Digestive Diseases and Sciences, Vol. 36. No. 2 (February" 1991), pp. 179-184
Modulation of Mediator Release from Human Intestinal Mast Cells by Sulfasalazine and 5-Aminosalicylic Acid CHARITY C. FOX, MD, WENDY C. MOORE, BS, and LAWRENCE M. LICHTENSTEIN, MD
Intestinal mast cells are thought to contribute to the mucosal inflammation in ulcerative colitis and Crohn' s disease through release of inflammatory mediators. Since sulfasaIazine and its metabolite 5-aminosalicylic acid are effective therapeutic agents in inflammatory bowel disease and have been shown to inhibit generation o f inflammatory products in other cells, we examined the effect of these agents in vitro on human intestinal mast cell mediator release. Sulfasalazine (5 x 10 -4 -10 -3 M) was found to significant~ enhance goat anti-human lgE-inducedhistamine release from intestinal mast cells, which is the same response as seen in human blood basophils, whereas its metabolite 5-aminosalicylic acid was an effect&e inhibitor of stimulated histamine release in both mast cells and basophils. 5-Aminosalicylic acid also inhibited production of prostaglandin D 2 by the stimulated intestinal mast cells. Sulfasalazine alone, without immunologic stimulation, did not induce histamine release from mast cells or basophils, but the enhancement of ongoing mast cell activation by sulfasalazine may explain some cases o f adverse reactions to the drug. The inhibition of mast ceil histamine release and prostaglandin generation by 5-aminosalicylic acid demonstrates a potential therapeutic modality o f this agent. KEY WORDS: intestinal mast cells; basophils; histamine; prostaglandin D2; sulfasalazine; 5-aminosalicylic acid.
Mast cell hyperplasia is a consistent feature of the idiopathic inflammatory bowel diseases (IBD) ulcerative colitis and Crohn's disease (1-3). While mast cells often are found to be increased in inflammatory foci, the number of mast cells found in the mucosa and submucosa in IBD is impressive and has led investigators to speculate on the role these Manuscript received January 24, 1990; revised manuscript received August 31, 1990; accepted September 6, 1990. From the Johns Hopkins Asthma and Allergy Center, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21224. This work was supported by the National Foundation for Ileitis and Colitis and the National Institute of Allergy and Infectious Disease grants AI07290 and AI08270. Address for reprint requests: Dr. Charity C. Fox, Department of Medical Microbiology and Immunology, The Ohio State University, 5072 Graves Hall, 333 W. 10th Avenue, Columbus, Ohio 43210-1239.
inflammatory cells play in the pathology of IBD (1-4). It is certain that the mast cell, with its array of inflammatory mediatorsmincluding histamine, prostaglandins, sulfidopeptide leukotrienes, proteases, chemotactic factors, and cytokines--is capable of contributing significantly to the tissue response in IBD (5-7). W e hypothesized that, if mast cells were inciting inflammation in IBD, drugs that were effective in the treatment of IBD should inhibit the release of mast cell mediators of inflammation such as histamine and the products of arachidonic acid metabolism. Sulfasalazine and its active metabolite 5-aminosalicylic acid (5-ASA) both have been found to be useful in treatment of mild to moderate attacks of ulcerative colitis and as maintenance therapy to
Digestive Diseases and Sciences, Vol. 36, No. 2 (Februar." 1991)
0163-2116/9110200-0179506.50/0 9 1991PlenumPublishingCorporation
179
FOX ET AL p r e v e n t relapses (8). 5-ASA also has been reported to be effective in treatment of active Crohn's disease (9, 10). While the exact mechanisms by which these agents act in vivo are not known, it has been suggested that their therapeutic activities are due to effects on the generation of inflammatory mediators (11-14). There are reports on the actions of sulfasalazine and 5-ASA on arachidonic acid metabolism in several human cell types. It is known that sulfasalazine inhibits thromboxane synthetase in human platelets (11) and Nso blocks products of the 5-1ipoxygenase pathway in human neutrophils (12). In o t h e r experiments using whole colonic mucosa biopsies, sulfasalazine and 5-ASA have been shown to inhibit calcium ionophore A23187-induced generation of both lipoxygenase and cyclooxygenase products (13, 