Roderick J. Flower

Measurement of lipocortin 1 levels in murine peripheral blood leukocytes by flow cytometry: modulation by glucocorticoids and inflammation

1 Lipocortin 1 (LC1) immunoreactivity in murine peripheral blood leukocytes was quantified by use of a flow cytometric technique associated with a permeabilisation protocol with saponin. Using specific antisera raised against the whole protein or against its N‐terminus peptide, cell‐associated LC1‐like immunoreactivity was easily detected in circulating neutrophils and monocytes, whereas very low levels were found in lymphocytes. Of the total protein measured 17.6% and 36% were associated with the external plasma membrane in neutrophils and monocytes, as assessed in the absence of cell permeabilisation, whereas no signal was detected on lymphocyte plasma membrane. 2 Treatment of mice with dexamethasone (Dex; 0.5‐5 μg per mouse corresponding to ∼ 0.015‐1.5 mg kg−1) increased LC1 levels in neutrophils and monocytes. The 2–3 fold increase in LC1 levels was time‐dependent with a peak at 2 h. Treatment of mice with the steroid antagonist, RU486 (two doses of 20 mg kg−1 orally) decreased LC1‐like immunoreactivity in all three types of circulating leukocytes by ≥50%. 3 Extravasation of blood neutrophils into inflamed tissue sites resulted in a consistent reduction (≥50%) in LC1 levels compared with circulating neutrophils. A high LC1‐like immunoreactivity was also measured in resident macrophages, of which approximately one third was membrane‐associated. Induction of an acute inflammatory response in the murine peritoneal cavity did not modify total LC1 levels measured in macrophages, but reduced membrane‐associated LC1 to a significant extent, i.e. up to 70%. 4 In conclusion, flow cytometric analysis is a rapid and convenient method for detecting and measuring LC1 in murine leukocytes. We confirmed that LC1 protein expression is controlled by exogenous and endogenous glucocorticoids. Amongst other factor(s) influencing protein concentrations, extravasation was found to be associated with a reduced LC1 expression in the emigrated cells.

Effect of interleukin-1β on the release of substance P from rat isolated spinal cord

Superfusion of rat spinal cord slices with rat interleukin-1 beta resulted in a significant enhancement of electrically evoked substance P-like immunoreactivity with a maximal effect (> 2-fold increase) at 0.1 ng/ml, whereas higher concentration (10-50 ng/ml) of the cytokine inhibited (approximately 50%) the release of the neuropeptide. Interleukin-1 beta (0.1 ng/ml) potentiation of substance P-like immunoreactivity release was abrogated by co-perfusion with interleukin-1 receptor antagonist (10-100 ng/ml) or with indomethacin (1 microM). Superfusion of spinal cord with interleukin-1 beta inhibited electrically evoked calcitonin gene-related peptide-like immunoreactivity release. Modulation of substance P-like immunoreactivity release from the spinal cord by interleukin-1 beta may represent a mechanism responsible for the hyperalgesic action of the cytokine characteristic of the inflammatory response.

Acute inflammatory response in the mouse: exacerbation by immunoneutralization of lipocortin 1.

