- Email: [email protected]
Resident intestinal macrophages are highly phagocytic but do not produce inflammatory cytokines
GASTROENTEROLOGY Vol. 118, No.4
A362 AGA ABSTRACTS
1932
1934
RESIDENT INTESTINAL MACROPHAGES ARE mGHLY PHAGOCYTIC BUT DO NOT PRODUCE INFLAMMATORY CY· TOKINES. Lesley E. Smythies, Marty Sellers, Martin Graham, Phillip D. Smith, Univ of Alabama, Birmingham, AL; Med Coli of Virginia, Richmond, VA. Intestinal lamina propria macrophages (LPM) lack surface CDI4, the LPS binding receptor, and CD89, the receptor for IgA. The functional implications of this unique macrophage phenotype have not yet been determined. Purpose: To determine whether their phentype confers upon intestinal macrophages unique functional characteristics, we compared the secretory and phagocytic capabilities of human LPM with those of blood monocytes. Methods: LPM isolated from normal small intestine and blood monocytes were cultured for 24 h in the presence or absence of LPS, IFN-y, IFN-y plus LPS, PMA or H. pylori urease. Because LPM are thought to be derived from blood monocytes, we also studied CD89 expression, cytokine secretion and phagocytosis in sorted CDI4+ and CDI4- monocytes. Results: Surprisingly, stimulated LPM (99% CDI4-/CD89-) released no IL-I, IL-6 or TNF-a and only pg amounts of IL-8, regardless of the length of incubation before and after stimulation. In contrast, stimulated unsorted monocytes (89% CD 14+/CD89+) produced ng levels of all four cytokines. Similar to LPM, CDI4- monocytes lacked CD89 and were incapable of producing lL-I, IL-6 and TNF-a and produced only pg levels of IL-8. Flow cytometric analysis for intracellular cytokines supported these findings. Nevertheless, LPM, CDI4- monocytes and CDI4 monocytes were all strongly phagocytic for FITC-Iabelled beads. Conclusions: These findings suggest that resident intestinal macrophages in normal small intestine are incapable of inducible inflammatory cytokine production but retain their avid phagocytic activity. The similar phenotype, cytokine secretion profile and phagocytic activity of CDI4- monocytes suggests that LPM are derived from the CDI4- subset of blood monocytes.
CYTOKINE INDUCTION OF NO RELEASE AND DEFENSIN EX· PRESSION IN INTESTINAL CELLS IS DEPENDENT ON NF·KB ACTIVATION. Thomas Witthoeft, Eduard F. Stange, Med Univ of Luebeck, Luebeck, Germany. Background: We recently showed that iNOSmRNA is being expressed and NO is being released in Caco-2 either after stimulation with proinflamrnatory cytokines or after infection with enteroinvasive bacteria. Defensins, antimicrobial peptides, are expressed by epithelial cells and might be involved in inflammatory diseases. It has been shown that sulfasalzine is a potent and specific inhibitor of NF-K{3 (JCI,101(5):1163-75). Since both NO and defensins have antibacterial properties we studied whether their modulation was coordinately regulated. Methods: Caco-2 cells were kept in DME medium for up to 7 days til monolayers were almost confluent. Cells were stimulated with cytokines (e.g. IL-I{3, TNF-a, IFN-y) alone or in combination with all three cytokines. Medium and cells were harvested after a 24hr period. In several experiments cell cultures were stimulated with proinflammatory cytokines (IOng/ml) and sulfasalazine (0.01, 0.1 or IrnM) was added concomitantly. Nitrite, the stable end product of NO, was measured in supernatants by GRIESS reaction and mRNA was extracted to perform RT-PCR for iNOS and HBD-2. Results: NO release into supernatants was induced in cell cultures stimulated with various proinflammatory cytokines compared to controls. The addition of sulfasalazine at the time of stimulation caused a significant inhibition (55% at the lowest and up to 100% at the highest concentration) of NO release in dose-dependent fashion. Defensin (HBD-2) was upregulated after stimulation with combined cytokines but was not consistently downregulated by sulfasalazine. Conclusion: Stimulation of Caco-Z cells with proinflammatory cytokines caused a dose- and time dependent increase of nitrite release into supernatants and HBD-2 mRNA expression. Addition of sulfasalazine, a specific inhibitor of NF-K{3, caused a dose-dependent inhibition of NOS, whereas HBD-2 was not consistently downregulated.