14). Testing the action of these drugs on human mast cells in vitro required the ability to isolate the mast cell from the tissue and to activate it. Using our m e t h o d for isolation of cells from human colonic m u c o s a (15), we produced single-cell suspensions of intestinal lamina propria cells, selectively stimulated the mast cells in these suspensions with goat anti-human IgE, and measured the mast cell specific p r o d u c t histamine as well as the inflammatory arachidonic acid metabolite prostaglandin D,. In this initial study, we chose to examine the action of 5-ASA on prostaglandin D2 (PGD) release since PGD is the predominant cyclooxygenase product of mast cells, and the intestinal mast cells make a larger quantity of this metabolite as compared to their predominant lipoxygenase product leukotriene C. T h e mast cells were preincubated with sulfasalazine and 5-ASA and subsequently challenged to test the effects of these drugs on the stimulated release of inflammatory mediators. Since sulfasalazine had been previously reported to enhance histamine release in human basophilmixed leukocyte preparations (16), we also studied the action of sulfasalazine and 5-ASA on human basophil histamine release. The basophil experiments functioned as controls for the mast cell experiments and, while the drugs would not necessarily have the same effects on both cell types, a similar basophil response would corroborate the mast cell response. MATERIALS AND METHODS T'.lssue Source. Human colonic mucosa specimens were obtained from routine surgical resections. Five resections
180
were performed for removal of adenocarcinoma, two for diverticulosis, and three for ulcerative colitis. The patients ranged in age from 19 to 73. There were four females and six males. None of the patients were currently taking the drugs studied. The patients with diverticulosis and adenocarcinoma received only the standard preoperative bowel preparations and sedatives. The inflammatory bowel disease patients were not on the drugs studied, but were maintained on low to moderate doses of prednisone, and received the standard bowel preparation and preoperative sedatives. In addition, one of the ulcerative colitis patients was maintained on hydrochlorothiazide and Tenormin, and another was on 1 g/day sulfasalazine that was changed on admission. Blood specimens for basophil histamine release were obtained from normal volunteers on no medications. Tissue and blood were obtained in protocols approved by the Joint Committee on Clinical Investigation. Isolation of Intestinal Mast Cells. Specimens of tissue 10 cm z were taken from grossly normal mucosa at least 4 cm from the diseased site. The mucosa was dissected carefully from the underlying muscularis and mechanically and enzymatically dispersed into a single-cell suspension as previously described (15). In brief, the strips of mucosa were chopped into 1 x l-cm pieces and incubated at 37~ C for 60 rain in calcium- and magnesium-free Hanks' balanced salt solution with 25 mM HEPES (CMF-HBSS/ HEPES) and 0.75 mM EDTA. The pieces were then washed with CMF-HBSS/HEPES, chopped further into 2 x 2-mm pieces and incubated in HBSS with 30 units/ml collagenase for 40 rain at 37~ C. At the end of the collagenase incubation, the supernatants were filtered through 150-p,m pore size Nitex cloth and the filtrate saved. The tissue then was incubated with pronase (1 I~g/g wet weight tissue) and chymopapain (0.5 I~g/g wet weight tissue) for two incubations of 30 rain each. After each incubation, the tissue was rinsed over Nitex with CMF-HBSS/HEPES and the filtrates saved. The filtrates from the collagenase and the pronase/chymopapain digestions were combined, centrifuged, and the cell pellet resuspended in buffer. The mast cells were counted using a low pH Alcian blue stain specific for mast cells and basophils (17), then centrifuged before placing in shortterm culture. Since basophils have been found to be rare in human intestinal mucosal preparations (15), all Alcian blue positive cells were counted as mast cells. Short-Term Culture. The cells were suspended in RPMI 1640 with 25 mM HEPES buffer (Gibco, Grand Island, New York), 2 mM L-glutamine, 100 izg/ml penicillin, and I00 p.g/ml gentamicin at a concentration of 2 x 106 cells/ml. The cell suspensions were incubated in a 75-cm2 tissue culture flask (Coming 25115) at 37~ in humidified 95% air and 5% COz. The cells were harvested after 12-18 hr, washed several times in PAG buffer (PIPES, 7.6 g/liter; NaCI, 6.4 g/liter; KCI, 0.37 g/liter; dextrose, 1.0 g/liter; human serum albumin, 3.0 mg/liter), pelleted, and recounted. Basophil Preparation. Basophil containing mixed leukocyte preparations were obtained from human whole blood as previously described (18). Blood was drawn and mixed with dextran-EDTA and allowed to sediment 90 min. The leukocyte-rich upper layer was drawn off and Digestive Diseases and Sciences. Vol. 36, No. 2 (February. 1991)
