Eosinophil infiltration and degranulation in normal human tissues: evidence for eosinophil degranulation in normal gastrointestinal tract.

Eosinophils play an important role in the pathogenesis of inflammatory diseases such as bronchial asthma and host immunity to parasitic infections. Deposition of eosinophil granule proteins and concomitant tissue damage have been documented in various diseases. Here, we review and summarize results of our immunofluorescence studies of eosinophil infiltration and degranulation in various normal human tissues. Furthermore, because eosinophil infiltration and degranulation are not normally present in healthy tissues, we examine whether eosinophil infiltration and degranulation normally occur in the small intestine and whether tissue procurement methods affect the extent of eosinophil infiltration and degranulation there. Hematopoietic and lymphatic tissues, including the thymus, showed eosinophil infiltration, but the only organ showing remarkable eosinophil infiltration and degranulation was the gastrointestinal (GI) tract. Eosinophil degranulation was significantly increased in specimens obtained by endoscopic forceps compared to those obtained by scalpel. These results suggest that tissue procurement methods affect the degree of eosinophil degranulation in the GI tract and that, among normal human body organs, both eosinophil infiltration and degranulation only occur in the GI tract.

Increased eosinophil infiltration and degranulation in colonic tissue from patients with collagenous colitis.

OBJECTIVE: Eosinophils infiltrate the colonic mucosa of patients with collagenous colitis (CC), although the pathogenetic implications are unknown, including whether these eosinophils are activated and degranulate in situ. We examined eosinophil infiltration and degranulation in the intestines of patients with CC by immunofluorescence for eosinophil granule major basic protein (MBP). METHODS: We used both conventional histology (hematoxylin and eosin) and indirect MBP immunofluorescence histochemistry on colon biopsy specimens from patients with CC (n = 21) and from healthy controls (n = 9). Scoring of histological features was performed on hematoxylin and eosin-stained sections on a 0 to 3 scale. Eosinophil infiltration and degranulation, as quantified by extracellular MBP staining, were scored in each specimen on a 0 to 4 scale. RESULTS: The inflammatory infiltrate of the lamina propria, the thickness of the collagen band, the numbers of intraepithelial lymphocytes, and degree of epithelial cell damage were all significantly increased in patients with CC as compared to controls (p < 0.0001). Scores for both eosinophil infiltration and degranulation were also significantly higher in the CC group compared to controls (p < 0.0001). The degree of infiltrating eosinophils by hematoxylin and eosin was correlated with eosinophil infiltration and degranulation by MBP immunostaining; however, no other correlations were found between eosinophil infiltration or degranulation by immunofluorescence and any of the histological parameters measured in the CC group. CONCLUSIONS: Eosinophil infiltration and degranulation are increased in the colonic mucosa of patients with CC compared to healthy controls. Eosinophils and their cytotoxic granule proteins could be involved in the pathogenesis of CC. Further studies will be necessary to elucidate the mechanisms of eosinophil activation in CC.

Reactivity of monoclonal antibodies EG1 and EG2 with eosinophils and their granule proteins.

Use of the murine monoclonal antibodies EG1 and EG2 has been based on the assumption that EG2 recognizes activated eosinophils. We examined the reactivity of EG1 and EG2 with eosinophil cationic protein (ECP), eosinophil-derived neurotoxin (EDN), and stimulated and nonstimulated eosinophils from normal donors. By radioimmunoassay, EG1 recognized only ECP, whereas EG2 recognized both ECP and EDN. By Western blot, EG1 reacted with ECP, EG2 reacted with both ECP and EDN, but EG2 could not distinguish between lysates of stimulated and nonstimulated eosinophils. By immunofluorescence, EG1 and EG2 at 20 microg/mL stained 95-100% of nonstimulated eosinophils, regardless of fixative; EG1 and EG2 at 0.1 microg/mL stained 61-90% of acetone- and paraformaldehyde-fixed and only 5-21% of methanol-fixed nonstimulated eosinophils. Thus, the reactivity of EG1 and EG2 with eosinophils depends on the method of fixation and antibody concentration; and EG2, in contrast to previous reports, cannot reliably discriminate between resting and activated eosinophils.

