Mobilizing lipocortin 1 in adherent human leukocytes downregulates their transmigration.

Polymorphonuclear leukocyte (PMN) migration into sites of inflammation is fundamental to the host defense response. Activation of endothelial cells and PMNs increases the expression or activation of adhesion molecules, culminating in rolling and subsequent adherence of these cells to the vascular wall. Further activation of adherent PMNs, possibly by endothelial cell ligands, leads, within a few minutes, to extravasation itself. This process is not clearly understood, but adhesion molecules or related proteins, as well as endogenous chemokines, may play an important role. The anti-inflammatory glucocorticoids delay extravasation, which implies that an inhibitory regulatory system exists. Resting PMNs contain abundant cytoplasmic lipocortin 1 (LC1, also called annexin I)', and the activity profile of this protein suggests that it could reduce PMN responsiveness. To investigate this we have assessed neutrophil transmigration both in vivo and in vitro and examined the content and subcellular distribution of LC1 in PMNs by fluorescence-activated cell-sorting (FACS) analysis, western blotting and confocal microscopy. We report that LC1 is mobilized and externalized following PMN adhesion to endothelial monolayers in vitro or to venular endothelium in vivo and that the end point of this process is a negative regulation of PMN transendothelial passage.

Glucocorticoids, lipocortins and the immune response.

Glucocorticoids are immunosuppressive and antiinflammatory agents that act through multiple mechanisms. This review highlights recent evidence that lipocortin-1, a member of the annexin family of calcium-binding proteins, can be induced by glucocorticoids, and may mediate some of the effects of glucocorticoids through putative lipocortin-1 receptors, which have been found on the surface of phagocytic cells. Recent advances in annexin biology have been drawn together to formulate a novel hypothesis for the regulation of inflammation.

Rhythms in the immune system.

This report reviews the evidence that cells within the immune system are subject to rhythmic influences that affect numbers of circulating cells and their function both in vitro and in vivo. It is concluded that, although periodicity has clearly been demonstrated for numbers of immunocompetent cells in the circulation, significant functional changes have not been consistently observed. A number of neuroendocrine hormones, which modulate immune responsiveness in vitro and which are released in a rhythmic manner, are considered as mediators of the observed effects on the immune system and this is related to changes in expression and activity of immune-mediated diseases such as rheumatoid arthritis.

Reversible hypercholesterolaemia produced by cholesterol-free fish meal protein diets.

Groups of rabbits were fed isonitrogenous diets containing 30% soya, milk or whitefish meal protein and at least 130 g/kg of added fat for 1 year. Mean serum cholesterol values in fish meal animals (13.4 +/- 2.4 mmol/l) were substantially greater than for soya (3.0 +/- 0.2 mmol/l) or milk (4.6 +/- 0.7 mmol/l). Fish meal rabbits developed extensive aortic atherosclerosis (c. 70% surface involvement) which histologically showed both fibrous and foam cell intimal thickening and destruction and calcification of medial elastic tissue. In a second experiment fish meal was given with either saturated (coconut oil) or unsaturated (maize oil) fat. A similar degree of hypercholesterolaemia developed in each group and was rapidly reversed when a soya protein-low fat diet was substituted. This model may therefore be of value in studies of the progression and regression of experimental atherosclerosis.

Immune tolerance and atherosclerosis in rabbits. Effect of high-fat and cholesterol-supplemented diets.

