Search

Regular bathing in the Blue Lagoon has beneficial effects on psoriasis. Previously, we showed that exopolysaccharides (EPS-Ca) secreted by Cyanobacterium aponinum, a dominating organism in the Blue Lagoon, increased IL-10 secretion by human dendritic cells (DCs). In addition, co-culturing allogeneic CD4+ T cells with DCs matured in the presence of EPS-Ca increased differentiation of T cells into T regulatory cells at the cost of the disease inducing Th17 cells. In the present study, EPS-Ca increased the proportion of DCs expressing CD141, a surface molecule linked to regulatory DCs, and the CD141+ cells secreted more IL-10 than the CD141- cells. EPS-Ca decreased T cell secretion of IL-17, IL-13 and IL-10 and the proportion of T cells expressing the activation marker CD69 that has also been linked to lymphocyte retention. In addition, EPS-Ca reduced keratinocyte secretion of CCL20 and CXCL10, chemokines implicated in recruitment of inflammatory cells. EPS-Ca decreased DC expression of Dectin-1/CLEC7A and SYK, keratinocyte expression of CLEC7A, SYK and CAMP (the gene for LL37), and T cell expression of phosphorylated Zap70. These results indicate that EPS-Ca may induce a regulatory phenotype of DCs, T cells that are less active/inflammatory and less prone to being retained in the skin, and keratinocytes that induce less recruitment of inflammatory cells to the skin and that these effects may be mediated by the effects of EPS-Ca on CLEC7A and SYK. Overall the results indicate that EPS-Ca may be involved in the beneficial effects psoriasis patients experience when bathing in the Blue Lagoon.

Polyunsaturated fatty acids (PUFAs) and their metabolites are potent regulators of inflammation. Generally, omega (n)-3 PUFAs are considered proresolving whereas n-6 PUFAs are classified as proinflammatory. In this study, we characterized the inflammatory response in murine peritonitis and unexpectedly found the accumulation of adrenic acid (AdA), a poorly studied n-6 PUFA. Functional studies revealed that AdA potently inhibited the formation of the chemoattractant leukotriene B4 (LTB4), specifically in human neutrophils, and this correlated with a reduction of its precursor arachidonic acid (AA) in free form. AdA exposure in human monocyte-derived macrophages enhanced efferocytosis of apoptotic human neutrophils. In vivo, AdA treatment significantly alleviated arthritis in an LTB4-dependent murine arthritis model. Our findings are, to our knowledge, the first to indicate that the n-6 fatty acid AdA effectively blocks production of LTB4 by neutrophils and could play a role in resolution of inflammation in vivo.

Natural killer (NK) cells and neutrophils engage in crosstalk that is important in inflammation and likely also for resolution of inflammation. NK cells activate neutrophils and induce their infiltration to the inflamed sites but may also influence their apoptosis and their subsequent efferocytosis by macrophages. Several studies indicate that docosahexaenoic acid (DHA) can inhibit NK cell cytotoxicity but the effects of DHA on the ability of NK cells to engage in crosstalk with neutrophils and affect their functions have not been described. This study explored the kinetics of the effects of NK cells and NK cells pre-treated with DHA on neutrophil surface molecule expression and apoptosis, as well as the ability of NK cells to affect other neutrophil functions. In addition, the study explored the effects of neutrophils on NK cell phenotype and function. Primary NK cells were pre-incubated with or without DHA, then stimulated and co-cultured with freshly isolated neutrophils. When co-cultured with NK cells, neutrophils had higher expression levels of CD11b and CD47; secreted more IL-8, IL-1ra, and CXCL10; had increased phagocytic ability; and their apoptosis was increased early after initiation of the co-culture while dampened at a later time-point. Pre-incubation of NK cells with DHA attenuated NK cell-induced upregulation of CD11b and CD47 on neutrophils, had minor effects on NK cell induction of cytokine/chemokine secretion or their phagocytic ability. Neutrophils also affected the function of NK cells, lowering the frequency of NKp46+ and CXCR3+ NK cells and increasing the concentrations of IFN-γ, TNF-α, and GM-CSF in the co-cultures. Pre-incubation of NK cells with DHA further decreased the frequency of NKp46+ NK cells in the co-culture with neutrophils and decreased the concentrations of IFN-γ, CCL3 and GM-CSF. These findings indicate that NK cells have mostly pro-inflammatory effects on neutrophils and that DHA can attenuate some of these pro-inflammatory effects. Neutrophils had both anti- and pro-inflammatory effects on NK cells. When NK cells had been pre-treated with DHA, the anti-inflammatory effects were increased and some of the pro-inflammatory effects attenuated. Overall, the results suggest that DHA may lead to a more anti-inflammatory microenvironment for NK cell and neutrophil crosstalk.