Ursodeoxycholic acid inhibits eosinophil degranulation in patients with primary biliary cirrhosis.

Eosinophilia is a distinctive feature of primary biliary cirrhosis (PBC), especially in its early stages. Intriguingly, treatment with ursodeoxycholic acid (UDCA) ameliorates eosinophilia as well as liver tests in patients with PBC. It remains unknown, however, whether eosinophils in PBC patients are functionally activated and whether UDCA inhibits eosinophil activation. In the present study, we systematically examined eosinophil dynamics in the blood and liver in patients with stage I to II PBC before and after UDCA treatment. We determined serum concentrations of eosinophil granule proteins (major basic protein [MBP] and eosinophil-derived neurotoxin [EDN]) by radioimmunoassay and quantitated eosinophil degranulation using computer-assisted morphometry after MBP immunohistochemistry. Before UDCA treatment, patients with PBC (n = 25) showed significantly higher circulating eosinophil counts (P <. 05) and serum concentrations of MBP (P <.0005) and EDN (P <.02) compared with patients with chronic viral hepatitis (n = 22), autoimmune hepatitis (n = 10), and obstructive jaundice (n = 12). Four-week UDCA treatment significantly reduced blood eosinophil counts (P <.0001) and serum MBP (P <.0001) and EDN (P <.0001) levels in PBC patients. MBP immunohistochemistry and computer-assisted quantitative morphometry showed infiltration and degranulation of eosinophils in the portal tract in patients with PBC and significant reductions in the number of sites and the area occupied by extracellular MBP deposits after UDCA treatment for 2 years (P <.02) but not in placebo-treated patients. Our results suggest that eosinophils in patients with PBC are not only increased in number, but also release granule proteins, and that UDCA treatment inhibits this eosinophil activation/degranulation.

Eosinophil infiltration and degranulation in normal human tissue.

Localization of eosinophil granule major basic protein by immunofluorescence permits recognition of both eosinophil infiltration and degranulation. Over the past decade and a half, our laboratory has shown that eosinophil infiltration and degranulation occur in many diseased tissues in humans; among normal tissues studied as controls, only the gut showed striking eosinophil infiltration and degranulation. Using an indirect immunofluorescence procedure for the detection of major basic protein, we extended our analyses of normal human tissues to include tissues from essentially all body organs; a total of 117 biopsy/autopsy specimens were analyzed. To determine whether the method of tissue procurement affected the level of eosinophil degranulation in the normal gastrointestinal tract, normal proximal jejunum from six patients was biopsied using either an endoscopic forceps or a scalpel at the time of elective surgery and examined by immunofluorescence. Spleen, lymph node, and thymus tissues showed eosinophil infiltration with scant evidence of degranulation, but the only organ showing both eosinophil infiltration and remarkable degranulation was the gastrointestinal tract. Eosinophil degranulation was significantly increased in specimens obtained by endoscopic forceps compared to those obtained by scalpel (P = 0.021). These results indicate that tissue procurement methods affect the degree of eosinophil degranulation in the gut. Thus, among normal human body organs, both eosinophil infiltration and appreciable degranulation consistently occur only in the gut.

Localization of eosinophil granule major basic protein in paracoccidioidomycosis lesions.

