadiazolo [4,3-a]quinoxalin-1-one

REFERENCES

  • 1.Coleman JW, Holliday MR, Kimber I, Zsebo KM, Galli SJ. Regulation of mouse peritoneal mast cell secretory function by stem cell factor, IL-3 or IL-4. J Immunol. 1993;150:556. [PubMed] [Google Scholar]
  • 2.Holliday MR, Banks EMS, Dearman RJ, Kimber I, Coleman JW. Interactions of IFN-γ with IL-3 and IL-4 in the regulation of serotonin and arachidonate release from mouse peritoneal mast cells. Immunology. 1994;82:70. [PMC free article] [PubMed] [Google Scholar]
  • 3.Lin T-J, Bissonnette EY, Hirsh A, Befus AD. Stem cell factor potentiates histamine secretion by multiple mechanisms, but does not affect tumour necrosis factor-α release from rat mast cells. Immunology. 1996;89:301. doi: 10.1046/j.1365-2567.1996.d01-733.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Lin T-J, Befus AD. Differential regulation of mast cell function by IL-10 and stem cell factor. J Immunol. 1997;159:4015. [PubMed] [Google Scholar]
  • 5.Bissonnette EY, Chin B, Befus AD. Interferons differentially regulate histamine and TNF-α in rat intestinal mucosal mast cells. Immunology. 1995;86:12. [PMC free article] [PubMed] [Google Scholar]
  • 6.Coleman JW, Buckley MG, Holliday MR, Morris AG. Interferon-γ inhibits serotonin release from mouse peritoneal mast cells. Eur J Immunol. 1991;21:2559. doi: 10.1002/eji.1830211037. [DOI] [PubMed] [Google Scholar]
  • 7.Bissonnette EY, Befus AD. Inhibition of mast cell-mediated cytotoxicity by IFN-α/β and -γ. J Immunol. 1990;145:3385. [PubMed] [Google Scholar]
  • 8.Bissonnette EY, Enciso JA, Befus AD. TGF-β1 inhibits histamine and tumor necrosis factor-α release by rat peritoneal mast cells through an autocrine pathway. Am J Respir Cell Mol Biol. 1997;16:275. doi: 10.1165/ajrcmb.16.3.9070612. [DOI] [PubMed] [Google Scholar]
  • 9.Bissonnette EY, Enciso JA, Befus AD. Inhibition of tumour necrosis factor-α (TNF-α) release from mast cells by the anti-inflammatory drugs, sodium cromoglycate and nedocromil sodium. Clin Exp Immunol. 1995;102:78. doi: 10.1111/j.1365-2249.1995.tb06639.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Hamawy MM, Oliver C, Mergenhagen SE, Siraganian RP. Adherence of rat basophilic leukemia (RBL-2H3) cells to fibronectin-coated surfaces enhances secretion. J Immunol. 1992;149:615. [PubMed] [Google Scholar]
  • 11.Ra C, Yasuda M, Yagita H, Okumura K. Fibronectin receptor integrins are involved in mast cell activation. J Allergy Clin Immunol. 1994;94:625. doi: 10.1016/0091-6749(94)90139-2. [DOI] [PubMed] [Google Scholar]
  • 12.Dastych J, Metcalfe DD. Stem cell factor induces mast cell adhesion to fibronectin. J Immunol. 1994;152:213. [PubMed] [Google Scholar]
  • 13.Ra C, Yasuda M, Kim Z. Fibronectin receptor (FNR) integrins on mast cells are involved in cellular activation. In: Kitamura Y, Yamamoto S, Galli S J, Greaves M W, et al., editors. Biological and Molecular Aspects of Mast Cell and Basophil Differentiation and Function. Vol. 239. New York: Raven Press; 1995. [Google Scholar]
  • 14.Dastych J, Hardison MC, Metcalfe DD. Aggregation of low affinity IgG receptors induces mast cell adherence to fibronectin. Requirement for the common FcR γ-chain. J Immunol. 1997;158:1803. [PubMed] [Google Scholar]
  • 15.Wyczolkowska J, Dastych J, Slusarczyk A, Kolago B. Relations between FcεRI crosslinking-induced mast cell activation and adhesion to fibronectin. J Physiol Pharmacol. 1994;45:501. [PubMed] [Google Scholar]
  • 16.Garcia-Pardo A, Wayner EA, Carter WG, Ferreira OC. Human B lymphocytes define an alternative mechanism of adhesion to fibronectin. J Immunol. 1990;144:3361. [PubMed] [Google Scholar]
  • 17.Ginsberg M, Pierschbacher MD, Ruoslahti E, Marguerie G, Plow E. Inhibition of fibronectin binding to platelets by proteolytic fragments and synthetic peptides which support fibroblast adhesion. J Biol Chem. 1985;260:3931. [PubMed] [Google Scholar]