Annotine is a lycopodane-type alkaloid isolated from the Icelandic club moss Lycopodium annotinum ssp. alpestre . Annotine does not inhibit acetylcholinesterase, as some other lycopodium alkaloids do, and other bioactivities have not been reported. The aim of this study was to determine the effects of annotine on maturation of dendritic cells (DCs) and their ability to activate allogeneic CD4 + T cells. Human monocyte-derived DCs were matured in the absence or presence of annotine at a concentration of 1, 10 or 100 μg/ml. The effect of the annotine on maturation of the DCs was determined by measuring concentration of cytokines in culture supernatant by ELISA and expression of surface molecules by flow cytometry. DCs matured in the absence or presence of annotine at 100 µg/ml were also co-cultured with allogeneic CD4 + T cells and concentration of cytokines in supernatants determined by ELISA and expression of surface molecules by flow cytometry. When cultured alone, DCs matured in the presence of annotine secreted less of the pro-inflammatory cytokines IL-6 and IL-23 and had a tendency toward less secretion of IL-12p40 than DCs matured in the absence of annotine. However, when DCs were matured in the presence of annotine and then co-cultured with allogeneic CD4 + T cells they secreted more IL-12p40 and had a tendency toward secreting more IL-6 than DCs matured in the absence of annotine and then co-cultured with T cells. Allogeneic CD4 + T cells co-cultured with DCs matured in the presence of annotine secreted more IL-13 than T cells co-cultured with DCs matured in the absence of annotine, but stimulating the DCs in the presence of annotine did not affect T cell secretion of IFN-γ and IL-17. There was also more IL-10 in co-cultures of T cells and DCs matured in the presence of annotine than in co-cultures of T cells and DCs matured in the absence of annotine. These results show that annotine increases the ability of DCs to direct the differentiation of allogeneic CD4 + T cells toward a Th2/Treg phenotype, which may be of interest in the development of new treatments for Th1- and/or Th17-mediated inflammatory diseases.

Murine zymosan-induced peritonitis is the model most frequently used to study resolution of inflammation. However, the antigen-induced peritonitis model may be better suited for studying resolution of inflammation and the adaptive phase that follows. The objective of this study was to provide an evaluation of the kinetics of cells and mediators during induction, resolution and the adaptive immune phases of a murine antigen-induced inflammation. Female C57BL/6 mice were immunized twice subcutaneously with mBSA and three weeks after the initial immunization they were injected intraperitoneally (i.p.) with mBSA, which induced peritonitis. Peritoneal cells were counted and expression of surface molecules and chemokine receptors analyzed with flow cytometry. Chemokine and cytokine concentrations in peritoneal fluid were determined by ELISA. Two neutrophil populations, differing in size and granularity and slightly in expression of surface molecules, were observed in the peritoneal cavity after induction of inflammation. Macrophages disappeared from the peritoneal cavity following i.p. administration of mBSA but appeared again as they differentiated from recruited monocytes and peaked in numbers at 48 h. At that time point, two distinct populations of macrophages were present in the peritoneal cavity; one with high expression of F4/80, also expressing the atypical chemokine receptor D6 as well as CCR7; the other expressing low levels of F4/80 and also expressing CD11c and CD138. Eosinophils appeared in the peritoneum 3h following i.p. administration of mBSA and peaked at 48 h. At that time point they had upregulated their expression of CCR3 but decreased their expression of CD11b. Peritoneal levels of CCL11 peaked at 6h and may have led to recruitment of the eosinophils. NK cells and T cells peaked at 48 h, whereas B cells peaked at 5 days, with the majority being B1 cells. Peritoneal concentrations of pro-inflammatory cytokines (IL-β and IL-6) and chemokines (CCL2 and CCL3) peaked at 3h, whereas IL-1ra peaked at 6h, sTNF-R at 24h and sIL-6R and TGF-β at 48 h. The results show kinetic alterations in cell populations and mediators in a murine model that may be an excellent model to study initiation and resolution of inflammation and the following adaptive phase.