Paracoccidioidomycosis is a chronic granulomatous disease caused by the fungus Paracoccidioides brasiliensis. Although eosinophils have long been associated with the immune defense against helminths, the role of eosinophils in the immune response to fungal diseases is not as well studied. The eosinophil granule major basic protein is toxic to helminths and mammalian cells in vitro, and its release has been used as a marker of eosinophil localization and degranulation. To determine whether eosinophil infiltration and degranulation, as evidenced by the deposition of major basic protein, occur in lesions of P. brasiliensis, we used an immunofluorescence technique to localize the P. brasiliensis organisms and eosinophils and major basic protein. Initially, all tissues were stained with polyclonal antibody to major basic protein; subsequently, colocalization of major basic protein and P. brasiliensis by double staining with mouse and rabbit antibodies, respectively, was performed. Nine biopsy tissues from seven patients were analyzed. All nine biopsies showed infiltration of intact eosinophils using both the monoclonal and the polyclonal anti-major basic protein antibodies, along with the presence of P. brasiliensis. Furthermore, using the polyclonal anti-major basic protein antibody, nine of nine tissues showed extracellular major basic protein deposition (granular or diffuse fluorescence staining outside of intact eosinophils). The double staining procedure using the anti-major basic protein monoclonal antibody showed extracellular deposition in five of eight biopsies; in these five biopsies, approximately 60% of the areas containing P. brasiliensis had extracellular major basic protein deposited on the organisms. These observations support the hypothesis that the eosinophil, through toxic granule proteins such as major basic protein, participates in the pathophysiology of paracoccidioidomycosis.

Localization of eosinophil-derived neurotoxin and eosinophil cationic protein in neutrophilic leukocytes.

Eosinophil-derived neurotoxin (EDN) and eosinophil cationic protein (ECP) are generally regarded as eosinophil-specific proteins. We tested whether EDN and ECP are present in mature neutrophils. By indirect immunofluorescence, both eosinophils and neutrophils stained with antibodies to EDN and ECP. Lysates of purified (<0.1% eosinophil contamination) neutrophils contained EDN, 112+/-4 ng/10(6) cells, and ECP, 163+/-2 ng/10(6) cells, whereas eosinophil major basic protein (MBP) was not detectable. Electron microscopic examination of immunogold-labeled buffy coat cells stained with EDN antibody showed that EDN is localized to neutrophil granules. Finally, EDN mRNA was detected in lysates of highly purified neutrophils (0.001% eosinophil contamination) by the reverse transcription-polymerase chain reaction. We conclude that proteins that are either identical to or immunologically cross-reactive with EDN and ECP are present in neutrophils and that EDN is synthesized and localized to neutrophil granules. Thus, caution must be exercised in interpreting the presence of EDN and ECP as specific markers of eosinophil-associated inflammation in human disease.

Sulfonylureas inhibit cytokine-induced eosinophil survival and activation.

Eosinophils play a key role in the pathogenesis of asthma and other allergic inflammatory diseases. We have previously shown that treatment of eosinophils with lidocaine preferentially inhibits IL-5-induced survival. This inhibition cannot be overcome by increasing concentrations of IL-5 and is not due to the blocking of Na+ channels by lidocaine. Here we report that one class of K+ channel blockers, the sulfonylureas, inhibits eosinophil survival in a manner similar to lidocaine. The sulfonylurea glyburide inhibits eosinophil survival even at high concentrations of IL-5. In contrast, increasing concentrations of IL-3 or granulocyte-macrophage CSF overcome glyburide inhibition. Glyburide also blocks cytokine-induced eosinophil superoxide production. Similar results were seen with the sulfonylureas tolbutamide and glipizide. Interestingly, the effects of glyburide are not antagonized by the ATP-sensitive K+ channel openers cromakalim, pinacidil, or diazoxide. Although Scatchard analysis of [3H]glyburide binding to eosinophil membranes indicated that the high affinity sulfonylurea receptor (SUR1) is not present on eosinophils, human eosinophils do express mRNA homologous to the sulfonylurea receptor family, in keeping with the presence of a sulfonylurea receptor. Finally, coculture of eosinophils with combinations of glyburide, lidocaine, and dexamethasone resulted in synergistic inhibition of cytokine-mediated eosinophil survival and superoxide production. These results have intriguing clinical implications for the treatment of eosinophil-associated diseases.