  • 18.Charo IF, Nannizzi L, Smith JW, Cheresh DA. The VNR alpha v beta 3 binds fibronectin and acts in concert with alpha 5 beta 1 in promoting cellular attachment and spreading on fibronectin. J Cell Biol. 1990;111:2795. doi: 10.1083/jcb.111.6.2795. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Radomski MW, Vallance P, Whitley G, Foxwell N, Moncada S. Platelet adhesion to human vascular endothelium is modulated by constitutive and cytokine induced nitric oxide. Cardiovasc Res. 1993;27:1380. doi: 10.1093/cvr/27.7.1380. [DOI] [PubMed] [Google Scholar]
  • 20.Gaboury J, Woodman RC, Granger DN, Reinhardt P, Kubes P. Nitric oxide prevents leukocyte adherence: role of superoxide. Am J Physiol. 1993;265:H862. doi: 10.1152/ajpheart.1993.265.3.H862. [DOI] [PubMed] [Google Scholar]
  • 21.Boehm U, Klamp T, Groot M, Howard JC. Cellular responses to interferon-γ. Annu Rev Immunol. 1997;15:749. doi: 10.1146/annurev.immunol.15.1.749. [DOI] [PubMed] [Google Scholar]
  • 22.Nafziger J, Arock M, Guillosson J-J, Wietzerbin J. Specific high-affinity receptors for Interferon-γ on mouse bone marrow-derived mast cells: inhibitory effect of Interferon-γ on mast cell precursors. Eur J Immunol. 1990;20:113. doi: 10.1002/eji.1830200117. [DOI] [PubMed] [Google Scholar]
  • 23.Hu Z-Q, Zenda N, Shimamura T. Down-regulation by IL-4 and up-regulation by IFN-γ of mast cell induction from mouse spleen cells. J Immunol. 1996;156:3925. [PubMed] [Google Scholar]
  • 24.Warbrick EV, Taylor AM, Botchkarev VA, Coleman JW. Rat connective tissue type mast cells express MHC class II: up-regulation by IFN-γ. Cell Immunol. 1995;163:222. doi: 10.1006/cimm.1995.1120. [DOI] [PubMed] [Google Scholar]
  • 25.Yanagida M, Fukamachi H, Takei M, et al. Interferon-γ promotes the survival and FcεRI-mediated histamine release in cultured human mast cells. Immunology. 1996;89:547. doi: 10.1046/j.1365-2567.1996.d01-768.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Schwartz MA, Schaller MD, Ginsberg MH. Integrins: emerging paradigms of signal transduction. Annu Rev Cell Dev Biol. 1995;11:549. doi: 10.1146/annurev.cb.11.110195.003001. [DOI] [PubMed] [Google Scholar]
  • 27.Burridge K, Chrzanowska-Wodnicka M. Focal adhesions, contractility, and signaling. Annu Rev Cell Dev Biol. 1996;12:463. doi: 10.1146/annurev.cellbio.12.1.463. [DOI] [PubMed] [Google Scholar]
  • 28.Moro MA, Russel RJ, Cellek S, et al. cGMP mediates the vascular and platelet actions of nitric oxide: confirmation using an inhibitor of the soluble guanylyl cyclase. Proc Natl Acad Sci USA. 1996;93:1480. doi: 10.1073/pnas.93.4.1480. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Plopper GE, McNamee HP, Dike LE, Bojanowski K, Ingber DE. Convergence of integrin and growth factor receptor signaling pathways within the focal adhesion complex. Mol Biol Cell. 1995;6:1349. doi: 10.1091/mbc.6.10.1349. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Rosales C, O’brien V, Kornberg L, Juliano R. Signal transduction by cell adhesion receptors. Biochim Biophys Acta. 1995;1242:77. doi: 10.1016/0304-419x(95)00005-z. [DOI] [PubMed] [Google Scholar]
  • 31.Beaven MA, Baumgartner RA. Downstream signals initiated in mast cells by FcεRI and other receptors. Curr Opin Immunol. 1996;8:766. doi: 10.1016/s0952-7915(96)80002-1. [DOI] [PubMed] [Google Scholar]
  • 32.Miyamoto S, Teramoto H, Silvio Gutkind J, Yamada KM. Integrins can collaborate with growth factors for phosphorylation of receptor tyrosine kinases and MAP kinase activation: roles of integrin aggregation and occupancy of receptors. J Cell Biol. 1996;135:1633. doi: 10.1083/jcb.135.6.1633. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Hamawy MM, Swaim WD, Minoguchi K, de Feijter AW, Mergenhagen SE, Siraganian RP. The aggregation of the high affinity IgE receptor induces tyrosine phosphorylation of paxillin, a focal adhesion protein. J Immunol. 1994;153:4655. [PubMed] [Google Scholar]
