Chercheurs


 Inflammatory Mechanisms in the CNS:Targets for Therapeutic Intervention.  


British Inflammation Research Association (BIRAs).  

December 1st 1995, The Governors Hall, St Thomas' Hospital, London  

Meeting report : 

Over 80 delegates, mostly from the U.K. but also from Continental Europe and the USA, attended this meeting, which was held in the historic Governors Hall of St Thomas' Hospital. Until relatively recently, inflammatory aspects of neurobiology have been little explored. However, there is now increasing interest in the developments of anti-inflammatory strategies for the treatment of stroke, head injury, and chronic neurodegenerative disease (see recent review by Rothwell & Hopkins, 1995, TINS, 18, 130-136). The distinguished panel of meeting speakers included representatives from industry and academia including two from the USA and one from Denmark, reinforcing the international flavour of the meeting. The meeting was organised by Adrian Payne (Institut de Recherche Jouveinal, Paris, France), Neville Punchard (University of Luton, Luton, U.K.) and Sandra Amor (St Thomas' Hospital, London, U.K.) The following report is a condensate of the meeting abstracts and subsequent discussion following the various presentations. ...

In the opening talk, Nancy Rothwell (University of Manchester) gave an overview of the current understanding of the role of cytokines in CNS inflammation. Cytokines are more usually associated with peripheral immune and inflammatory responses but virtually every cytokine so far identified has been found in the brain. Indeed, both pro-inflammatory and anti-inflammatory cytokines and their receptors are synthesised by resident brain cells; glia, neurones and endothelial cells, and also by peripheral immune cells which can enter the brain. It has been established that these molecules act as important mediators of many neu-roimmune interactions and of responses to systemic inflammation and disease that are controlled by the CNS.

Furthermore, they may be involved in relationships between disease, stress or psychological status and immune or endocrine status. Recent findings have shown that cytokines are directly involved in local inflammatory responses in the brain. They have also been implicated in many acute and chronic neurological diseases. In particular the role of specific cytokines (IL-1, TNF-a,_b and IFNg) in multiple sclerosis is well established. This probably explains the benefit of IFNb in this disease. In addition, some cytokines (notably IL-1) apparently mediate acute inflammatory and neurodegenerative responses to stroke, brain trauma, infection, heat stroke and excitotoxic insults although they may not in themselves cause overt neuronal death. Marked increases in expression of IL-1, IL-6 and TNF have been observed in rat brains after tissue damage. Although the evidence is still circumstantial, expression of cytokines may also be involved in chronic neurodegenerative disease such as Alzheimers disease, which is now thought to have an inflammatory basis.

Possible therapeutic approaches to mitigating the pro-inflammatory effects of cytokines in the CNS include synthetic IL-1 receptor antagonists (IL-1ra) and inhibition of cytokine production. In animal models, anti-IL-1ra antibody dramatically enhances neuronal damage following ischaemia and IL-1ra has been shown to inhibit cortical damage after middle cerebral artery occlusion. The role of cytokines in the CNS is undoubtably complex and future studies will hopefully determine more precisely their relative contributions to physiological regulation and disease pathology.

Brenda Shivers (Parke-Davis, Ann Arbor) developed further the possible role of IL-1b in CNS inflammation. This molecule is a product of the brains macrophage, the microglia. It is cleaved to a bioactive form by the activity of the cysteine protease, interleukin-1b converting enzyme, otherwise known by the acronym ICE. Inhibition of the activation of microglia, including the inhibition of IL-1b production by ICE inhibitors may reduce the resultant pathophysiology of both acute (e.g. stroke) and chronic (e.g. Alzheimers Disease) neurodegeneration. Studies in Dr Shivers' lab had shown that ICE activity was increased by LPS, quinolinic acid and hypoxia/ischaemia in rat brains and by LPS in mouse microglia. Moreover, peptidic inhibitors of ICE were demonstrated to inhibit IL-1b production from these microglia as well as from human THP-1 monocytes. Importantly, the high degree of structural conservation of ICE between rodents (mice, rats) and man suggests that rodent models of neuroinflammation should be predictive of the efficacy of ICE inhibitors in man. Homologs of ICE (ICH-1, Nedd 2) have recently been shown to be involved in apoptotic cell death. Furthermore, in fas-mediated killing, ICE itself may play an important role. However, the ICE inhibitors referred to above failed to inhibit apoptosis induced by lymphokine withdrawal in a T-cell line and in K+ deprived cerebellar granule neurons. Together, these observations suggest that inhibiting ICE may selectively reduce IL-1b driven pathologies without impacting on cell death initiated by homologs of this cytokine. Indeed, apoptosis was a normal event in the course of development and it was perhaps reassuring that ICE knockout mice developed normally. Experiments were in course to determine the effect of ICE knockout on experimental neurodegeneration and the results were awaited with interest.