Localization of eosinophils to airway nerves and effect on neuronal M2 muscarinic receptor function.

Neuronal M2 muscarinic receptors inhibit acetylcholine release from pulmonary parasympathetic nerves but are dysfunctional in antigen-challenged animals and asthmatics. Deletion of pulmonary eosinophils protects M2 receptor function in antigen-challenged guinea pigs. Therefore, the association of eosinophils with airway nerves was investigated. Nerve-associated eosinophils were significantly increased in challenged animals compared with controls (0.75 +/- 0.05 vs. 0.28 +/- 0.05 eosinophils/nerve). In antigen-challenged animals, eosinophil density was greatest around airway nerves, suggesting recruitment to the nerves. M2 receptor function was inversely correlated with the number of eosinophils per nerve, thus eosinophils are associated with airway nerves in antigen-challenged guinea pigs, where they impair M2 receptor function. In airways from three patients with fatal asthma, 196 of 637 eosinophils (30%) were associated with nerves, and release of eosinophil major basic protein was evident; conversely, in three control patients 1 of 11 (9%) eosinophils were in contact with nerves. Thus eosinophils and their granule proteins are also seen in association with airway nerves in patients with asthma.

Localization of pregnancy-associated plasma protein-A and colocalization of pregnancy-associated plasma protein-A messenger ribonucleic acid and eosinophil granule major basic protein messenger ribonucleic acid in placenta.

BACKGROUND: The human eosinophil granule major basic protein (MBP), a 13.8 kilodalton cationic polypeptide constituting the core of the eosinophil granule, is cytotoxic to parasites and numerous mammalian cells. Concentrations of a molecule immunochemically similar to eosinophil granule MBP are present in maternal plasma, and MBP mRNA has been localized to placental X cells by in situ hybridization. Eosinophil granule MBP is initially translated as a nontoxic precursor (proMBP), containing a 9.9 kilodalton acidic pro-portion that is believed to neutralize MBP toxicity. Recent analyses of sera from pregnant women have revealed that pregnancy-associated plasma protein-A (PAPP-A), previously thought to be a homotetramer of PAPP-A subunits, is actually composed of PAPP-A subunits bound by disulfide bonds to equimolar amounts of proMBP molecules to form a complex, PAPP-A/proMBP. In addition, the PAPP-A subunit nucleotide and deduced amino acid sequence have been determined from cloned cDNA. The PAPP-A monomer found in plasma contains 1547 amino acid residues. EXPERIMENTAL DESIGN: Because of the new evidence that PAPP-A is complexed with proMBP, previous studies on the localization of PAPP-A using antibodies to PAPP-A must be questioned. To determine the localization of the PAPP-A subunit, immunofluorescence was performed on normal placental tissues using proMBP absorbed anti-PAPP-A antibody. Furthermore, the expression of PAPP-A mRNA was investigated by in situ hybridization. RESULTS: Immunofluorescence staining with proMBP absorbed anti-PAPP-A antibody showed that PAPP-A is localized to placental septa, anchoring villi, and the syncytia of chorionic villi, whereas MBP is localized only to septa and anchoring villi. By in situ hybridization, PAPP-A mRNA is detected in placental X cells and syncytiotrophoblasts, but MBP mRNA is localized only to placental X cells. CONCLUSIONS: The presence of PAPP-A mRNA and PAPP-A subunit protein in placental X cells and syncytiotrophoblasts indicates that both X cells and syncytiotrophoblasts synthesize the PAPP-A subunit, whereas only X cells synthesize proMBP.

Pregnancy-associated major basic protein: deposition of protein and expression of mRNA at the maternal-fetal junction in early and late gestation.

