Ocular albumin fluorophotometric quantitation of endotoxin-induced vascular permeability

Blocking the inflammasome: A novel approach to treat uveitis

Uveitis is a complex ocular inflammatory disease often accompanied by bacterial or viral infections (infectious uveitis) or underlying autoimmune diseases (non-infectious uveitis). Treatment of the underlying infection along with corticosteroid-mediated suppression of acute inflammation usually resolves infectious uveitis. However, to develop more effective therapies for non-infectious uveitis and to better address acute inflammation in infectious disease, an improved understanding of the underlying inflammatory pathways is needed. In this review, we discuss the disease aetiology, preclinical in vitro and in vivo uveitis models, the role of inflammatory pathways, as well as current and future therapies. In particular, we highlight the involvement of the inflammasome in the development of non-infectious uveitis and how it could be a future target for effective treatment of the disease.

Innate immune signals in autoimmune and autoinflammatory uveitis

Although traditionally the adaptive immune system has been viewed as the essential contributor to autoimmune diseases, the evidence implicating the innate immune system has grown considerably in recent years. Several multisystem inflammatory diseases affect the uvea and occur as a result of a mutation in a gene coding for a component of the innate immune system. Diseases associated with uveitis such as ankylosing spondylitis, sarcoidosis, Behcet's disease and inflammatory bowel disease can best be conceptually understood by hypotheses that consider microbial infection and innate immunity as contributing factors.

Evaluation of therapeutic effectiveness using MR imaging in a rabbit model of anterior uveitis

Magnetic resonance imaging (MRI) has been used to examine conditions that alter the permeability of the blood-retinal barrier. Our goal was to determine if blood-aqueous barrier permeability could be similarly assessed, because MRI offers the theoretical advantage of providing quantitative data directly from inflamed uveal tissues rather than from the aqueous humor into which the inflammatory reaction spills. As an additional challenge, we sought to use MRI to measure differences between the inflamed uveal tissues of corticosteroid-treated and placebo-treated uveitic eyes. Anterior uveitis was induced in one eye of eight rabbits by subcutaneous injection of Mycobacterium tuberculosis, followed after 10 days with intravitreal challenge. One rabbit of each pair was treated with topical 1% prednisolone acetate while control rabbits were treated with artificial tears. Contrast-enhanced MRI studies were performed prior to uveitis induction, one day after induction and then weekly for at least 2 weeks. MR image data were analyzed to determine percent change in peak enhancement of the ciliary body and anterior chamber. The initial rate of change of enhancement of the anterior chamber was also measured. Extensive contrast agent-induced MR image enhancement of both the anterior chamber and the ciliary processes was measured following the induction of uveitis. More rapid improvement was measured for the 1% prednisolone acetate-treated rabbit eyes (P < 0.001). MR signal enhancement data obtained from the ciliary processes proved to be the most reliable indicator of disease activity in this rabbit model of uveitis. Such data can only be obtained using MRI.

A laboratory evaluation of the Kowa laser flare-cell meter for the study of uveitis

The Kowa FC-1000 laser flare-cell meter (LFCM) has been described as an instrument which will objectively quantify inflammation of the anterior chamber of the eye. We evaluated the LFCM using the intravenous endotoxin-induced uveitis (EIU) rabbit model of ocular inflammation. In vitro flare and cell calibration measurements utilizing bovine serum albumin (BSA) and latex particles, respectively, were also performed. A linear relationship between the flare measurements and BSA concentrations was noted. In addition, the time course of the LFCM flare count in EIU was comparable to previously published fluorophotometric data. However, the LFCM reported cells in the anterior chamber of the EIU rabbits despite negative cytology and histology results. The LFCM also recorded cells in BSA solutions which contained neither cells nor latex particles. Our results suggest that although the LFCM may be useful for evaluating flare, its cell measurements are not accurate in cases of severe uveitis.

Ocular permeability after systemic administration of endotoxin in humans

Acute anterior uveitis in response to the administration of systemic gram-negative endotoxin was studied in humans. The blood-aqueous barrier was evaluated in eight normal human subjects at 8 or 24 hours after systemic administration of purified gram-negative endotoxin. No significant changes in the blood-aqueous barrier were found, as evaluated by permeability to fluorescein, number of aqueous cells, flare, or intraocular pressure, despite profound endotoxin-induced cardiac, pulmonary, and circulatory effects. Gram-negative endotoxin does not appear to affect the human blood-aqueous barrier in doses that can safely be given to humans.

Role of bacterial debris in inflammatory diseases of the joint and eye

Several distinct rheumatic conditions (including Lyme arthritis, Reiter's syndrome and rheumatic fever) as well as certain forms of the blinding disease, uveitis, may share a common etiology. In each instance specific bacterial pathogens may infect a distant site, which on interaction with the immune system, leads to a sterile inflammation in the joint or eye. These "reactive" conditions may result, in some cases, from prior localization of non-viable bacterial remnants (including the cell wall or peptidoglycan) or alternatively "dormant" fastidious bacteria in the affected joint or eye where they act as persisting antigens. Classical culture techniques, would not detect the presence of these putative microbial antigens. Alternative approaches for detection of ubiquitous components of bacteria in the host (using appropriate chemical, molecular and immunological techniques) are discussed.