The effects of omega-3 fatty acids on the adaptive immune response have mainly been analysed in vitro with varying results. How omega-3 fatty acids affect the adaptive immune response in vivo is largely unknown. This study examined the effects of dietary fish oil on the adaptive immune response in antigen-induced inflammation in mice, focusing on its effects on B cells and B cell subsets. Mice were fed a control diet with or without 2.8% fish oil, immunized twice with methylated BSA (mBSA) and peritonitis induced by intraperitoneal injection of mBSA. Serum, spleen and peritoneal exudate were collected prior to and at different time points after induction of peritonitis. Serum levels of mBSA-specific antibodies were determined by ELISA and the number of peritoneal and splenic lymphocytes by flow cytometry. The levels of germinal center B cells and IgM + , IgG + and CD138 + cells in spleen were evaluated by immunoenzyme staining. Mice fed the fish oil diet had more peritoneal B1 cells, more IgM + cells in spleen and higher levels of serum mBSA-specific IgM antibodies compared with that in mice fed the control diet. However, dietary fish oil did not affect the number of peritoneal B2 cells, splenic IgG + or CD138 + cells or serum levels of mBSA-specific IgG antibodies in mice with mBSA-induced peritonitis. These results indicate that dietary fish oil can enhance the adaptive immune response, specifically the B1 cell response, which may lead to better protection against secondary infection as well as improvement in reaching homeostasis following antigenic challenge.

Omega-3 polyunsaturated fatty acids may have beneficial effects in inflammation, where neutrophil migration and activation are of importance. The effects of dietary fish oil on neutrophil numbers and subpopulations in healthy mice and mice with endotoxin-induced inflammation were determined. Mice were fed a control diet with or without 2.8% fish oil, and half of them were injected intraperitoneally with endotoxin. Blood, peritoneal lavage, bone marrow and spleen were collected. Expression of cell surface molecules was analyzed by flow cytometry, and chemokine concentrations were determined by enzyme-linked immunosorbent assay. Dietary fish oil did not alter the proportion of total neutrophils in blood but increased the proportion of a specific subpopulation of neutrophils 48 h following endotoxin administration. This subpopulation of neutrophils expressed higher levels of CD11b, Ly6G and major histocompatibility complex-II, suggesting a different role for these neutrophils in the inflammatory response. Dietary fish oil did not affect neutrophil numbers in the peritoneum of healthy mice, but 12 h after endotoxin administration, there were fewer neutrophils in the peritoneum of mice fed the fish oil diet than in mice fed the control diet. However, 48 h after endotoxin administration, mice fed the fish oil diet had more neutrophils in peritoneum than mice fed the control diet. These results indicate that, although dietary fish oil may delay recruitment of neutrophils from blood to the peritoneum early in inflammation, it has the potential to increase the number of peritoneal neutrophils later, which may be of benefit as impaired neutrophil migration and activation have been associated with immunosuppression late in inflammation.

This study examined whether NK cells are important for resolution of antigen-induced inflammation. C57BL/6 mice were immunized twice with methylated BSA (mBSA) and inflammation induced by intraperitoneal injection of mBSA. Mice were injected intravenously with anti-asialo GM1 (αASGM1) or a control antibody 24h prior to peritonitis induction and peritoneal exudate collected at different time points. Expression of surface molecules and apoptosis on peritoneal cells was determined by flow cytometry and concentration of chemokines, cytokines, soluble cytokine receptors and lipid mediators by ELISA and LC-MS/MS. Apoptosis in parathymic lymph nodes and spleens was determined by TUNEL staining. Mice administered αASGM1 had lower peritoneal NK cell numbers and a higher number of peritoneal neutrophils 12h after induction of inflammation than control mice. The number of neutrophils was still high in the αASGM1 treated mice when their number had returned to baseline levels in the control mice, 48h after induction of inflammation. Peritoneal concentrations of the neutrophil regulators G-CSF and IL-12p40 were higher at 12h in the αASGM1 treated mice than in the control mice, whereas concentrations of lipid mediators implicated in resolution of inflammation, i.e. LXA4 and PGE2, were lower. Reduced apoptosis was detected in peritoneal neutrophils as well as in draining lymph nodes and spleens from the αASGM1 treated mice compared with that in the control mice. In addition, αASGM1 treated mice had lower number of peritoneal NK cells expressing NKp46 and NKG2D, receptors implicated in NK cell-induced neutrophil apoptosis. Furthermore, αASGM1 treatment completely blocked the increase in CD27+ NK cells that occurred in control mice following induction of inflammation, but CD27+ NK cells have been suggested to have a regulatory role. These results indicate a crucial role for NK cells in resolution of antigen-induced inflammation and suggest their importance in tempering neutrophil recruitment and maintaining neutrophil apoptosis.