Breeding rabbits were fed diets supplemented with 5% milk or soya protein for 1 month before conception and throughout pregnancy and lactation. Four groups of 5 or 6 of their offspring were given high-energy isonitrogenous diets containing either 30% soya or milk protein. Half of the animals from each litter received the same protein as their dams; the remainder were fed the alternative (i.e. dams soya; offspring milk, and vice versa). 0.75% cholesterol was added to the diets between 30 and 120 days after weaning. Animals given the same protein as their dams formed substantially lower amounts of food antigen-specific antibody than rabbits fed a novel protein at weaning but the extent of aortic atherosclerosis was similar in all groups. In a second experiment groups of 5-8 weanling rabbits were fed a cholesterol-free diet containing 30% soya and 17% saturated fat for 1 year. Animals in group 1 were bred from dams given soya but those in groups 2 and 3 were derived from a colony fed a soya-free diet. Rabbits in group 3 were immunised by repeated parenteral injections of soya protein and developed high levels of antisoya antibodies. Group 1 and 2 animals were injected with saline only but antisoya antibodies were substantially higher in animals derived from the soya-free breeding colony. Although serum cholesterol and triglyceride levels were similar in all groups, animals from dams fed soya (group 1) had significantly less aortic atherosclerosis (10% involvement of ascending aorta) than those reared from a colony fed a soya-free diet (group 2; 32%). Parenteral immunization with soya protein (group 3) was not associated with significantly increased atherosclerosis (37%). These findings indicate that perinatal exposure to dietary antigen in rabbits may be important in modulating the systemic immune reaction to food antigens. The magnitude of this systemic response is unlikely to alter the nature or distribution of cholesterol-induced atherosclerosis but may have an important influence on the development of aortic disease produced by prolonged feeding of high-fat, cholesterol-free diets.

C1q binding and Raji immune complex assays: a comparison using defined immunoglobulin aggregates.

Aggregated IgG is frequently employed as a standard in systems for the measurement of immune complexes in man and animals. In this paper aggregates prepared by heat or alkali denaturation of human IgG were fractionated by column chromatography through LKB AcA 22 Ultrogel. Heat aggregation yields preparations containing considerably more monomer than alkali treatment (47% and 6.3% respectively). The bulk of aggregated material prepared by both methods was of size 19 S or greater. Smaller aggregates were present in assayable quantities only in the alkali aggregated material. The sized fractions of aggregates IgG were tested in the presence of a human complement source for their efficiency in the C1q binding and Raji radioimmunoassay for immune complexes. Both techniques efficiently measured large aggregates (greater than or equal to 19 S) but the C1q binding assay measured smaller material with greater efficiency than did the Raji cell assay. Neither technique detected monomeric IgG. The date presented is relevant to the binding characteristics of the 2 assay systems studied and suggests that when used together they are capable of measuring immune complexes present over a wide range of sizes.

An assay for the assessment of lipocortin 1 levels in human lung lavage fluid.

The physiological function of the lipocortins, proteins which are thought to be glucocorticoid-regulated, is unclear. An improved assay for lipocortins might help to elucidate their role. A rapid and specific sandwich enzyme-linked immunosorbent assay (ELISA) for lipocortin 1 with a working range of 1-2000 ng/ml and an interrun coefficient of variation of less than 10% is described and used in this pilot study to quantify human lipocortin 1 for the first time in acellular bronchoalveolar lavage fluid (BALF), and in media conditioned by BAL cells, from control patients and those with pulmonary sarcoidosis. Using this assay a statistically significant relationship, not previously observed in man, has been demonstrated between concentrations of lipocortin 1/ml of BALF and serum cortisol levels (n = 10, rs = 0.6939, P less than 0.05). Although lipocortin 1 levels in acellular BALF were the same in control and sarcoid patients, significantly more lipocortin 1 was released from sarcoid BAL cells in culture (median 21.6, range 8.1-45.4 ng lipocortin/10(6) cells/h in culture) than from control cells (2.5, 1.5-7.6 ng lipocortin/10(6) cells/h in culture). The possible clinical significance of these data is discussed, but remains to be established.

Transfection of annexin 1 in monocytic cells produces a high degree of spontaneous and stimulated apoptosis associated with caspase-3 activation.

Transfection of the pre-monomyelocytic U937 cell line with a plasmid coding for full-length annexin 1 (ANX1, 347 amino acid) leads to cell death by promoting apoptosis. In addition, over-expression of the N-terminal and the first domain of the protein (144 amino acids, clone ANX1-S), which does not contain the Ca2+ binding sites, gives susceptibility to cell apoptosis following activation by either 5 ng ml(-1) tumour necrosis factor (TNF)-alpha or 1 - 40 microg ml-1 etoposide. This was demonstrated by using the fluorescent labelled annexin V, cell cycle and nuclear staining analyses. Transfection with an empty plasmid (clone CMV) or with a plasmid carrying the cDNA antisense for ANX1 (clone ANX1-AS) did not alter U937 cells to the degree of apoptosis promoted by either stimulant. Treatment of CMV U937 cells with TNF-alpha increased ANX1 mRNA and protein expression in a time-dependent manner, with maximal increases at 3 and 6 h, respectively. Clone ANX1-S showed higher constitutive (more than 2 fold) and activated caspase-3 activity, associated with higher phospholipase A2 (PLA2) activity (in the region of +50 - 100%), whereas expression of cytosolic PLA2 Bax and Bcl-2 were similar in all cell clones, as determined by Western blotting. In conclusion, this study demonstrates a complex regulatory role of cell apoptosis for ANX1, at least with regards to cells of the myelo-monocytic lineage.