SULFASALAZINE, 5-ASA, AND INTESTINAL MAST CELLS centrifuged. The resultant pellet was washed twice in PAG and pelleted before adding final buffer for histamine release assay. Histamine Release Assay. Duplicate samples of 2 x 104 mast cells or basophils were challenged in PAGCM buffer (PAG buffer with CaCI_,-2H,O, 0.14 g/liter and MgC1, - 6H,O, 0.20 g/liter). Cells suspended in 0.3 ml of PAGCM were added to 0.1-ml volumes of buffer containing drug and preincubated for 10 rain. Sulfasalazine, sulfapyridine, and 5-aminosalicylic acid (Sigma, St. Louis, Missouri) were tested in concentrations ranging from 10-~ M to 10-3 M. At the end of the preincubation period, the mast cells or basophils in the cell suspension were selectively challenged with the stimulus--goat antihuman IgEnand incubated further for 20 rain (mast cells) or 30 rain (basophils). Supernatants were assayed for histamine content by automated fluorometric analysis (19). Results are expressed as percent of total histamine content as determined by lysis of cells in 2% perchloric acid and indicate specific stimulated histamine release since the background spontaneous release is subtracted from the total released histamine. Radioimmunoassay. Prostaglandin D,_ (PGD) was measured by radioimmunoassay (RIA) as previously described (20) in which the bound ligand is separated from free with dextran-coated charcoal. Supernatants from challenged cells were decanted and aliquots frozen at -20 ~ C. The PGD assay was performed on unextracted aliquots neat and at 1:5 dilution. The PGD antiserum had a 4% cross-reactivity with PGF,a and less than 1% with all other heterologous ligands (20). The sensitivity limit of the assay was 20 pg PGD/0.1 ml sample. Results are expressed as nanograms of immunoreactive PGD per 106 mast cells. Statistical Tests. Data for the histamine and prostaglandin release experiments were compared by paired twosample Student's t test and the Wilcoxon matched-pairs signed-rank test. Values of P < 0.05 were considered significant. RESULTS Sulfasalazine (10-7-10 -3 M) caused enhancement of goat anti-human IgE-induced release from human intestinal mast cells (Figure 1) that was statistically significant at 10-4-10 -3 M. This enhancement of stimulated histamine release was equivalent to that also seen in human basophils (Figure 1) and was not a result of direct stimulation by the drug itself. Sulfasalazine alone caused a slight enhancement of s p o n t a n e o u s histamine r e l e a s e - - t h e background histamine release in absence o f immunological stimulation such as anti-IgE from the intestinal mast cells, which was 5% at the highest concentration of drug (10 -3 M), but this did not account for the potentiation seen with anti-IgE. There was no significant difference between IBD specimens and non-IBD specimens in the enhancement of either anti-IgE-induced or spontaneous histamine release Digestive Diseases and Sciences, Vol. 36, No. 2 (February 1991)
1oo m
$o
@
80 7O E m
60 5O 4O
30 20 ~c
10
o
10 .8
10 .7
10 .6
10 -5
10'-4
10-3
10-2
Sulfasalazine (M) Fig 1. Sulfasalazine enhancement of anti-IgE-stimulated histamine release from human intestinal mast cells (111) and human blood basophils (El). Each point represents the mean - SEM of duplicate determinations from seven preparations of intestinal mucosal cells and six preparations of human blood basophils. Mean histamine release from sulfasalazine-treated intestinal mast cells was statistically different from mean control release of 24.8% (• 3.2% SEM) at 10-3 -5 x 10 -4 M (P < 0.03). Mean histamine release from sulfasalazine-treated basophits was statistically different from control basophil release (48.5% - 5 SEM) at 10-3-5 x 10-5 M with p < 0.001 andat 10-5 M w i t h P < 0.05.