1933
1935
A THREE DIMENSIONAL ORGANOTYPIC MODEL FOR THE STUDY OF CELL INTERACTION IN THE INTESTINAL MU· COSA. Tanja Spoettl, Martin Hausmann, Marina Kreutz, Werner Falk, Juergen Schoelmerich, Tilo Andus, Gerhard RogIer, Dept of Internal Medicine I, Regensburg, Germany; Dept of Internal Medicine I, Univ of Regensburg, Regensburg, Germany.
DESENSITIZING THE CD2·MEDIATED ACTIVATION OF LAM· INA PROPRIA T CELLS (LPT) BY IL·2. Virginia L. Wong, Kimberly A. Krivacic, Andrew E. Schade, Alan D. Levine, Case Western Reserve Univ, Cleveland, OH. LPT perform dual functions, remaining tolerant to antigens derived from dietary proteins and resident microflora, yet providing immunological protection against pathogens. The mechanisms that regulate these dual activities are not known. In vitro LPT are hyporesponsive to stimulation via the CD3 complex. Activation of LPT is dominated by signaling through an alternate pathway mediated by CD2. We recently reported a model in which conditioning LPT with IL-2 restores CD3-responsiveness (GE, 1999, 116:A-762). Using this model, we investigated the impact of reconstituted CD3 signaling on LPT activation through the CD2 receptor. Methods: LPT, isolated from normal mucosa of surgical specimens, were conditioned with 20 U/rnl IL-2. Fresh and conditioned LPT were stimulated with antibody against CD3 or against CD2 (aCD2). Proliferation at 72 h was measured by [3Hl-thymidine incorporation, cytokine production at 48 h by ELISA, CD2 surface expression by flow cytometry, and signal transduction by Western blot with an anti-phophotyrosine mAb. Results: As we and others reported, CD2 stimulation of freshly isolated LPT resulted in greater than 2-fold higher levels of proliferation and 5-fold higher levels of IFNysynthesis than that observed with CD3 stimulation. However, as LPT were conditioned with IL-2, the mean effective concentration (ECso) of aCD2 which stimulated 50% maximal response increased 6-fold over baseline, as the cell surface expression of CD2 also dramatically increased. The combined increase in ECso and surface protein indicates that more CD2 stimulation is required to achieve a similar response. Since the peak proliferative response to CD2 is unchanged, LPT therefore have become desensitized to CD2-mediated activation. A similar increase in ECso for CD3-mediated proliferation of LPT or for CD2- or CD3mediated proliferation of PBT was not observed. In addition, the pattern of tyrosine phosphorylation of proteins at 105,76,70,52-54,35, and 22 kDa after CD2-activation is markedly different between fresh and IL-2 conditioned LPT. Conclusions: Conditioning of LPT with IL-2 not only reestablishes their ability to respond through the TCRlCD3 complex, but also decreases their sensitivity to CD2-mediated activation. This results in an overall loss of dominance of the CD2 activation pathway, as LPT switch from a tolerant to an immune state. These findings suggest that regulation of intracellular signaling initiated through the CD2 receptor in LPT may be a mechanism for modulating mucosal tolerance and immunity.
Background: Recently we demonstrated that freshly elutriated monocytes differentiate into macrophages with a phenotype similar to that of intestinal macrophages from non inflamed mucosa in a three dimensional model (multicellular spheroids) of intestinal epithelial cells. Now we established a more organotypic model to study cell interactions in the intestinal mucosa. Methods: Primary intestinal fibroblasts from not inflamed mucosa and freshly elutriated blood monocytes (ratio 1:1) were embedded in a collagen I (from rat tail)/medium mixture and cultured on sixwell filter inserts for 5 days. At day 5 intestinal epithelial cells (HT-29, WiDr) were seeded on top of the collagen gels (4,5x lOS cells/gel) and cultures were refed every two days with medium containing 2% human AB serum. After further seven days collagen gels were harvested and fixed for immunohistochemical analysis. For immunohistochemistry cells were either isolated by collagenase and fixed on glass slides or cryosections of the aggregates were prepared. Staining procedure was carried out according to a standard APAAP-technique. Results: Collagen gels were contracted by the intestinal fibroblasts to a diameter of about l cm and formed stable three dimensional aggregates within the first five days. Intestinal epithelial cells formed a monolayer on top of the gels about three days after seeding as seen by positive staining for EP-4 in cryosections. Fibroblasts were distributed over the whole three dimensional aggregate (shown by positive staining with an anti-fibroblast antibody). Monocytes inside the aggregates could be identified by positive staining for CD68, but no signal for CDI4 or CDllb could be detected after the seven day culture period. Conclusion: In the three dimensional organotypic cell culture model resembling intestinal mucosa monocytes differentiate into intestinal like rnacrophages.