Failed export of the adrenocorticotrophin receptor from the endoplasmic reticulum in non-adrenal cells: evidence in support of a requirement for a specific adrenal accessory factor.

Difficulty in expressing the adrenocorticotrophin (ACTH) receptor (melanocortin 2 receptor; MC2R) after transfection of various MC2R expression vectors has been experienced by many researchers. Reproducible evidence for expression has been obtained only in the Y6/OS3 corticoadrenal cell lines or in cells expressing endogenous melanocortin receptors. In order to determine the cause of this failure of expression we have undertaken the following studies. An MC2R expression plasmid was constructed in which the green fluorescent protein (GFP) coding region had been added to the C-terminus of the mature protein. Transfection of this plasmid into Y6 cells with a cAMP-responsive reporter plasmid demonstrated normal function of this receptor. Imaging of CHO cells expressing MC2R-GFP revealed perinuclear expression, although a cholecystokinin receptor (CCKR)-GFP construct was efficiently expressed at the cell surface. Y6 cells, in contrast, showed cell surface fluorescence after transfection with MC2R-GFP. Several other cell types showed a similar pattern of GFP distribution characteristic of retention in the endoplasmic reticulum. Counterstaining with an anti-KDEL antibody confirmed this location. Co-expression of the MC2R and the CCKR-GFP did not impair CCKR trafficking to the cell surface, implying a receptor-specific impairment to trafficking in the CHO cell which was absent in the Y6 cell.

Lipocortin-1 autoantibody concentration in children with inflammatory bowel disease.

BACKGROUND: Corticosteroids are widely used to treat children with inflammatory bowel disease although the response is variable, side-effects are common, and many patients develop a partial or complete steroid resistance. The mechanism underlying these phenomena are unclear. Corticosteroids mediate some of their actions through lipocortin-1, and the induction of autoantibodies to lipocortin has been proposed as a possible mechanism by which steroid efficacy is suboptimal in vivo. PATIENTS AND METHODS: We have measured serum lipocortin-1 antibody concentration by ELISA in 38 children with Crohn's disease, 12 with ulcerative colitis and in 15 controls. RESULTS: IgG and IgA anti-lipocortin-1 antibody levels were higher in the Crohn's group than in the ulcerative colitis or control groups. Elevated concentrations did not relate to disease activity, history of steroid therapy or steroid-responsiveness. Lipocortin IgM antibody status was similar in all three groups. CONCLUSION: It is therefore unlikely that serum antibodies to lipocortin-1 have a role in the development of steroid-resistance in children with inflammatory bowel disease.

Microbial and immunological investigations of chronic non-ulcerative blepharitis and meibomianitis.

Concentrations of tear lysozyme, lactoferrin, ceruloplasmin, IgG, and IgA have been measured by enzyme linked immunosorbent assay (ELISA) in patients with chronic non-ulcerative blepharitis and meibomianitis at the same time as the lid and conjunctivae were cultured for bacteria and fungi by a semiquantitative method. A group of normal controls aged 20 to 80 were similarly sampled, when strains of Staphylococcus epidermidis from their eyes and the patients' eyes were biotyped according to Baird-Parker's scheme. 5% of blepharitis cases had increased numbers of Staph. aureus present on the lids, compared with only a scanty growth obtained from 5% of normals. 7% of blepharitis cases had increased numbers of Staph. epidermidis type VI (coagulase-negative, mannitol-fermenting) present compared with a scanty growth obtained from 6% of normals. Isolation rates of other types of Staph. epidermidis did not differ from those in normals; no types were associated with meibomianitis. Tear protein profiles were normal in most patients, and there was no increase in tear IgA or IgG, which is expected with chronic infection. Overall our evidence suggests that in 88% of cases these lid conditions have an inflammatory aetiology not associated with infection. Staphylococcal isolates often found in the eye usually represent a normal commensal rather than pathogenic flora.