by sulfasalazine. Sulfasalazine alone had no effect on the spontaneous release of the blood basophils. It is significant that sulfasalazine increased the histamine release from mast cells that were already maximally stimulated by the immunologic challenge of anti-IgE. (The peak mast cell histamine release occurs at 1 g.g/ml anti-IgE). While it would ordinarily be best to stimulate the mast cells with a suboptimal anti-IgE stimulus in order to allow for response to a second stimulus, it was necessary to challenge the mast cells for the sulfasalazine experiments with an anti-IgE concentration of 1 Ixg/ml in order to have sufficient specific release above the background spontaneous release (baseline background plus sulfasalazine enhancement was 9% at 10 -3 M). Paradoxically, the active metabolite of sulfasalazinc, 5-aminosalicylic acid (5-ASA), was found to be an inhibitor of anti-IgE-stimulated histamine release at concentrations from 3 x 10 -s to 10 .3 M (Figure 2). Effective inhibition of histamine release was seen even though the mast cells were maximally stimulated with 1 o,g/ml anti-IgE, and at that level of stimulation the IDso of 5-ASA was found to be 5 x 10 -4 M. There was no effect of 5-ASA on the spontaneous histamine release of either mast cells or basophils. 5-ASA also inhibited anti-IgE stimulated histamine release from human blood basophils in a similar d o s e - r e s p o n s e (10-7-10 -3 M) (Figure
] 81
FOX ET AL 8O t~
7O
n-
60
r-
"~
50
.~ "1" "6 e,-
40
Sulfapyridine had no significant effect on anti-IgE induced histamine release from human intestinal mast cells (N = 4, data not shown). DISCUSSION
30
.0
2o
~e,.
10
_=
0
.
10"8
.
.
10-7
.
.
10 .6
.
.
.
10 .5
.
.
.
.
10 .4
10"3
.
j
10-2
5 - A m i n o s a l i c y l i c A c i d (M)
Fig 2. 5-Aminosalicylic acid (5-ASA) inhibition o f anti-lgEstimulated histamine release from human intestinal mast cells (11) and human blood basophils (2). Each point is the mean = SEM of duplicate determinations from eight different intestinal mast cell and six different human blood basophil preparations. Mean histamine release from 5-ASA-treated intestinal mast cells was statistically different from control histamine release (t8.9% *-- 2.6 SEM) at 10.3 -10 .4 M (P < 0.03). The mean histamine release from 5-ASA-treated basophils was statistically different from controls (mean 42.5 +-- 9.3 SEN) at 10 ~3 -3 x 10 -~ M (P
Modulation of Mediator Release from Human Intestinal Mast Cells by Sulfasalazine and 5-Aminosalicylic Acid CHARITY C. FOX, MD, WENDY C. MOORE, BS, and LAWRENCE M. LICHTENSTEIN, MD
Intestinal mast cells are thought to contribute to the mucosal inflammation in ulcerative colitis and Crohn' s disease through release of inflammatory mediators. Since sulfasaIazine and its metabolite 5-aminosalicylic acid are effective therapeutic agents in inflammatory bowel disease and have been shown to inhibit generation o f inflammatory products in other cells, we examined the effect of these agents in vitro on human intestinal mast cell mediator release. Sulfasalazine (5 x 10 -4 -10 -3 M) was found to significant~ enhance goat anti-human lgE-inducedhistamine release from intestinal mast cells, which is the same response as seen in human blood basophils, whereas its metabolite 5-aminosalicylic acid was an effect&e inhibitor of stimulated histamine release in both mast cells and basophils. 5-Aminosalicylic acid also inhibited production of prostaglandin D 2 by the stimulated intestinal mast cells. Sulfasalazine alone, without immunologic stimulation, did not induce histamine release from mast cells or basophils, but the enhancement of ongoing mast cell activation by sulfasalazine may explain some cases o f adverse reactions to the drug. The inhibition of mast ceil histamine release and prostaglandin generation by 5-aminosalicylic acid demonstrates a potential therapeutic modality o f this agent. KEY WORDS: intestinal mast cells; basophils; histamine; prostaglandin D2; sulfasalazine; 5-aminosalicylic acid.