  • 34.Hamawy MM, Mergenhagen SE, Siraganian RP. Cell adherence to fibronectin and the aggregation of the high affinity immunoglobulin E receptor synergistically regulate tyrosine phosphorylation of 105-kDa proteins. J Biol Chem. 1993;268:5227. [PubMed] [Google Scholar]
  • 35.Hamawy MM, Mergenhagen S, Siraganian RP. Tyrosine phosphorylation of pp125FAK by the aggregation of high affinity immunoglobulin E receptors requires cell adherence. J Biol Chem. 1993;268:6851. [PubMed] [Google Scholar]
  • 36.Kawasugi K, French PW, Penny R, Ludowyke RI. Focal adhesion formation is associated with secretion of allergic mediators. Cell Motil Cytoskeleton. 1995;31:215. doi: 10.1002/cm.970310305. [DOI] [PubMed] [Google Scholar]
  • 37.Salvemini D, Masini E, Anggard E, Mannaioni PF, Vane J. Synthesis of a nitric oxide-like factor from l-arginine by rat serosal mast cells: stimulation of guanylate cyclase and inhibition of platelet aggregation. Biochem Biophys Res Commun. 1990;169:596. doi: 10.1016/0006-291x(90)90372-t. [DOI] [PubMed] [Google Scholar]
  • 38.Salvemini D, Masini E, Pistelli A, Mannaioni PF, Vane J. Nitric oxide: a regulatory mediator of mast cell reactivity. J Cardiovasc Pharmacol. 1991;17(Suppl. 3S):258. [Google Scholar]
  • 39.Bissonnette EY, Hogaboam CM, Wallace JL, Befus AD. Potentiation of TNF-α-mediated cytotoxicity of mast cells by their production of nitric oxide. J Immunol. 1991;147:3060. [PubMed] [Google Scholar]
  • 40.Hogaboam CM, Befus AD, Wallace JL. Modulation of rat mast cell reactivity by IL-1β: divergent effects on nitric oxide and platelet-activating factor release. J Immunol. 1993;151:3767. [PubMed] [Google Scholar]
  • 41.Eastmond NC, Banks EMS, Coleman JW. Nitric oxide inhibits IgE-mediated degranulation of mast cells and is the principal intermediate in IFN-γ induced suppression of exocytosis. J Immunol. 1997;159:1444. [PubMed] [Google Scholar]
  • 42.Bidri M, Ktorza S, Vouldoukis I, et al. Nitric oxide pathway is induced by FcεR1 and up-regulated by stem cell factor in mouse mast cells. Eur J Immunol. 1997;27:2907. doi: 10.1002/eji.1830271124. [DOI] [PubMed] [Google Scholar]
  • 43.Bacci S, Arbi-Riccardi R, Mayer B, Rumio C, Borghi-Cirri MB. Localization of nitric oxide synthase immunoreactivity in mast cells of human nasal mucosa. Histochemistry. 1994;102:89. doi: 10.1007/BF00269011. [DOI] [PubMed] [Google Scholar]
  • 44.Shimizu Y, Sakai M, Umemura Y, Ueda H. Immunohistochemical localization of nitric oxide synthase in normal human skin: expression of endothelial-type and inducible-type nitric oxide synthase in keratinocytes. Dermatology. 1997;24:80. doi: 10.1111/j.1346-8138.1997.tb02748.x. [DOI] [PubMed] [Google Scholar]
  • 45.Kitajima I, Kawahara K, Nakajima T, Soejima Y, Matsuyama T, Maruyama I. Nitric oxide-mediated apoptosis in murine mastocytoma. Biochem Biophys Res Commun. 1994;204:244. doi: 10.1006/bbrc.1994.2451. [DOI] [PubMed] [Google Scholar]
  • 46.Bidri M, Becherel P-A, Le Goff L, et al. Involvement of cyclic nucleotides in the immunomodulatory effects of nitric oxide on murine mast cells. Biochem Biophys Res Commun. 1995;210:507. doi: 10.1006/bbrc.1995.1689. [DOI] [PubMed] [Google Scholar]
  • 47.Mannaioni PF, Bello MG, Di Bello MG, et al. Interaction between histamine and nitric oxide in rat mast cells and in isolated guinea pig hearts. Int Arch Allergy Immunol. 1997;113:297. doi: 10.1159/000237580. [DOI] [PubMed] [Google Scholar]
  • 48.Clancy RM, Rediske J, Tang X, et al. Outside-in signaling in the chondrocyte: nitric oxide disrupts fibronectin-induced assembly of a subplasmalemmal actin/rho A/focal adhesion kinase signaling complex. J Clin Invest. 1997;100:1789. doi: 10.1172/JCI119706. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Bingisser RM, Tilbrook PA, Holt PG, Kees UR. Macrophage-derived nitric oxide regulates T cell activation via reversible disruption of the Jak3/STAT5 signaling pathway. J Immunol. 1998;160:5729. [PubMed] [Google Scholar]

Articles from Immunology are provided here courtesy of British Society for Immunology

RESOURCES