Pregnancy-associated major basic protein (pMBP) has previously been isolated from human placenta and localized to the X cell. Here we used immunofluorescence staining and in situ hybridization to determine the distribution of pMBP and pMBP mRNA throughout the maternal-fetal junction in both early gestation tissues and at term. In early gestation tissues, pMBP was present only at the placental insertion site. Specifically, pMBP was present in (a) the decidua basalis (in the extracellular space, in interstitial pools and inside endometrial glands) and (b) intracellularly within extravillous interstitial trophoblasts in the decidua, in the myometrium and surrounding but not within luminal cells of spiral arteries. At term, the placental bed showed intense extracellular pMBP staining with little intracellular pMBP. In situ hybridization showed the presence of pMBP mRNA in both the early and late gestational tissues. pMBP mRNA was present in cells in the decidua, at the decidual-myometrial junction and in cell islands. Quantitative image analysis showed statistically significant hybridization signals with the pMBP antisense probe as compared to the control/sense probe. These results indicate that pMBP mRNA is expressed and pMBP is extensively deposited at the maternal-fetal junction in early pregnancy and at term.

The eosinophil as an effector cell of the immune response during hepatic allograft rejection.

OBJECTIVE: To evaluate the role of the eosinophil granulocyte during hepatic allograft rejection. DESIGN: (a) A retrospective case-control study and (b) a prospective study of consecutive liver transplant recipients. PATIENTS: In the retrospective study, eight patients with severe rejection in the first month after liver transplantation were compared with six patients without rejection. In the prospective study, 20 consecutive patients were studied for the presence of liver allograft rejection between March 1989 and October 1989. MEASUREMENTS: Absolute eosinophil counts were determined whenever blood was drawn. Serum was analyzed for the presence of two eosinophil granule proteins, major basic protein and eosinophil-derived neurotoxin, on days 7, 14 and 21 after transplantation. Liver biopsy specimens were stained for the presence of major basic protein by means of immunofluorescence using a double-antibody staining technique. The degree of eosinophil infiltration and degranulation was graded using a panel of representative slides. RESULTS: Blood eosinophilia was increased in patients with hepatic allograft rejection (p < 0.05). Serum major basic protein and eosinophil-derived neurotoxin concentrations were similar in patients with and without rejection. Many portal tracts of patients with rejection contained an abundance of eosinophils, and staining for major basic protein revealed the presence of intact eosinophils. In addition, extracellular major basic protein was seen, sometimes in the absence of intact eosinophils or an extensive infiltrate. In patients with severe allograft rejection, major basic protein staining was present in littoral cells lining the sinusoids. CONCLUSIONS: Patients with severe rejection in the first month after liver transplantation often have blood eosinophilia and marked infiltration of portal tracts with eosinophils or evidence of eosinophil degranulation. The presence of major basic protein likely is direct evidence of tissue destruction and may indicate active rejection (major basic protein in eosinophils and extracellular major basic protein, presence of portal infiltrate) or the immediate postinflammatory rejection state (extracellular major basic protein and major basic protein inside littoral cells, absence of portal infiltrate and eosinophils, bile ducts damaged or vanished). These findings underline the importance of the eosinophil as an integral part of the rejection process. We conclude that the presence of eosinophils or their secretion products in the first month after liver transplantation is an indicator of ongoing or recent allograft rejection.

Eosinophil granule major basic protein inhibition of corneal epithelial wound healing.

PURPOSE: Human eosinophil major basic protein (MBP) was studied in an established organ culture model for rat corneal epithelial wound healing to elucidate further the role of the protein in vernal keratopathy. METHODS: Epithelial migration rates were tested for five MBP concentrations (10, 25, 50, 100, and 200 micrograms/ml MBP). RESULTS: Significantly less epithelial migration than control (P < 0.05) was observed in all tested groups. Histologic examination revealed abnormally heaped-up leading epithelial edges in all test groups compared to the normal tapered edges in all controls. Immunofluorescence disclosed MBP deposition on deepithelialized cornea. CONCLUSIONS: These results suggest that MBP may contribute to vernal corneal ulcerations by inhibiting corneal epithelial migration.