Mast cell hyperplasia is a consistent feature of the idiopathic inflammatory bowel diseases (IBD) ulcerative colitis and Crohn's disease (1-3). While mast cells often are found to be increased in inflammatory foci, the number of mast cells found in the mucosa and submucosa in IBD is impressive and has led investigators to speculate on the role these Manuscript received January 24, 1990; revised manuscript received August 31, 1990; accepted September 6, 1990. From the Johns Hopkins Asthma and Allergy Center, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21224. This work was supported by the National Foundation for Ileitis and Colitis and the National Institute of Allergy and Infectious Disease grants AI07290 and AI08270. Address for reprint requests: Dr. Charity C. Fox, Department of Medical Microbiology and Immunology, The Ohio State University, 5072 Graves Hall, 333 W. 10th Avenue, Columbus, Ohio 43210-1239.
inflammatory cells play in the pathology of IBD (1-4). It is certain that the mast cell, with its array of inflammatory mediatorsmincluding histamine, prostaglandins, sulfidopeptide leukotrienes, proteases, chemotactic factors, and cytokines--is capable of contributing significantly to the tissue response in IBD (5-7). W e hypothesized that, if mast cells were inciting inflammation in IBD, drugs that were effective in the treatment of IBD should inhibit the release of mast cell mediators of inflammation such as histamine and the products of arachidonic acid metabolism. Sulfasalazine and its active metabolite 5-aminosalicylic acid (5-ASA) both have been found to be useful in treatment of mild to moderate attacks of ulcerative colitis and as maintenance therapy to
Digestive Diseases and Sciences, Vol. 36, No. 2 (Februar." 1991)
0163-2116/9110200-0179506.50/0 9 1991PlenumPublishingCorporation
179
FOX ET AL p r e v e n t relapses (8). 5-ASA also has been reported to be effective in treatment of active Crohn's disease (9, 10). While the exact mechanisms by which these agents act in vivo are not known, it has been suggested that their therapeutic activities are due to effects on the generation of inflammatory mediators (11-14). There are reports on the actions of sulfasalazine and 5-ASA on arachidonic acid metabolism in several human cell types. It is known that sulfasalazine inhibits thromboxane synthetase in human platelets (11) and Nso blocks products of the 5-1ipoxygenase pathway in human neutrophils (12). In o t h e r experiments using whole colonic mucosa biopsies, sulfasalazine and 5-ASA have been shown to inhibit calcium ionophore A23187-induced generation of both lipoxygenase and cyclooxygenase products (13, 14). Testing the action of these drugs on human mast cells in vitro required the ability to isolate the mast cell from the tissue and to activate it. Using our m e t h o d for isolation of cells from human colonic m u c o s a (15), we produced single-cell suspensions of intestinal lamina propria cells, selectively stimulated the mast cells in these suspensions with goat anti-human IgE, and measured the mast cell specific p r o d u c t histamine as well as the inflammatory arachidonic acid metabolite prostaglandin D,. In this initial study, we chose to examine the action of 5-ASA on prostaglandin D2 (PGD) release since PGD is the predominant cyclooxygenase product of mast cells, and the intestinal mast cells make a larger quantity of this metabolite as compared to their predominant lipoxygenase product leukotriene C. T h e mast cells were preincubated with sulfasalazine and 5-ASA and subsequently challenged to test the effects of these drugs on the stimulated release of inflammatory mediators. Since sulfasalazine had been previously reported to enhance histamine release in human basophilmixed leukocyte preparations (16), we also studied the action of sulfasalazine and 5-ASA on human basophil histamine release. The basophil experiments functioned as controls for the mast cell experiments and, while the drugs would not necessarily have the same effects on both cell types, a similar basophil response would corroborate the mast cell response. MATERIALS AND METHODS T'.lssue Source. Human colonic mucosa specimens were obtained from routine surgical resections. Five resections
180