A362 AGA ABSTRACTS
1932
1934
RESIDENT INTESTINAL MACROPHAGES ARE mGHLY PHAGOCYTIC BUT DO NOT PRODUCE INFLAMMATORY CY· TOKINES. Lesley E. Smythies, Marty Sellers, Martin Graham, Phillip D. Smith, Univ of Alabama, Birmingham, AL; Med Coli of Virginia, Richmond, VA. Intestinal lamina propria macrophages (LPM) lack surface CDI4, the LPS binding receptor, and CD89, the receptor for IgA. The functional implications of this unique macrophage phenotype have not yet been determined. Purpose: To determine whether their phentype confers upon intestinal macrophages unique functional characteristics, we compared the secretory and phagocytic capabilities of human LPM with those of blood monocytes. Methods: LPM isolated from normal small intestine and blood monocytes were cultured for 24 h in the presence or absence of LPS, IFN-y, IFN-y plus LPS, PMA or H. pylori urease. Because LPM are thought to be derived from blood monocytes, we also studied CD89 expression, cytokine secretion and phagocytosis in sorted CDI4+ and CDI4- monocytes. Results: Surprisingly, stimulated LPM (99% CDI4-/CD89-) released no IL-I, IL-6 or TNF-a and only pg amounts of IL-8, regardless of the length of incubation before and after stimulation. In contrast, stimulated unsorted monocytes (89% CD 14+/CD89+) produced ng levels of all four cytokines. Similar to LPM, CDI4- monocytes lacked CD89 and were incapable of producing lL-I, IL-6 and TNF-a and produced only pg levels of IL-8. Flow cytometric analysis for intracellular cytokines supported these findings. Nevertheless, LPM, CDI4- monocytes and CDI4 monocytes were all strongly phagocytic for FITC-Iabelled beads. Conclusions: These findings suggest that resident intestinal macrophages in normal small intestine are incapable of inducible inflammatory cytokine production but retain their avid phagocytic activity. The similar phenotype, cytokine secretion profile and phagocytic activity of CDI4- monocytes suggests that LPM are derived from the CDI4- subset of blood monocytes.
CYTOKINE INDUCTION OF NO RELEASE AND DEFENSIN EX· PRESSION IN INTESTINAL CELLS IS DEPENDENT ON NF·KB ACTIVATION. Thomas Witthoeft, Eduard F. Stange, Med Univ of Luebeck, Luebeck, Germany. Background: We recently showed that iNOSmRNA is being expressed and NO is being released in Caco-2 either after stimulation with proinflamrnatory cytokines or after infection with enteroinvasive bacteria. Defensins, antimicrobial peptides, are expressed by epithelial cells and might be involved in inflammatory diseases. It has been shown that sulfasalzine is a potent and specific inhibitor of NF-K{3 (JCI,101(5):1163-75). Since both NO and defensins have antibacterial properties we studied whether their modulation was coordinately regulated. Methods: Caco-2 cells were kept in DME medium for up to 7 days til monolayers were almost confluent. Cells were stimulated with cytokines (e.g. IL-I{3, TNF-a, IFN-y) alone or in combination with all three cytokines. Medium and cells were harvested after a 24hr period. In several experiments cell cultures were stimulated with proinflammatory cytokines (IOng/ml) and sulfasalazine (0.01, 0.1 or IrnM) was added concomitantly. Nitrite, the stable end product of NO, was measured in supernatants by GRIESS reaction and mRNA was extracted to perform RT-PCR for iNOS and HBD-2. Results: NO release into supernatants was induced in cell cultures stimulated with various proinflammatory cytokines compared to controls. The addition of sulfasalazine at the time of stimulation caused a significant inhibition (55% at the lowest and up to 100% at the highest concentration) of NO release in dose-dependent fashion. Defensin (HBD-2) was upregulated after stimulation with combined cytokines but was not consistently downregulated by sulfasalazine. Conclusion: Stimulation of Caco-Z cells with proinflammatory cytokines caused a dose- and time dependent increase of nitrite release into supernatants and HBD-2 mRNA expression. Addition of sulfasalazine, a specific inhibitor of NF-K{3, caused a dose-dependent inhibition of NOS, whereas HBD-2 was not consistently downregulated.