1 An immuno‐neutralization strategy was employed to investigate the role of endogenous lipocortin 1 (LC1) in acute inflammation in the mouse. 2 Mice were treated subcutaneously with phosphate‐buffered solution (PBS), non‐immune sheep serum (NSS) or with one of two sheep antisera raised against LC1 (LCS3), or its N‐terminal peptide (LCPS1), three times over a period of seven days. Twenty four hours after the last injection several parameters of acute inflammation were measured including zymosan‐induced inflammation in 6‐day‐old air‐pouches, zymosan‐activated serum (ZAS)‐induced oedema in the skin, platelet‐activating factor (PAF)‐induced neutrophilia and interleukin‐1β (IL‐1β‐induced corticosterone (CCS) release. 3 At the 4 h time‐point of the zymosan inflamed air‐pouch model, treatment with LCS3 did not modify the number of polymorphonuclear leucocytes (PMN) recruited: 7.84 ± 1.01 and 7.00 ± 0.77 × 106 PMN per mouse for NSS‐ and LCS3 group, n = 7. However, several other parameters of cell activation including myeloperoxidase (MPO) and elastase activities were increased (2.2 fold, P < 0.05, and 6.5 fold, P < 0.05, respectively) in the lavage fluids of these mice. Similarly, a significant increase in the amount of immunoreactive prostaglandin E2 (PGE2; 1.81 fold, P < 0.05) and IL‐1α (2.75 fold, P < 0.05), but not tumour necrosis factor‐α (TNF‐α), was also observed in LCS3‐treated mice. 4 The recruitment of PMN into the zymosan inflamed air‐pouches by 24 h had declined substantially (4.13 ± 0.61 × 106 PMN per mouse, n = 12) in the NSS‐treated mice, whereas high values were still measured in those treated with LCS3 (9.35 ± 1.20 × 106 PMN per mouse, n = 12, P < 0.05). A similar effect was also found following sub‐chronic treatment of mice with LCPS1: 6.48 ± 0.10 × 106 PMN per mouse, vs. 2.77 ± 1.20 and 2.64 ± 0.49 × 106 PMN per mouse for PBS‐ and NSS‐treated groups (n = 7, P < 0.05). Most markers of inflammation were also increased in the lavage fluids of LCS3‐treated mice: MPO and elastase showed a 2.47 fold and 17 fold increase, respectively (P < 0.05 in both cases); TNF‐α showed a 11.1 fold increase (P < 0.05) whereas the IL‐1α levels were not significantly modified. PGE2 was still detectable in most (5 out of 7) of the mice treated with LCS3 but only in 2 out of 7 of the NSS‐ treated mice. 5 Intradermal injection of 50% ZAS caused a significant increase in the 2 hoedema formation in the skin of LCS3‐treated mice in comparison to PBS‐ and NSS‐treated animals: 16.7 ± 1.5 μl vs. 10.8 ± 1.2 μl and 10.2 ± 1.0 μl, respectively (n = 14 mice per group, P < 0.05). ZAS‐induced oedema had subsided by 24 h in control animals but a residual significant amount of extravasation was still detectable in LCS3‐ treated mice: 4.4 ± 0.8 μl (P < 0.05). 6 A recently described model driven by endogenous glucocorticoids is the blood neutrophilia observed following administration of PAF. In our experimental conditions, a single bolus of PAF (100 ng, i.v.) provoked a marked neutrophilia at 2 h (2.43 and 2.01 fold) in NSS‐ and PBS‐treated mice (n = 11), respectively, which was significantly attenuated in the animals treated with LCS3: 1.26 fold increase in circulating PMN (n = 11, P < 0.01 vs. NSS‐ and PBS‐groups). 7 Intraperitoneal injection of IL‐1β (5 μg kg−1) caused a marked increase in circulating plasma CCS by 2 h, to a similar extent in all experimental groups. In contrast, measurement of CCS levels in the plasma of mice bearing air‐pouches inflamed with zymosan revealed significant differences between LCS3 and NSS‐treated mice at the 4 h time‐point: 198 ± 26 ng ml−1 vs. 110 ± 31 ng ml−1 (n = 8, P < 0.05). 8 In conclusion, we found a remarkable exacerbation of the inflammatory process with respect to both humoral and cellular components in mice passively immunised against LC1, suggesting the existence of a negative modulatory role for this protein in the normal regulation of the host defence mechanism.

Selective inhibition of neutrophil function by a peptide derived from lipocortin 1 N-terminus

A multi-faceted approach was used to investigate the effect of an anti-inflammatory peptide derived from human lipocortin 1 N-terminus region (amino acid 2-26; termed human Ac2-26) on human neutrophil activation in vitro. When incubated with purified human neutrophils. human Ac2-26 produced a concentration-dependent inhibition of elastase release stimulated by formyl-Met-Leu-Phe (fMLP), platelet-activating factor, or leukotriene B4, with an approximate EC50 of 33 microM (100 micrograms/ml). At this concentration, human Ac2-26 also inhibited (77%) the release of [3H]-arachidonic acid from neutrophils stimulated with fMLP. The peptide, however, did not inhibit the up-regulation of the beta 2-integrin CD11b and the concomitant shedding of L-selectin from neutrophil plasma membrane induced by fMLP. In adhesion experiments, human Ac2-26 inhibited neutrophil adhesion to endothelial monolayers when this was stimulated with fMLP, but not when this followed endothelial cell activation with histamine or platelet-activating factor. Again, the effect of the peptide was concentration-dependent, and an approximate EC50 of 33 microM was calculated. When a preparation of 125I-labeled human Ac2-26 was incubated with the neutrophils, the peptide was internalised in an energy-dependent fashion. All together, these observations lead us to propose a model in which this peptide derived from the N-terminus of human lipocortin 1 alters a common cellular mechanism producing a selective inhibition of neutrophil activation.