were performed for removal of adenocarcinoma, two for diverticulosis, and three for ulcerative colitis. The patients ranged in age from 19 to 73. There were four females and six males. None of the patients were currently taking the drugs studied. The patients with diverticulosis and adenocarcinoma received only the standard preoperative bowel preparations and sedatives. The inflammatory bowel disease patients were not on the drugs studied, but were maintained on low to moderate doses of prednisone, and received the standard bowel preparation and preoperative sedatives. In addition, one of the ulcerative colitis patients was maintained on hydrochlorothiazide and Tenormin, and another was on 1 g/day sulfasalazine that was changed on admission. Blood specimens for basophil histamine release were obtained from normal volunteers on no medications. Tissue and blood were obtained in protocols approved by the Joint Committee on Clinical Investigation. Isolation of Intestinal Mast Cells. Specimens of tissue 10 cm z were taken from grossly normal mucosa at least 4 cm from the diseased site. The mucosa was dissected carefully from the underlying muscularis and mechanically and enzymatically dispersed into a single-cell suspension as previously described (15). In brief, the strips of mucosa were chopped into 1 x l-cm pieces and incubated at 37~ C for 60 rain in calcium- and magnesium-free Hanks' balanced salt solution with 25 mM HEPES (CMF-HBSS/ HEPES) and 0.75 mM EDTA. The pieces were then washed with CMF-HBSS/HEPES, chopped further into 2 x 2-mm pieces and incubated in HBSS with 30 units/ml collagenase for 40 rain at 37~ C. At the end of the collagenase incubation, the supernatants were filtered through 150-p,m pore size Nitex cloth and the filtrate saved. The tissue then was incubated with pronase (1 I~g/g wet weight tissue) and chymopapain (0.5 I~g/g wet weight tissue) for two incubations of 30 rain each. After each incubation, the tissue was rinsed over Nitex with CMF-HBSS/HEPES and the filtrates saved. The filtrates from the collagenase and the pronase/chymopapain digestions were combined, centrifuged, and the cell pellet resuspended in buffer. The mast cells were counted using a low pH Alcian blue stain specific for mast cells and basophils (17), then centrifuged before placing in shortterm culture. Since basophils have been found to be rare in human intestinal mucosal preparations (15), all Alcian blue positive cells were counted as mast cells. Short-Term Culture. The cells were suspended in RPMI 1640 with 25 mM HEPES buffer (Gibco, Grand Island, New York), 2 mM L-glutamine, 100 izg/ml penicillin, and I00 p.g/ml gentamicin at a concentration of 2 x 106 cells/ml. The cell suspensions were incubated in a 75-cm2 tissue culture flask (Coming 25115) at 37~ in humidified 95% air and 5% COz. The cells were harvested after 12-18 hr, washed several times in PAG buffer (PIPES, 7.6 g/liter; NaCI, 6.4 g/liter; KCI, 0.37 g/liter; dextrose, 1.0 g/liter; human serum albumin, 3.0 mg/liter), pelleted, and recounted. Basophil Preparation. Basophil containing mixed leukocyte preparations were obtained from human whole blood as previously described (18). Blood was drawn and mixed with dextran-EDTA and allowed to sediment 90 min. The leukocyte-rich upper layer was drawn off and Digestive Diseases and Sciences. Vol. 36, No. 2 (February. 1991)
SULFASALAZINE, 5-ASA, AND INTESTINAL MAST CELLS centrifuged. The resultant pellet was washed twice in PAG and pelleted before adding final buffer for histamine release assay. Histamine Release Assay. Duplicate samples of 2 x 104 mast cells or basophils were challenged in PAGCM buffer (PAG buffer with CaCI_,-2H,O, 0.14 g/liter and MgC1, - 6H,O, 0.20 g/liter). Cells suspended in 0.3 ml of PAGCM were added to 0.1-ml volumes of buffer containing drug and preincubated for 10 rain. Sulfasalazine, sulfapyridine, and 5-aminosalicylic acid (Sigma, St. Louis, Missouri) were tested in concentrations ranging from 10-~ M to 10-3 M. At the end of the preincubation period, the mast cells or basophils in the cell suspension were selectively challenged with the stimulus--goat antihuman IgEnand incubated further for 20 rain (mast cells) or 30 rain (basophils). Supernatants were assayed for histamine content by automated fluorometric analysis (19). Results are expressed as percent of total histamine content as determined by lysis of cells in 2% perchloric acid and indicate specific stimulated histamine release since the background spontaneous release is subtracted from the total released histamine. Radioimmunoassay. Prostaglandin D,_ (PGD) was measured by radioimmunoassay (RIA) as previously described (20) in which the bound ligand is separated from free with dextran-coated charcoal. Supernatants from challenged cells were decanted and aliquots frozen at -20 ~ C. The PGD assay was performed on unextracted aliquots neat and at 1:5 dilution. The PGD