On a different tack, Sarah Piddlesden (University of Wales College of Medicine, Cardiff) described the use of complement regulatory molecules in the therapy of experimental demyelination. It is now widely accepted that the complement system plays an important role in the pathogenesis of many neurological diseases. Active fragments released during complement activation and fixation can act as chemotactic agents. The formation of the membrane attack complex, while not only physically disrupting the cell membrane has other important sequelae. These include the stimulation of the release of reactive oxygen metabolites and the induction of cell activation via a rise in the level of intracellular calcium. Certainly complement components seen in the CNS may be derived from the systemic circulation, particularly when the blood brain barrier is disrupted. However, recent work has shown that these components can also be produced by cells within the CNS. A potential role for complement inhibitory protein (CIP) CD59 in the therapy of demyelinating disease of the CNS is suggested by the unusual sensitivity of cultured rat oligodendrocytes to complement mediated lysis, possibly resulting from a relative lack of CIP on their surface. Accordingly, the effects of soluble recombinant human CR1 (sCR1), a CIP acting at an early stage of the complement cascade, were examined in a Lewis rat model of multiple sclerosis (MS). This CIP is a 200kd glycoprotein which binds C3b and C4b and accelerates the decay of C3 and C5 convertases of both pathways of complement activation.

Animals treated with sCR1 showed significant reductions in the levels of clinical disease, inflammation, demyelination and complement deposition. Preliminary studies directed towards reducing the dose of CIP required and in so doing limit its effects to the CNS, have shown that this molecule can be effectively targeted using antibody and furthermore that this targeted construct retains its complement-inhibitory activity despite the increased distance from the site of complement attack on the cell membrane. Even if complement itself proved not to be a primary event in the process of neurodegeneration it certainly had the potential to exacerbate the situation. It would certainly be interesting to monitor whether relapses in MS were correlated with evidence of complement activation. Finally, Dr Piddlesden drew the audiences attention to the fact that herbal tea had recently been found to contain a complement inhibitor, perhaps this should be served at future BIRAs meetings !

In the next presentation, Giti Garthwaite (Wellcome Research Laboratories, Beckenham) introduced to the concept of sodium channel involvement in neurodegenerative processes. Cerebral ischaemia resulting from occlusion or haemorrhage of cerebral arteries, head trauma, cardiac arrest or complications arising from certain surgical procedures can cause death or disability as a result of neuronal loss. Both grey matter (neuronal cell bodies and dendrites) and white matter (myelinated axon tracts) are potentially at risk. In experimental models, antagonists of glutamate receptors (NMDA and AMPA receptors) have been found to be neuroprotective, suggesting that glutamate is an important player in the pathoge-nesis of ischaemic neurodegeneration. Unfortunately such agents have their shortcomings. They interfere with normal synaptic transmission and furthermore, they cannot protect white matter, which is devoid of glutamate receptors. Superior anti-ischaemic agents would ideally protect both grey and white matter without interfering with either synaptic transmission or axonal conduction. In this regard, a recently identified alternative target for neuroprotection was the voltage dependent sodium channel. Compounds which block this channel, such as BW 1003C87 and BW 618C89 (now in phase II for stroke) and in so doing, inhibit glutamate release, have been shown to be at least as effective as glutamate antagonists in animal models of focal and transient global (4 vessel occlusion) ischaemia. These agents also conserve cellular ATP.