Dexamethasone Suppresses the Release of Prolactin from the Rat Anterior Pituitary Gland by Lipocortin 1 Dependent and Independent Mechanisms

Glucocorticoids have been shown repeatedly to inhibit the release of prolactin (PRL) in the rat but their site and mode of action is unknown. In the present study, we used an in vitro model to examine the requirement for protein synthesis for dexamethasone to suppress the release of immunoreactive (ir)-PRL release from the rat pituitary gland. In addition we have performed a series of in vitro and in vivo experiments to investigate the potential role in this regard of lipocortin 1 (LC1), a protein shown previously not only to mediate aspects of the anti-inflammatory and anti-proliferative actions of the glucocorticoids but also to contribute to the regulatory actions of the steroids in the brain-neuroendocrine system. In vitro, the release of ir-PRL from rat anterior pituitary tissue initiated by submaximal concentrations of VIP (10 nM). TRH (10 nM) or the adenyl cyclase activator forskolin (100 microM) was reduced significantly (p < 0.01) by preincubation (2 h) of the tissue with dexamethasone (0.1 microM). By contrast, ir-PRL release evoked by a submaximal concentration of the L-Ca2+ channel opener BAY K8644 (10 microM) was unaffected by the steroid although readily antagonised (p < 0.01) by nifedipine (1-100 microM). Exposure of the pituitary tissue to dexamethasone (0.1 microM) also caused a pronounced and highly significant increase in de novo protein synthesis, as assessed by the incorporation of 14C-lysine into the tissue (p < 0.001). This response was reduced markedly by the inclusion of the RNA and protein synthesis inhibitors, actinomycin-D (0.5 micrograms/ml) or cycloheximide (1.0 micrograms/ml), in the incubation medium (p < 0.001), both of which also effectively abrogated (p < 0.01) the dexamethasone-induced inhibition of the release of ir-PRL evoked by TRH. VIP and forskolin. Lipocortin I was readily detectable by Western blotting in protein extracts of freshly excised anterior pituitary tissue: a small proportion of the protein was found to be attached to the outer surface of the cells where it was retained by a Ca(2+)-dependent mechanism. Exposure of the tissue in vitro to dexamethasone (0.1 microM) or corticosterone (0.1 microM) but not 17 beta-oestradiol (0.1 microM) caused a pronounced increase in the amount of LC1 attached to the outer surface of the cells and concomitant decrease in the LC1 content of the intracellular LC1 pool. Addition of an N-terminal LC1 fragment. LC11-188 (10 pg-10 ng/ml), to the incubation medium reduced significantly (p < 0.01) the increases in ir-PRL release induced in vitro by VIP (10 nM) and forskolin (100 microM). By contrast, at all concentrations tested. LC11-188 (10 pg-10 ng/ml) failed to influence (p < 0.05) the highly significant (p < 0.01) ir-PRL response to TRH (10 nM). Similarly, the inhibitory actions of dexamethasone (0.1 microM) on the release of ir-PRL induced by VIP (10 nM) or forskolin (100 microM) but not by TRH (10 nM) were substantially reversed (p < 0.01) by a specific monoclonal anti-LC1 antibody while an isotype-matched control antibody was without effect. In vivo, rats pretreated with either a polyclonal anti LC1 antiserum (anti-LC1 pAb, 1 ml/day s.c. for 2 days) or a corresponding volume of non-immune sheep serum (NSS) responded to stress (laparotomy under ether anaesthesia) with significant (p < 0.05) increases in the serum ir-PRL concentration. In the NSS-treated group, the ir-PRL response to stress was effectively inhibited by dexamethasone (100 micrograms/kg i.p.) which had no effect on the pre-stress serum ir-PRL concentration. By contrast, in rats pretreated with anti-LC1 pAb dexamethasone failed to block the stress-induced release of ir-PRL. The results show clearly that the inhibitory actions of dexamethasone on PRL release are dependent on de novo protein synthesis and provide novel evidence for the involvement of both LC1-dependent and LC1-independent mechanisms.

Modulation of the Hypothalamo-Pituitary-Adrenocortical Responses to Cytokines in the Rat by Lipocortin 1 and Glucocorticoids: A Role for Lipocortin 1 in the Feedback Inhibition of CRF-41 Release?