Bacteriology and tear protein profiles of the dry eye.

The concentrations of tear lysozyme, lactoferrin, ceruloplasmin, IgA, and IgG have been estimated in patients with dry eyes at the same time as semiquantitative bacterial culture was performed of the conjunctivae and lids. Staphylococcal isolations were quantified and biotyped. There was no increased conjunctival colonisation by any particular biotype of Staphylococcus aureus or Staph. epidermidis, and similar numbers of conjunctivae were sterile as in controls (33%); neither were any pathogens such as pneumococci or haemophili isolated. We consider that the conjunctiva of the dry eye, without the lacrimal secretion components of lysozyme and lactoferrin, has an alternative protective antibacterial mechanism which is derived from serum proteins via chronically inflamed vessels.

Increased apoptosis in U937 cells over-expressing lipocortin 1 (annexin I).

The potential involvement of endogenous lipocortin 1 in the process of cellular apoptosis, particularly in cells of the myelo-monocytic lineage, has been investigated. U937 cells were transfected either with an antisense or a sense DNA for lipocortin 1 and the stable clones 36.4AS clone (20-40% lower lipocortin 1 levels) and 15S (30% higher lipocortin 1 levels) were obtained. Cell apoptosis was induced by incubation with tumor necrosis factor-alpha: optimal responses were observed within a 24 h incubation period at a 5 ng/ml concentration. Apoptosis was assessed both morphologically, by annexin V binding and cell cycle analysis with propidium iodide. Whilst no consistent difference was seen between wild type cells and clone 36.4AS, a higher incidence of apoptosis (ranging from +30% to + 60%) was observed in the 15S clone. Release of arachidonic acid from loaded cells was promoted by 24 h incubation with the cytokine, and a higher degree of release was measured in the 15S clone. These data indicate that endogenous intracellular lipocortin 1 is involved in the promotion of apoptosis in cells of the myelo-monocytic derivation.

Circulating autoantibodies to recombinant lipocortin-1 in asthma.

One of the postulated mechanisms of corticosteroid action is through the de novo synthesis and release of lipocortins. We assayed circulating antibodies to lipocortin-1 in sera obtained from normal (n = 67) and asthmatic (n = 57) subjects using an ELISA technique. Asthmatic subjects with a wide range of severity, with the mildest needing only occasional inhaled beta-agonist therapy to the most severe needing maintenance oral corticosteroid treatment, were recruited from our Asthma Clinic and classified into five categories according to the need of therapy. Median values of IgM and IgG lipocortin-1 antibody for normal subjects were 19.3 (interquartile range (r) = 11.0-30.4) and 16.9 (r = 10.54-29.4) ELISA units (EU) ml-1 respectively. These levels were significantly elevated in asthmatic subjects: IgM = 43.9 EU ml-1 (r = 31.7-64.5) and IgG = 29.0 EU ml-1 (r = 21.2-44.7) (P less than 0.001). There was no significant relationship between the levels of lipocortin antibody and the clinical severity of asthma. Asthmatics with significantly raised levels of antibody were found within all five categories of severity. We conclude that the level of this antibody is not related to severity of asthma, to previous or current corticosteroid therapy or to the development of corticosteroid resistance.

Lipocortin 1 binding to human leukocytes correlates with its ability to inhibit IgG interactions with Fc gamma receptors.

The anti-inflammatory protein lipocortin 1 has been proposed as a mediator of some of the anti-inflammatory actions of glucocorticoid hormones. In view of reported glucocorticoid effects on leukocyte Fc gamma receptors, the effect of short-term lipocortin 1 pre-incubation on expression and IgG binding capacity of human Fc gamma receptors was examined in vitro. The formation of erythrocyte-antibody rosettes, binding of fluoresceinated IgG ligand and the expression of three defined types of Fc gamma receptors were observed following lipocortin 1 treatment. Maximal inhibition of EA rosetting (70%) by peripheral blood mononuclear cells and a 30-50% inhibition of binding of fluoresceinated human IgG1 to purified human monocytes occurred in the presence of 400 nM lipocortin 1 (p < 0.01). However, there was no accompanying decrease in expression of the three known Fc gamma receptor types measured by specific monoclonal antibodies. Similar observations were made for peripheral blood polymorphonuclear cells. On the other hand, IgG binding was not inhibited by lipocortin 1 in lymphocytes or in a panel of cell lines which express Fc gamma receptors and none of these cell types had the capacity to bind lipocortin.