Modulation of the blood-aqueous barrier by gram positive and gram negative bacterial cell wall components in the rat and rabbit

Acute anterior uveitis in man is related to Gram negative bacterial infection occurring at sites distant to the eye. This could involve intraocular localization of inflammatory bacterial cell wall constituents. Modulation of the blood-aqueous barrier in rabbit and rat, by muramyl dipeptide (the monomer of peptidoglycan) and lipopolysaccharide (and its monomer lipid A) was studied. The rabbit eye was found to be highly susceptible to MDP and LPS, although without cellular infiltration. In contrast the rat eye was demonstrated to be totally refractory to MDP. The response to LPS in the rat was modest, required high dosages and ocular changes were slow to occur, but cellular infiltration was readily apparent. Since MDP is found in Gram positive (as well as Gram negative) bacterial cell walls it is hypothesized that Gram positive bacteria might also play a role in causing uveitis in man.

Endotoxin-induced uveitis in the rat. A study of the role of inflammation mediators

Intraocular inflammation was induced in the rat by footpad injection of salmonella endotoxin in order to study the influence of chemical inflammation mediators in this uveitis model. Ocular inflammation was assessed 1, 6, 18, 24 and 72 h after endotoxin administration as well as in control rats, by measuring aqueous protein concentration, aqueous inflammatory cell content, and pupillary diameter. Thromboxane B2 (TXB2), prostaglandin E2 (PGE2), prostaglandin F2 alpha (PGF2-alpha), leukotriene B4 (LTB4), and substance P were simultaneously measured in the aqueous humor by radioimmunoassay. Inflammation parameters peaked at 18 h. TXB2 was already significantly elevated at 1 h. PGE2 peak values of 2.7 ng/ml were reached at 18 h. PGF2-alpha was never significantly raised over control values. LTB4 peaked at 18 h, together with a polymorphonuclear peak. Substance P was significantly elevated after 6 h. It is concluded that maximal uveitis in this model occurs at 18 h. TXB2 is an early mediator, and PGE2 is probably implicated in blood-ocular barrier disruption for which levels as high as 2.7 ng/ml in aqueous seem necessary. PGF2-alpha does not play a major role in this model, while LTB4 seems to be the main chemotactic factor for polymorphonuclears (PMNs) in the anterior chamber and substance P is clearly related to pupil miosis.

Pharmacologic effects on the expression of class II histocompatibility antigen in experimental endotoxin-induced uveitis

Systemic dexamethasone, cyclosporine A, colchicine, indomethacin, and cobra venom factor were administered to Lewis rats that were then challenged with 100 micrograms of Salmonella typhimurium endotoxin. The effects of these five pharmacologic agents in endotoxin-induced uveitis were studied by analyzing their effects on ocular inflammation and on the expression of class II antigen of the major histocompatibility complex (MHC) on the epithelium of the ciliary body and iris. All the agents used in our study exhibited anti-inflammatory effects but only dexamethasone was able to inhibit significantly the expression of Ia antigen on the uveal epithelium. This demonstrates a dissociation between the inflammation and class II antigen expression and supports our previous data that class II antigen presentation may precede the inflammatory infiltrates. The alteration of ocular inflammation and especially the variation of Ia antigen expression in the uveal tissue may provide insight into the mechanism of action and the success of different pharmacologic agents used in the treatment of uveitis.

Expression of class II antigen in endotoxin induced uveitis

Uveitis can be induced by systemic or intravitreal administration of endotoxin (lipopolysaccharide, LPS). In this study we correlated the expression of class II antigens (la in rat) of the Major Histocompatibility Complex (MHC), with this experimental model of uveitis. Ia antigen was detected by immunohistochemistry using the Avidin-Biotin-Peroxidase Complex (ABC) method and the monoclonal antibody OX6. Ia antigen was not expressed in normal eyes. However, Ia was expressed in the anterior uvea epithelial cells in all eyes with LPS induced uveitis. This study demonstrates that the ocular Ia expression is a localized process in the anterior uvea in response to systemic or intravitreal LPS. This response appears to be distinct from the action of LPS on macrophage Ia expression, where LPS has been shown to inhibit the induction of Ia antigen in macrophages by gamma interferon.