Two potentially important pathological mechanisms which may explain the neuroprotective effect of sodium channel blockers have recently been identified. The first of these involved the use of an in vitro model of white matter ischaemia. Isolated rat optic nerves were exposed to a period of glucose and oxygen deprivation and then allowed to recover 1-2h. As a consequence, substantial axonal degeneration occurs, which is dependent on extracellular sodium and calcium, consistent with excessive sodium influx, followed by calcium loading due to reversal of the sodium/calcium exchanger. Classical sodium channel blockers like tetrodotoxin, as well as more selective compounds such as BW 619C89 and related compounds dose-dependently inhibit the resulting axonopathy. Hence, under conditions in which glutamate does not participate, Na+ channels appear to be a primary transducer of neurodegeneration. The second mechanism was identified using primary cultures of rat striatum. When these cultures were exposed briefly to NMDA, the neurones underwent a delayed gradual degeneration over several hours (50% degeneration in 9h) resulting from the persistent secondary release of glutamate, which subsequently stimulates the generation of nitric oxide through acting at NMDA receptors. Tetrodotoxin strongly inhibited glutamate release, nitric oxide generation, and cell death. As one might expect, similar results were obtained with BW 619C89. Thus Na+ channels may mediate neurodegeneration and cell death in two ways, by directly mediating the loss of transmembrane ionic homeostasis and, by inducing the release of glutamate in sufficient quantities to provoke the generation of toxic amounts of nitric oxide. As BW 619C89 and related compounds inhibit Na+ channels in a state-dependent manner (unlike tetrodotoxin) they are able to have neuroprotective properties without interfering with either synaptic transmission or axonal conduction.

Cerebral embolism is the most common cause of acute ischaemic stroke in man. Consequently, there is much current interest in the therapeutic potential of thrombolytics in this situation. Per Meden (Neurovascular Research Laboratory, Copenhagen) opened the afternoon session by describing a model of embolic stroke that he and co-workers had developed in rats. In this model, white clots, made in vitro from autologous blood, are injected into the cerebral circulation. Autoradiographic studies showed that the subsequent embolisation initially causes severe ischaemia in an area of variable size, surrounded by a larger penumbra. Intervention with thrombolytic therapy abolished the presence of areas of severe ischaemia, but not those of moderate ischaemia (cerebral blood flow 10-30 ml/ 100g/min ), which were still present in a significant proportion of the hemisphere affected.

Two days after embolisation the areas of ischaemic tissue damage appeared either hyperaemic or they displayed a core in which there was no blood flow. The size of this core was reduced when arterial blood pressure was raised. Hyperaemia was associated with less severe degrees of tissue damage compared to that in areas with a core of no blood-flow. Perivascular infiltration of leukocytes was abundant in hyperaemic areas. In some cases perivascular infiltration of monocytes was found in tissue adjacent to a large infarct, which otherwise displayed no other signs of tissue damage. The plugging of cerebral capillaries by leukocytes during early reperfusion may exaggerate tissue damage and could explain the neuroprotective effect of antibodies directed towards adhesion molecules. Experience with this experimental model strongly suggested that in thromboembolic stroke, the therapeutic potential of thrombolytics with/without anti-inflammatory therapy was very likely to be dependent on the time of administration after the initial infarct.

In the following presentation, Keith Muir (Institute of Neurological Sciences, Southern General Hospital, Glasgow) reviewed the current status of clinical trials of neuroprotective agents in stroke. Such trials stemmed from the recognition that stroke was an evolving, rather than completed, event. Referring first to the previous presentation, Dr Muir commented that three out of five major trials with thrombolytics had been stopped because of an increase in mortality. With streptokinase, the key limitation appeared to be the incidence of intracranial haemorrhage. More recent studies in which thrombolytics were given 3h after the infarct were suggestive of a drop in mortality. Dr Muir reminded us that while the excitotoxic hypothesis of neurodegeneration was generally accepted, many assumptions based on animal models, particularly the existence in man of a salvageable penumbra, remained unproven at this time. However it seemed reasonably certain that a key event in the process of neurodegeneration following ischaemia was an overload of cellular calcium.

Currently the most advanced drugs in clinical development for stroke were antagonists of the glutamate NMDA receptor. These studies stemmed from the consistent observation of a reduction cerebral infarct volumes in animal studies with such molecules. Also of relevance was the observation that brain glutamate levels rose over 1,000 fold during temporal lobe resection. Competitive and non competitive antagonists for the glutamate and glycine sites, polyamine site and magnesium are currently being trialled (table 1). Drugs which appear to influence presynaptic glutamate release, notably Na+ channel blockers such as BW 619C89 (see above) are also in advanced development. Few of these compounds were without dose-limiting side effects, of which potentially the most problematic were CNS disturbances (NMDA antagonists) and were cardiovascular (QT interval increases; Na+ channel blockers). Cardiovascular side effects, notably hypotension, probably explained why calcium antagonists like nimodipine had not been found to be effective despite being the subject of 18 trials in 3500 stroke patients.