Our recent studies indicate that lipocortin 1 (LC1), a putative second messenger protein for the anti-inflammatory steroids in peripheral tissues, may also contribute to the regulatory actions of the glucocorticoids on the hypothalamo-pituitary-adrenal axis. In the present study we have used in vitro and in vivo models to compare the effects of adrenalectomy, LC1 and dexamethasone on the cytokine-induced secretion of the 41-amino acid corticotrophin releasing factor (CRF-41) and arginine vasopressin (AVP) by the rat hypothalamus. In addition, western blot analysis was used to examine the influence of dexamethasone on the expression of LC1 in the hypothalamus. In vitro, interleukins- (IL-) 1 alpha (100 and 200 pg/ml), 1 beta (0.5 and 1.0 ng/ml) and 8 (0.25-1.0 ng/ml) readily initiated the release of CRF-41 and AVP from hypothalami removed from intact rats. IL-6 (10 and 20 ng/ml) was also an effective CRF-41 secretagogue but, unlike the other interleukins tested, it was ineffective with regard to AVP. Adrenalectomy 7-14 days prior to autopsy increased significantly (p < 0.01) the magnitude of the CRF-41 responses to IL-1 alpha, IL-1 beta and IL-6 but not to IL-8. In contrast however, while hypothalamic tissue from adrenalectomized rats, unlike that from intact animals, responded to IL-6 (5-20 ng/ml) with a pronounced hypersecretion of AVP, the AVP responses to IL-1 alpha and IL-1 beta were largely unaffected by adrenalectomy as too were those to IL-8. The marked increases in CRF-41 and AVP release from hypothalami from adrenalectomized rats initiated in vitro by IL1 alpha, IL-1 beta, IL-6 and IL-8 were readily overcome by preincubation of the tissue with dexamethasone (10(-7) M). In addition, the steroid caused externalization of two species of immunoreactive (ir-) LC1 (37 and 58 kDa) by the hypothalamic cells but failed to influence the total LC1 content of the tissue. The inhibitory effects of dexamethasone on the cytokine-induced release of CRF-41 in vitro were mimicked by LC1 (10 ng/ml) which alone had no effect on the basal release of the peptide. However, unlike dexamethasone, LC1 failed to influence the concomitant release of AVP from the hypothalamic tissue elicited by IL1 alpha, IL-1 beta or IL-8 and potentiated that evoked by IL-6.(ABSTRACT TRUNCATED AT 400 WORDS)

The local anti-inflammatory action of dexamethasone in the rat carrageenin oedema model is reversed by an antiserum to lipocortin 1

1 A local pre‐injection of 1 μg dexamethasone sodium phosphate strongly inhibited (> 60% inhibition at 3 h; P < 0.001 at all time points) the development of carrageenin‐induced paw oedema in the rat induced by a subplantar injection of 0.1 ml, 2% carrageenin. 2 Coinjection of a polyclonal rabbit antiserum raised against human 1–188 recombinant lipocortin 1, which also recognised the rat protein, reversed the inhibitory action of dexamethasone (P < 0.05 at 4 h and 5 h). At the highest volume used (40 μl) control antisera were without any effect. 3 These data further support the concept that lipocortin 1 is involved in the anti‐inflammatory mechanism of action of the glucocorticoids.

Serum corticosterone, interleukin-1 and tumour necrosis factor in rat experimental endotoxaemia: comparison between Lewis and Wistar strains.

1 Circulating corticosterone, interleukin‐1 (IL‐1) and tumour necrosis factor‐α (TNFα) activities in serum of Lewis and Wistar rats were measured following injection of lipopolysaccharide (LPS). IL‐1 was measured as ‘lymphocyte activation factor’ (LAF) activity following precipitation of inhibitory activity with polyethylene glycol. TNFα activity was measured as cytotoxic activity. 2 Compared to the Wistar, the Lewis rat had higher circulating LAF and TNF activities following LPS, and release of both cytokines was prolonged in this strain. 3 Corticosterone increases in response to LPS were less in the Lewis than in the Wistar rat following the initial peak at 1 h; basal corticosterone was lower in the Lewis rat. 4 Adrenalectomized Lewis rats had even greater amounts of circulating LAF and TNF activities following LPS than did intact animals; the effect of adrenalectomy was not however mimicked by acute treatment with the steroid receptor antagonist, RU486, suggesting that endogenous corticosteroids did not acutely control cytokine release. 5 Although in vivo administration of anti‐murine IL‐1α antiserum significantly lowered LAF activity of serum, circulating corticosterone in response to LPS was not affected. Similarly, treatment with anti‐murine TNFα monoclonal antibody (mAb) abrogated TNF activity without affecting corticosterone, suggesting that other mediators may be responsible for corticosterone release following LPS. 6 This ‘overproduction’ of inflammatory cytokines together with lower circulating corticosterone may contribute to the susceptibility of the Lewis rat to diseases such as adjuvant arthritis or experimental allergic encephalomyelitis.