Expression of eosinophil-granule major basic protein messenger ribonucleic acid in placental X cells.

BACKGROUND: The human eosinophil-granule major basic protein (MBP) is a 13.8-kilodalton cationic polypeptide constituting the core of the eosinophil granule. MBP is cytotoxic to parasites and numerous mammalian cells and is a potent secretagogue for platelets, basophils, mast cells, and neutrophils. Concentrations of a molecule immunochemically similar to eosinophil granule MBP are present in maternal plasma, and MBP has been localized by immunofluorescence to placental X cells. EXPERIMENTAL DESIGN: To determine whether X cells produce MBP, the expression of MBP messenger RNA (mRNA) was investigated in placentas by Northern blot analyses and by in situ hybridization with 35S-labeled RNA probes. RESULTS: Northern blot analyses of RNA from placental septa and villi showed the existence of a 1.0-kb RNA band that hybridized with the MBP anti-sense probe; no MBP mRNA was detected in whole blood of normal or pregnant women or in cord blood. Analyses of placentas by in situ hybridization showed MBP mRNA in X cells of placental septa and anchoring villi, but not in other cellular elements such as syncytiotrophoblasts, cytotrophoblasts, villous stromal cells, and fetal endothelial cells. RNase pretreatment abolished X-cell hybridization signals; treatment of sections with an excess of nonradiolabeled anti-sense RNA also blocked binding of the 35S-labeled anti-sense RNA probe. Additional evidence supporting the production of MBP by X cells was obtained using a combination of in situ hybridization and immunofluorescence, which showed colocalization of MBP and its mRNA. CONCLUSIONS: The presence of MBP mRNA and MBP protein in placental X cells indicates that X cells synthesize this biologically active molecule.

Sudden-onset fatal asthma. A distinct entity with few eosinophils and relatively more neutrophils in the airway submucosa?

To determine the histologic differences in the airways of patients who died from sudden-onset asthma and the more common slow-onset asthma, we studied seven cases of fatal asthma. The numbers of eosinophils and neutrophils, as well as extracellular deposition of their respective granule contents in the airway mucosa and submucosa, were determined and statistically analyzed. Four of the seven patients had slow-onset asthma attacks in which the time interval between onset of asthma and death was more than 2.5 h. In contrast, three patients had sudden-onset asthma in which the time interval between onset of asthma attack and death was less than 1 h. The four patients with slow-onset fatal asthma had more eosinophils (34.1 +/- 6.3 in slow-onset; 9.7 +/- 3.5 in sudden-onset; p = 0.002) and fewer neutrophils (4.8 +/- 2.0 in slow-onset; 16.8 +/- 5.4 in sudden-onset; p = 0.008) in the airway submucosa than did patients with sudden-onset fatal asthma. In addition, within the slow-onset fatal asthma group, eosinophils exceeded neutrophils in the airway submucosa (eosinophils > neutrophils, p = 0.002). By contrast, within the sudden-onset fatal asthma group, neutrophils exceeded eosinophils (neutrophils > eosinophils, p = 0.04). We suggest that sudden-onset fatal asthma is immunohistologically distinct from slow-onset fatal asthma and that it is characterized by a relative paucity of eosinophils in the face of an excess of neutrophils in the airway submucosa.(ABSTRACT TRUNCATED AT 250 WORDS)

Eosinophil granule major basic protein deposition in corneal ulcers associated with vernal keratoconjunctivitis.

An indirect immunofluorescence assay detected eosinophil granule major basic protein in the inflammatory debris covering deepithelialized cornea in two patients with vernal keratoconjunctivitis. A slight degree of non-specific fluorescence was present in the control autopsy corneas. High concentrations of the eosinophil granule major basic protein inhibit epithelial migration and protein synthesis, whereas low concentrations affect epithelial migration. The results suggest participation of eosinophil granule major basic protein in the ulcerative process.