antiserum had a 4% cross-reactivity with PGF,a and less than 1% with all other heterologous ligands (20). The sensitivity limit of the assay was 20 pg PGD/0.1 ml sample. Results are expressed as nanograms of immunoreactive PGD per 106 mast cells. Statistical Tests. Data for the histamine and prostaglandin release experiments were compared by paired twosample Student's t test and the Wilcoxon matched-pairs signed-rank test. Values of P < 0.05 were considered significant. RESULTS Sulfasalazine (10-7-10 -3 M) caused enhancement of goat anti-human IgE-induced release from human intestinal mast cells (Figure 1) that was statistically significant at 10-4-10 -3 M. This enhancement of stimulated histamine release was equivalent to that also seen in human basophils (Figure 1) and was not a result of direct stimulation by the drug itself. Sulfasalazine alone caused a slight enhancement of s p o n t a n e o u s histamine r e l e a s e - - t h e background histamine release in absence o f immunological stimulation such as anti-IgE from the intestinal mast cells, which was 5% at the highest concentration of drug (10 -3 M), but this did not account for the potentiation seen with anti-IgE. There was no significant difference between IBD specimens and non-IBD specimens in the enhancement of either anti-IgE-induced or spontaneous histamine release Digestive Diseases and Sciences, Vol. 36, No. 2 (February 1991)
1oo m
$o
@
80 7O E m
60 5O 4O
30 20 ~c
10
o
10 .8
10 .7
10 .6
10 -5
10'-4
10-3
10-2
Sulfasalazine (M) Fig 1. Sulfasalazine enhancement of anti-IgE-stimulated histamine release from human intestinal mast cells (111) and human blood basophils (El). Each point represents the mean - SEM of duplicate determinations from seven preparations of intestinal mucosal cells and six preparations of human blood basophils. Mean histamine release from sulfasalazine-treated intestinal mast cells was statistically different from mean control release of 24.8% (• 3.2% SEM) at 10-3 -5 x 10 -4 M (P < 0.03). Mean histamine release from sulfasalazine-treated basophits was statistically different from control basophil release (48.5% - 5 SEM) at 10-3-5 x 10-5 M with p < 0.001 andat 10-5 M w i t h P < 0.05.
by sulfasalazine. Sulfasalazine alone had no effect on the spontaneous release of the blood basophils. It is significant that sulfasalazine increased the histamine release from mast cells that were already maximally stimulated by the immunologic challenge of anti-IgE. (The peak mast cell histamine release occurs at 1 g.g/ml anti-IgE). While it would ordinarily be best to stimulate the mast cells with a suboptimal anti-IgE stimulus in order to allow for response to a second stimulus, it was necessary to challenge the mast cells for the sulfasalazine experiments with an anti-IgE concentration of 1 Ixg/ml in order to have sufficient specific release above the background spontaneous release (baseline background plus sulfasalazine enhancement was 9% at 10 -3 M). Paradoxically, the active metabolite of sulfasalazinc, 5-aminosalicylic acid (5-ASA), was found to be an inhibitor of anti-IgE-stimulated histamine release at concentrations from 3 x 10 -s to 10 .3 M (Figure 2). Effective inhibition of histamine release was seen even though the mast cells were maximally stimulated with 1 o,g/ml anti-IgE, and at that level of stimulation the IDso of 5-ASA was found to be 5 x 10 -4 M. There was no effect of 5-ASA on the spontaneous histamine release of either mast cells or basophils. 5-ASA also inhibited anti-IgE stimulated histamine release from human blood basophils in a similar d o s e - r e s p o n s e (10-7-10 -3 M) (Figure
] 81
FOX ET AL 8O t~
7O
n-
60
r-
"~
50
.~ "1" "6 e,-
40
Sulfapyridine had no significant effect on anti-IgE induced histamine release from human intestinal mast cells (N = 4, data not shown). DISCUSSION
30
.0
2o
~e,.
10
_=
0
.
10"8
.
.
10-7
.
.
10 .6
.
.
.
10 .5
.
.
.
.
10 .4
10"3
.
j
10-2
5 - A m i n o s a l i c y l i c A c i d (M)
Fig 2. 5-Aminosalicylic acid (5-ASA) inhibition o f anti-lgEstimulated histamine release from human intestinal mast cells (11) and human blood basophils (2). Each point is the mean = SEM of duplicate determinations from eight different intestinal mast cell and six different human blood basophil preparations. Mean histamine release from 5-ASA-treated intestinal mast cells was statistically different from control histamine release (t8.9% *-- 2.6 SEM) at 10.3 -10 .4 M (P < 0.03). The mean histamine release from 5-ASA-treated basophils was statistically different from controls (mean 42.5 +-- 9.3 SEN) at 10 ~3 -3 x 10 -~ M (P