Immunoneutralization of Lipocortin 1 Reverses the Acute Inhibitory Effects of Dexamethasone on the Hypothalamo-Pituitary-Adrenocortical Responses to Cytokines in the Rat in vitro and in vivo

Our recent studies suggest that lipocortin 1 (LC1), a potential mediator of the anti-inflammatory, antiproliferative and anti-fever actions of glucocorticoids in peripheral tissues, may also contribute to the powerful negative feedback actions of the steroids on the hypothalamo-pituitary-adrenal (HPA) axis. In the present study we have used (1) an in vitro model to examine the influence of a specific neutralizing monoclonal anti-LC1 antibody (anti-LC1 mAb) on the capacity of dexamethasone to suppress the cytokine-induced release of the 41-amino acid corticotropin-releasing factor (CRF-41) and arginine vasopressin (AVP) from the rat hypothalamus and (2) a passive immunization protocol to assess the contribution of LC1 to the inhibitory actions of dexamethasone on the HPA responses to immunological (i.p. injection of interleukin 1 beta, IL-1 beta) and surgical (laparotomy under ether anaesthesia) stress. In vitro, Il-1 alpha (0.2 ng/ml), IL-1 beta (0.5 ng/ml), IL-6 (10 ng/ml) and IL-8 (1 ng/ml) each caused significant increases in the release of immunoreactive (ir)-CRF-41 and ir-AVP from hypothalami removed from rats adrenalectomized 10-12 days before autopsy; these responses were readily inhibited by preincubation of the tissue with dexamethasone (10(-7) M). The inhibitory actions of the steroid were attenuated and, in many instances, abolished by inclusion in the medium of a monoclonal anti-LC1 antibody (LC1 mAb, diluted 1:15,000); an isotype-matched control antibody (antispectrin alpha+beta, diluted 1:15,000) was ineffective in this regard. IL-1 alpha (0.2 ng/ml), IL-1 beta (0.5 ng/ml) and IL-6 (10 ng/ml) also initiated similar increases in the release of CRF-41 and AVP from hypothalami from intact rats which were effectively blocked by dexamethasone (10(-7) M). However, although the inhibitory actions of the steroid on the pharmacologically evoked release of CRF-41 were specifically overcome by anti-LC1 mAb (diluted 1:15,000), the steroid blockade of AVP release was not. In vivo, rats pretreated with either a polyclonal anti-LC1 antibody (anti-LC1 pAb, 1 ml/day s.c. for 2 days) or a corresponding volume of a nonimmune sheep serum (NSS) responded to immunological (IL-1 beta, 3 micrograms/kg i.p.) or surgical (laparotomy under ether anaesthesia) trauma with significant increases in the serum ACTH and corticosterone concentrations. In the NSS-treated groups, dexamethasone (100 micrograms/kg), which had no effect on the prestress concentrations of ACTH and corticosterone in the blood, completely prevented the HPA responses to both IL-1 beta and laparotomy.(ABSTRACT TRUNCATED AT 400 WORDS)

CYTOKINES, GLUCOCORTICOIDS AND NEUROENDOCRINE FUNCTION

Activation of the immune system is normally associated with widespread alterations in neuroendocrine activity, the profile of which depends upon the species and on the severity and duration of the stimulus. Particularly important in this regard is the activation of the hypothalamo-pituitary-adrenocortical (HPA) axis for the consequent rise in circulating glucocorticoids serves to contain the ensuing pathophysiological responses and thus to restore homeostasis. In the present study, in vivo and in vitro techniques have been used to examine the influence of various immunokines on the HPA axis and to determine whether their actions are modulated by glucocorticoids and lipocortin 1 (LC1). In vivo interleukin-1 beta (IL-1 beta), given orally or peripherally, produced increases (P < 0.01) the serum corticosterone concentration which were reversed by pretreatment with dexamethasone. IL-1 beta also produced glucocorticoid reversible increases in the release of the two corticotrophin releasing factors, CRF-41 and AVP, from the hypothalamus in vitro (P < 0.01) as also did IL-1 alpha, IL-6 and IL-8. By contrast, none of these cytokines influenced directly the release of ACTH from pituitary tissue in vitro. The inhibitory actions of the glucocorticoids on the HPA responses to cytokines observed in vivo and in vitro were mimicked by LC1 and reversed by neutralizing anti-LC1 antisera. Our results demonstrate a role of cytokines, glucocorticoids and LC1 in effecting the interplay between the brain-neuroendocrine and immune system which may be critical to host defence in conditions of both health and disease.