Ethnopharmacological relevance: Menyanthes trifoliate and Achillea millefolium have been used in traditional medicine to ameliorate chronic inflammatory conditions. The aim of this study was to identify the effects of ethanol and aqueous extracts of Menyanthes trifoliate and Achillea millefolium on maturation of dendritic cells (DCs) and their ability to activate allogeneic CD4+ T cells. Materials and methods Human monocyte-derived DCs were matured in the absence or presence of lyophilised aqoueous or ethanol extracts from Menyanthes trifoliate or Achillea millefolium and their expression of surface molecules analysed with flow cytometry and cytokine secretion measured by ELISA. DCs matured in the presence of aqueous extracts from Menyanthes trifoliate and Achillea millefolium were co-cultured with allogeneic CD4+ T cells and the expression of surface molecules by T cells and their cytokine secretion and cell proliferation determined. Results Maturation of DCs in the presence of aqueous extracts from Menyanthes trifoliate or Achillea millefolium did not affect expression of the surface molecules examined but reduced the ratio of secreted IL-12p40/IL-10, compared with that by DCs matured in the absence of extracts. Allogeneic CD4+ T cells co-cultured with DCs matured in the presence of aqueous extract from Menyanthes trifoliate secreted less IFN-γ, IL-10 and IL-17 than CD4+ T cells co-cultured with DCs matured without an extract. Maturation of DCs in the presence of aqueous extract from Achillea millefolium decreased IL-17 secretion but did not affect IFN-γ and IL-10 secretion by allogeneic CD4+ T cells. Conclusions Aqueous extract from Menyanthes trifoliate induces a suppressive phenotype of DCs that has reduced capacity to induce Th1 and Th17 stimulation of allogeneic CD4+ T cells, whereas aqueous extract from Achillea millefolium reduces the capacity of DCs to induce a Th17 response.

Extracts and fractions from birch bark have been used to treat various diseases, such as skin disorders and rheumatism, and for analgesic effects. Results from studies in vitro and in vivo have shown that birch bark extracts can have immunoregulatory effects. These effects have mainly been attributed to the various triterpenes found in birch bark. The effects of birch bark from Betula pubescens on immune responses have not been reported. Ethanol extract was prepared from dry birch bark (DBBEE) and five fractions made using various ratios of dichloromethane and methanol (fractions I-V). Human monocyte-derived dendritic cells (DCs) were matured with or without DBBEE or fractions I-V at several concentrations. The effects of the extract and fractions on DC maturation were determined by measuring cytokine secretion by ELISA and expression of surface molecules by flow cytometry. DBBEE and fractions III and IV reduced DC secretion of IL-6, IL-10 and IL-12p40 and expression of CD83, CD86, CCR7 and DC-SIGN compared with control DCs. Proliferation of allogeneic CD4(+) T cells co-cultured with DCs matured with fraction IV, as measured by (3)H-thymidine incorporation, was similar to proliferation of allogeneic CD4(+) T cells co-cultured with control DCs. However, IFN-γ secretion was reduced and IL-10 and IL-17 secretion was increased, a cytokine profile consistent with a Th17 regulatory phenotype. These results indicate that bark from Betula pubescens contains compound(s) that can modulate DCs so that their interaction with T cells leads to an immunoregulatory response.