Impairment of neutrophil Fc gamma receptor mediated transmembrane signalling in active rheumatoid arthritis.

Neutrophil Fc gamma receptor (Fc gamma R) signalling responses were compared in healthy subjects, patients with definite rheumatoid arthritis (RA), ankylosing spondylitis, and osteoarthritis. The patients with A were subdivided into those with active synovitis and those with quiescent disease. Basal intracellular calcium ion concentrations in patients with inactive RA were significantly higher than in control subjects, which in turn were greater than in patients with active RA. Transient cytosolic calcium ion fluxes were observed after binding Fc gamma RII or Fc gamma RIII with specific monoclonal antibodies and cross linking with the F(ab')2 fragment of antimouse IgG. Response times were significantly faster for Fc gamma RII than for Fc gamma RIII. Peak concentrations of intracellular calcium ions after neutrophil stimulation were comparable for Fc gamma RII and RIII in healthy subjects. Neutrophils in patients with ankylosing spondylitis and osteoarthritis responded to Fc gamma R triggering, but in the group with active RA fluxes of calcium ions were severely depressed. Neutrophils isolated from patients with RA with quiescent disease showed exaggerated responses when compared with controls. Expression of all three Fc gamma R types on neutrophils from patients with active RA, as measured by monoclonal antibody binding, was comparable with control cells. Impairment of neutrophil Fc gamma R cytosolic signalling in active RA could reflect a receptor signalling defect with potential effects on Fc mediated functions, or a fundamental defect in calcium ion homeostasis within these cells.

Regulation of inflammation by lipocortin 1.

The mechanisms that regulate inflammatory responses, and their dysfunction in chronic inflammatory diseases, are poorly understood. Lipocortin 1 is a potential mediator of the anti-inflammatory effects of glucocorticoid hormones. Following the discovery of putative receptors on phagocytes for lipocortin 1, Nick Goulding and Paul Guyre here offer one explanation for the self-limitation of inflammation and the impairment of the mechanism in chronic inflammatory disease.

The ability of murine leukocytes to bind lipocortin 1 is lost during acute inflammation.

Lipocortin 1, a member of the annexin superfamily of calcium and phospholipid binding proteins, mediates some of the anti-inflammatory actions of the glucocorticoid hormones. Lipocortin 1 binds to the surface of murine peripheral blood monocytes and polymorphonuclear leukocytes (Kd estimate 2 x 10(-8) M) but not lymphocytes. Resident peritoneal macrophages exhibit binding (Kd estimate 1.3 x 10(-8) M) but lymphocytes do not. A 95-98% reduction in lipocortin 1 binding was observed to leukocytes obtained from air pouch or peritonitis models of inflammation. When given intravenously, lipocortin 1 binds rapidly to murine leukocytes within 5 min but disappears before 10 min, leaving the binding capacity of the cells unaltered. Modulation of lipocortin 1 binding sites could be an important step in regulating the function of inflammatory cells.

Lack of involvement of lipocortin 1 in dexamethasone suppression of IL-1 release.

The annexin lipocortin 1 is reported to mediate some anti-inflammatory effects of glucocorticoids, but the mechanisms of this mediation are incompletely understood. The involvement of lipocortin 1 in glucocorticoid inhibition of monocyte interleukin 1beta (IL-1beta) release has been investigated. Treatment of peripheral blood monocytes with 2 mug/ml lipopolysaccharide potently increased IL-1beta release (p = 0.001) and dexamethasone (10(-7) M) significantly reduced both resting and stimulated IL-1beta release (p = 0.009). A neutralizing monoclonal antibody to lipocortin 1 (0.5-50.0 mug/ml) was unable to inhibit this effect and recombinant lipocortin 1 (2 x 10(-6) M) and 188aa lipocortin 1 fragment (10(-8)-10(-6) M) had no effect. It is concluded that lipocortin 1 is not involved in the inhibition of monocyte IL-1beta release by glucocorticoids.