1933
1935
A THREE DIMENSIONAL ORGANOTYPIC MODEL FOR THE STUDY OF CELL INTERACTION IN THE INTESTINAL MU· COSA. Tanja Spoettl, Martin Hausmann, Marina Kreutz, Werner Falk, Juergen Schoelmerich, Tilo Andus, Gerhard RogIer, Dept of Internal Medicine I, Regensburg, Germany; Dept of Internal Medicine I, Univ of Regensburg, Regensburg, Germany.
DESENSITIZING THE CD2·MEDIATED ACTIVATION OF LAM· INA PROPRIA T CELLS (LPT) BY IL·2. Virginia L. Wong, Kimberly A. Krivacic, Andrew E. Schade, Alan D. Levine, Case Western Reserve Univ, Cleveland, OH. LPT perform dual functions, remaining tolerant to antigens derived from dietary proteins and resident microflora, yet providing immunological protection against pathogens. The mechanisms that regulate these dual activities are not known. In vitro LPT are hyporesponsive to stimulation via the CD3 complex. Activation of LPT is dominated by signaling through an alternate pathway mediated by CD2. We recently reported a model in which conditioning LPT with IL-2 restores CD3-responsiveness (GE, 1999, 116:A-762). Using this model, we investigated the impact of reconstituted CD3 signaling on LPT activation through the CD2 receptor. Methods: LPT, isolated from normal mucosa of surgical specimens, were conditioned with 20 U/rnl IL-2. Fresh and conditioned LPT were stimulated with antibody against CD3 or against CD2 (aCD2). Proliferation at 72 h was measured by [3Hl-thymidine incorporation, cytokine production at 48 h by ELISA, CD2 surface expression by flow cytometry, and signal transduction by Western blot with an anti-phophotyrosine mAb. Results: As we and others reported, CD2 stimulation of freshly isolated LPT resulted in greater than 2-fold higher levels of proliferation and 5-fold higher levels of IFNysynthesis than that observed with CD3 stimulation. However, as LPT were conditioned with IL-2, the mean effective concentration (ECso) of aCD2 which stimulated 50% maximal response increased 6-fold over baseline, as the cell surface expression of CD2 also dramatically increased. The combined increase in ECso and surface protein indicates that more CD2 stimulation is required to achieve a similar response. Since the peak proliferative response to CD2 is unchanged, LPT therefore have become desensitized to CD2-mediated activation. A similar increase in ECso for CD3-mediated proliferation of LPT or for CD2- or CD3mediated proliferation of PBT was not observed. In addition, the pattern of tyrosine phosphorylation of proteins at 105,76,70,52-54,35, and 22 kDa after CD2-activation is markedly different between fresh and IL-2 conditioned LPT. Conclusions: Conditioning of LPT with IL-2 not only reestablishes their ability to respond through the TCRlCD3 complex, but also decreases their sensitivity to CD2-mediated activation. This results in an overall loss of dominance of the CD2 activation pathway, as LPT switch from a tolerant to an immune state. These findings suggest that regulation of intracellular signaling initiated through the CD2 receptor in LPT may be a mechanism for modulating mucosal tolerance and immunity.
Background: Recently we demonstrated that freshly elutriated monocytes differentiate into macrophages with a phenotype similar to that of intestinal macrophages from non inflamed mucosa in a three dimensional model (multicellular spheroids) of intestinal epithelial cells. Now we established a more organotypic model to study cell interactions in the intestinal mucosa. Methods: Primary intestinal fibroblasts from not inflamed mucosa and freshly elutriated blood monocytes (ratio 1:1) were embedded in a collagen I (from rat tail)/medium mixture and cultured on sixwell filter inserts for 5 days. At day 5 intestinal epithelial cells (HT-29, WiDr) were seeded on top of the collagen gels (4,5x lOS cells/gel) and cultures were refed every two days with medium containing 2% human AB serum. After further seven days collagen gels were harvested and fixed for immunohistochemical analysis. For immunohistochemistry cells were either isolated by collagenase and fixed on glass slides or cryosections of the aggregates were prepared. Staining procedure was carried out according to a standard APAAP-technique. Results: Collagen gels were contracted by the intestinal fibroblasts to a diameter of about l cm and formed stable three dimensional aggregates within the first five days. Intestinal epithelial cells formed a monolayer on top of the gels about three days after seeding as seen by positive staining for EP-4 in cryosections. Fibroblasts were distributed over the whole three dimensional aggregate (shown by positive staining with an anti-fibroblast antibody). Monocytes inside the aggregates could be identified by positive staining for CD68, but no signal for CDI4 or CDllb could be detected after the seven day culture period. Conclusion: In the three dimensional organotypic cell culture model resembling intestinal mucosa monocytes differentiate into intestinal like rnacrophages.