Omega-3 polyunsaturated fatty acids (PUFA) affect immune response, partly by affecting cytokine secretion. Omega-3 PUFA decrease tumor necrosis factor (TNF)-alpha secretion by RAW 264.7 macrophages but increase TNF-alpha secretion by primary elicited peritoneal macrophages in vitro. In this study, the effects of omega-3 and omega-6 PUFA on lipopolysaccharide induced TNF-alpha and interleukin (IL)-10 secretion by murine primary resident and elicited peritoneal macrophages and by RAW 264.7 macrophages, were examined in vitro using an enzyme-linked immunosorbent assay. In addition, the effects of dietary omega-3 PUFA on the number of cells secreting these cytokines were examined with enzyme-linked immunospot assay. All cell types secreted more TNF-alpha but similar amounts of IL-10 when incubated with the omega-3 PUFA, eicosapentaenoic acid or docosahexaenoic acid, compared with that when incubated with the omega-6 PUFA, linoleic acid or arachidonic acid. Dietary fish oil did not affect the number of TNF-alpha secreting resident peritoneal macrophages but decreased the number of macrophages secreting IL-10 ex vivo. These results show that dietary omega-3 PUFA and omega-3 PUFA added to cells in vitro increase TNF-alpha secretion by resident peritoneal macrophages, probably by a direct effect on the cells. In contrast, omega-3 PUFA did not affect IL-10 secretion by the cells but decreased the number of cells secreting IL-10 ex vivo, possibly by affecting cell recruitment, maturation or proliferation.

Dietary n-3 polyunsaturated fatty acids (PUFA) influence the inductive phase of inflammation but less is known about their effects on the resolution phase. This study examined the effects of dietary fish oil on induction and resolution of antigen-induced inflammation in mice. Mice were fed a control diet with or without 2.8% fish oil, immunized twice with methylated BSA (mBSA) and inflammation induced by intraperitoneal injection of mBSA. Prior to and at different time points after mBSA administration, peritoneal cells were analyzed and expression of surface molecules determined by flow cytometry. Concentration of chemokines, cytokines and soluble cytokine receptors was determined by ELISA. Mice fed the fish oil diet had fewer peritoneal neutrophils, shorter resolution interval and lower levels of pro-inflammatory cytokines and chemokines than mice fed the control diet. In mice fed the fish oil diet there was an early peak in peritoneal levels of the immunosuppressive molecules sIL-6R and TGF-β, that was not seen in mice fed the control diet. In the resolution phase, peritoneal macrophages from mice fed the fish oil diet expressed more of the atypical chemokine receptor D6 and peritoneal TGF-β levels were higher than that in mice fed the control diet. Furthermore, in the late-resolution phase there were more peritoneal eosinophils and macrophages in mice fed the fish oil diet than in mice fed the control diet. These results demonstrate a suppressive effect of n-3 PUFA on the inductive phase of inflammation and indicate an enhancing effect of n-3 PUFA on resolution of inflammation.

Fish oil can have beneficial effects in health and disease. In healthy individuals, reduction of the inflammatory status may be of benefit, whereas in patients with systemic inflammation, such as sepsis, it is important to diminish the immunosuppression that is thought to contribute to poor outcome. The objective of this study was to determine the effects of dietary fish oil on monocytes/macrophages in blood, bone marrow, spleen, and peritoneum and chemokine concentrations in blood and peritoneum in healthy mice and mice with endotoxin-induced inflammation. Mice were fed a Western-type diet without fish oil (C) or with 2.8% fish oil (FO) for 6 wk and then either killed (healthy mice) or injected i.p. with endotoxin (LPS) and killed after 3, 8, 12, 24, or 48 h. Blood, bone marrow, spleen, and peritoneal lavage were collected. Expression of cell surface molecules and chemokine receptors was analyzed by flow cytometry and chemokine concentrations measured by ELISA. Healthy mice in the FO group had lower proportions of classical monocytes in blood than healthy mice in the C group. LPS administration increased the proportion of classical monocytes in blood in mice in the FO group but not in those in the C group. Healthy mice in the FO group had lower serum concentrations of CCL2 than mice in the C group, but in inflamed mice, CCL2 concentrations were higher in the FO group than in the C group. These results indicate that dietary fish oil can attenuate the inflammatory status in homeostasis but intensify the immune response upon inflammation.