Endotoxin tolerance diminishes certain antiinflammatory effects of endotoxin

Endotoxin (bacterial lipopolysaccharide, LPS) is paradoxically both inflammatory and antiinflammatory. A single intravenous injection of 100 micrograms Escherichia coli LPS markedly inhibits the inflammatory changes associated with cutaneous reversed passive Arthus (RPA) reactions in New Zealand white rabbits. Polymorphonuclear (PMN) leukocytes from LPS-treated rabbits exhibit diminished responsiveness in vitro to complement (C5) -derived peptides. Repeated injections of LPS render animals "tolerant", that is, refractory to the toxic and inflammatory effects of LPS. We examined whether tolerance would enhance the ability of LPS to inhibit inflammation not attributable to LPS. Surprisingly, as compared with rabbits receiving a single dose of LPS, tolerant rabbits demonstrated greater inflammatory changes (i.e., PMN exudation, vascular permeability) associated with RPA reactions. PMNs from LPS-tolerant rabbits responded in vitro to C5-derived peptides significantly more than PMNs from rabbits that received a single dose of LPS. We speculate that some antiinflammatory effects of LPS require the toxic or inflammatory effects of LPS itself. These observations might relate to the limited efficacy of fever therapy and the variable effects of gram-negative sepsis on functions of human PMNs.

Endotoxin-induced uveitis in the rat: observations on altered vascular permeability, clinical findings, and histology

A single intraperitoneal dose of endotoxin (lipopolysaccharide or LPS) induces an acute inflammatory response in the uveal tract of rats. This inflammation is characterized by a breakdown of the blood/aqueous barrier within 3 hr after the LPS and the subsequent development of clinical disease and a cellular infiltrate. Early change in vascular permeability, clinical, and pathological changes were dose dependent with the two highest doses (100 micrograms or 500 micrograms) producing more severe pathology. Clinical and histopathologic abnormalities peaked at 24 hr and were resolving by 48 hr. Although clinical and histologic changes correlated well, the degree of breakdown of the blood/aqueous barrier at 3 hr failed to predict the extent of the cellular exudate measured by either clinical or histologic criteria. In addition, pharmacologic suppression of the early vascular permeability changes with indomethacin, cyproheptadine, or both agents failed to protect the animals consistently from subsequently developing significant clinical disease or cellular infiltrates on histopathology. LPS-induced uveitis in the rat provides a simple, reproducible model for ocular inflammation without requiring direct eye manipulation. The mediators responsible for the early vascular permeability in this model appear to be distinct from the mediators primarily responsible for the subsequent cellular exudate.

Antiinflammatory effects of endotoxin. Inhibition of rabbit polymorphonuclear leukocyte responses to complement (C5)-derived peptides in vivo and in vitro

Although capable of provoking a variety of inflammatory effects, endotoxin (bacterial lipopolysaccharide) paradoxically has been reported to be antiinflammatory. The authors have found that single intravenous injections of Escherichia coli endotoxin, 24 hours before challenge, inhibit almost completely the vascular permeability changes and exudation of polymorphonuclear leukocytes induced in rabbit skin by reversed passive Arthus reactions. Whereas intravenous injections of endotoxin also caused modest inhibition of the vascular permeability changes induced in rabbit skin by the synthetic peptide N-formyl-methionyl-leucyl-phenylalanine (FMLP), exudation of polymorphonuclear leukocytes was unaffected. Polymorphonuclear leukocytes from rabbits given single injected doses of endotoxin exhibited markedly diminished chemotactic and degranulation responses to complement (C5)-derived peptides in vitro. Responses of these cells to FMLP, however, were normal. These data suggest that selective suppression of polymorphonuclear leukocyte responses to C5-derived peptides accounts, in part, for the antiinflammatory effects of endotoxin.

The effect of endotoxin and endotoxin tolerance on inflammation induced by mycobacterial adjuvant

Peptidoglycan, the substance in mycobacteria thought to be responsible for inducing adjuvant arthritis, and endotoxin (lipopolysaccharide or LPS) share many inflammatory properties. Since repeated administration of LPS produces tolerance, i.e., resistance to the toxic and inflammatory effects of LPS, we tested whether LPS and/or LPS tolerance might influence inflammation due to mycobacterial adjuvant. Male Sprague-Dawley rats were injected with Escherichia coli LPS or saline intraperitoneally and then challenged with 100 micrograms killed Mycobacteria butyricum (adjuvant) in the footpad. A single dose of 100 micrograms LPS three or 24 hours before adjuvant markedly, but transiently, reduced the local footpad swelling that begins within hours of the adjuvant injection and histologically resembles a sterile abscess. Animals that received multiple doses of LPS and were therefore tolerant or animals that received LPS 72 hours before adjuvant demonstrated adjuvant-induced footpad swelling nearly equal to controls. The anti-inflammatory effect of LPS was transient since footpad swelling in all groups was nearly comparable six days after the adjuvant injection and LPS failed to inhibit consistently the arthritis that develops two or more weeks after adjuvant injection. These studies establish that LPS can markedly inhibit the prodrome of adjuvant arthritis (footpad swelling due to M. butyricum), that inhibition of this prodrome does not prevent the subsequent development of arthritis, and that LPS tolerance diminishes this anti-inflammatory effect of LPS.