Agents with effects on "downstream" processes, such as nitric oxide synthase inhibitors, gangliocides, free radical scavengers and anti-inflammatory agents either remain in preclinical development or have yet to demonstrate clinical efficacy. Key issues in relation to the design and interpretation of clinical trials in stroke included :

Only when these issues were adequately addressed would we be able to get a clear feel for the real therapeutic potential of current and future neuroprotective agents in stroke.

 

(special table to be supplied later)

 

In the penultimate presentation, Bruce Lyeth (Medical College of Virginia, Richmond, USA) described the pathomechanisms in traumatic brain injury (TBI) and the current options for therapy. TBI is a significant health problem that produces death and disability as well as considerable social and financial burdens. As in stroke, there are complex injury processes encompassing structural and neurochemical alterations in brain function. Diffuse axonal injury, a hallmark of TBI is a progressive process evolving for days after the initial injury and culminates in deafferentation. TBI is often accompanied by secondary insults such as raised intracranial pressure, and pyrexia. These secondary insults can themselves initiate biochemical cascades that lead to further brain damage. Furthermore, TBI is often associated with sites of vascular haemorrhage and more subtle alterations in blood brain barrier permeability together with altered cerebrovascular autoregulation. Key mechanistic aspects of TBI are acidosis, cytokine (IL-1, IL-6) release due to TNF_a production, glutamate release and liberation of free radicals. However the relative importance of these aspects and their exact inter-relationship is still not wholly understood. Trials with calcium antagonists (nimodipine) and free radical scavengers (polyethylene-glycol linked superoxide dismutase) have so far proved ineffective and definitive results with other therapies (glutamate antagonists, inhibitors of lipid pero-xidation, hypothermia) are awaited. Notable in the discussion that followed this presentation was the lurid description by the speaker and certain members of the audience alike of their varied experiences of patients with TBI. The most memorable of these descriptions were of one unfortunate individual who presented with a metal spike through his head and of another with a self-inflicted bullet wound straight through his cranium. Remarkably perhaps, both individuals were conscious and apparently lucid after sustaining their injuries.

Peter Goadsby (Institute of Neurology, London) brought the meeting to its conclusion with a presentation on slightly different theme, addressing the issue of whether the role of inflammation in migraine was overemphasised. Migraine is a common and debilitating disorder characterised by episodic, sometimes throbbing, severe headache associated with one or more of the following: nausea, sensitivity to light, noise; or head movement. The anatomy and physiology of the pain-sensitive innervation of the cranium has been characterised relatively recently and there is much current interest in identifying and developing new therapies for prophylactic and therapeutic use in migraine. This interest has been fuelled further by the recent clinical application of selective 5-HT- 1-like receptor agonists such as sumitriptan in migraine. Moskowitz (Ann.Rev. Med. 1993, 44, 145-154) has provided an elegant series of experiments to suggest that the pain of migraine may result from a form of sterile neurogenic inflammation. Stimulation of the trigeminal ganglion in rats leads to neurogenic plasma extravasation. This process can be blocked by a range of compounds including ergot alkaloids, indomethacin, acetylsalicylic acid, 5HT-1-like agonists, GABA agonists, neurosteroids, substance P antagonists and the endothelin antagonist bosentan and so appears relatively non-specific and possibly not predictive of clinical efficacy. To support this contention, bosentan has recently been found to be ineffective in a double blind placebo-controlled trial. Similarly the substance P antagonist RPR 100893 has also been found to be no more effective than placebo. Migraine may in fact result from a failure of inhibitory neuronal control mechanisms leading to referred vascular pain. Dr Goadsby ended his presentation by drawing attention to the possible occurrence of migraine headache in various forms of extraterrestial beings. With the consolidation that is now occurring within the industry at present, this possibility (and the corresponding market opportunities for migraine and other disorders) should certainly not be taken lightly. At the very least, research into this eventuality should be highly fundable !

Adrian N. Payne

Dept. Allergie-Inflammation Institut de Recherche Jouveinal 94265 Fresnes Cedex

.MenuAsmanetSuiteCongrès Conçue et réalisée par: Michel Godard (at) TopTop
Date de création: 12 Juillet1995 - Dernière mise à jour: 21/07/98
Le secret des correspondances transmises sur le réseau Internet n'est pas garanti.
Toute personne citée dispose d'un droit d'accès, de modification, de rectification et de suppression des données le concernant (art. 34 de la loi "Informatique et Libertés" n° 78-17 du 6 janvier 1978). Pour l'éxercer adressez-vous à Michel Godard (at)