Recent studies indicate that neutrophils are heterogeneous and may have an immunosuppressive role in addition to their well-known phagocytic and bactericidal function. This study examined neutrophil subpopulations in the circulation, peritoneum, spleen and bone marrow from mice at various time points after induction of acute inflammation. MATERIAL, TREATMENT AND METHODS: Female C57BL/6 mice were injected intraperitoneally with lipopolysaccharide (LPS). Blood, peritoneal, spleen and bone marrow cells were collected and counted and expression of surface molecules and chemokine receptors analyzed with flow cytometry. Chemokine and cytokine concentrations in serum and peritoneal fluid were determined by ELISA.Neutrophil numbers in the circulation decreased following administration of LPS but reached similar numbers to those prior to inflammation at 8 h. At that time point, two distinct neutrophil populations were present in the circulation. These two neutrophil populations differed in size, granularity and expression of CD11b and Ly6G. Few neutrophils were recruited into the peritoneum until 24 h after administration of LPS at a time when the neutrophils in the circulation had increased their expression of the chemokine receptor CXCR2.Induction of acute inflammation leads to the appearance of two circulating neutrophil subpopulations, which may differ in their activation state and function.

Mucosal tolerance has been shown to reduce disease severity in animal models mimicking human autoimmune diseases. The objective of this study was to examine whether mucosal tolerance against keyhole limpet haemocyanin (KLH) could be used to reduce bovine serum albumin (BSA)-induced arthritis in rats and whether anti-inflammatory drugs or passive cigarette smoke affected tolerance induction. Arthritis was induced by immunizing rats with BSA and then injecting BSA into one knee and saline into the other knee for comparison. Prior to BSA immunization, the rats were treated intranasally with KLH or saline and KLH then injected in the knee joints at the time of BSA injection, or the rats were treated with or without anti-inflammatory drugs or subjected to cigarette smoke prior to and during intranasal treatment with BSA. The rats that received intranasal treatment with KLH had a significantly less inflammation in their left knee joint compared to rats that received intranasal saline treatment. Beclamethasone increased the tolerance effect of BSA, whereas passive cigarette smoke abrogated the mucosal tolerance. This data suggests that bystander suppression can be used to treat arthritis and other autoimmune diseases, even when the autoantigen is not known.

Altered hepatic expression of apolipoproteins occurs during the acute phase response. Here we examined whether the acute phase response alters extra hepatic expression of apolipoproteins. Syrian hamsters were injected with endotoxin (LPS), tumor necrosis factor (TNF), interleukin (IL)-1, or the combination of TNF + IL-1 and mRNAs for serum amyloid A (apoSAA), apolipoprotein (apo) J, apo E. apo A-I, and apo D, were analyzed. LPS increased mRNA levels for apoSAA in all tissues examined. LPS and TNF + IL-1 increased mRNA levels for apo J in kidney, heart, stomach, intestine, and muscle. Individually, TNF and IL-1 were less potent than the combination of the two cytokines. LPS decreased mRNA levels for apo E in all tissues, except for mid and distal intestine. TNF and IL-1 were less effective than LPS. LPS, TNF + IL-1 and TNF decreased mRNA levels for apo A-I in duodenum. mRNA for apo D decreased in heart, were unchanged in brain and increased in muscle, following LPS. The widespread extra hepatic regulation of the apolipoproteins during the acute phase response may be important for the alterations in lipid metabolism that occur during infection and inflammation as well as the immune response.

Tumor necrosis factor (TNF) is a macrophage derived peptide that has an antitumor action and modulates immune and inflammatory reactions. Dietary fatty acids may modulate TNF production as dietary n-3 polyunsaturated fatty acids suppress human monocyte TNF production, but enhance its secretion by murine peritoneal macrophages. Mice were maintained for 5 weeks on diets containing different amounts of n-3 and n-6 fatty acids. TNF, PGE2 and 6-keto PGF1 alpha production was monitored following in vitro stimulation of resident peritoneal macrophages with lipopolysaccharide. Macrophages from mice fed the high n-3 diet produced 8-fold more TNF and half the PGE2 produced by macrophages from mice on the other diets. Indomethacin caused an increase in the TNF production by macrophages from mice on all diets but macrophages from mice on the high n-3 diet produced more TNF than macrophages from mice on the other diets. Exogenous PGE2 (100 nM) greatly decreased TNF production by macrophages from mice on all diets, but macrophages from mice on the high n-3 diet secreted 70% more TNF than macrophages from mice fed the other diets, indicating that PGE2 is only partly responsible for the effects observed. The results show that feeding n-3 polyunsaturated fatty acids may cause enhanced TNF production by resident peritoneal macrophages and that PGE2 is partly responsible for the effect.