Liver Kupffer cells control the magnitude of the inflammatory response in the injured brain and spinal cord.

The CNS inflammatory response is regulated by hepatic chemokine synthesis, which promotes leukocytosis and facilitates leukocyte recruitment to the site of injury. To understand the role of the individual cell populations in the liver during the hepatic response to acute brain injury, we selectively depleted Kupffer cells (KC), using clodronate-filled liposomes, and assessed the inflammatory response following a microinjection of IL-1beta into the rat brain or after a compression injury in the spinal cord. We show by immunohistochemistry that KC depletion reduces neutrophil infiltration into the IL-1beta-injected brain by 70% and by 50% into the contusion-injured spinal cord. qRT-PCR analysis of hepatic chemokine mRNA expression showed that chemokine expression in the liver after brain injury is not restricted to a single cell population. In non-depleted rats, CXCL-10, IL-1beta, CCL-2, and MIP-1alpha mRNAs were increased up to sixfold more than in KC depleted rats. However, CXCL-1 and MIP-1beta were not significantly affected by KC depletion. The reduction in chemokine mRNA expression by the liver was not associated with decreased neutrophil mobilisation as might have been expected. These findings suggest that in response to CNS injury, KC mediated mechanisms are responsible for increasing neutrophil entry to the site of CNS injury, but that neutrophil mobilisation is dependent on other non-KC mediated events. However, the suppression of KC activity may prevent secondary damage after acute brain injury.

Immunomodulatory effects of etanercept in a model of brain injury act through attenuation of the acute-phase response.

TNF-alpha has proved to be a successful target in the treatment of many peripheral inflammatory diseases, but the same interventions worsen immune-mediated CNS disease. However, anti-TNF-alpha strategies may offer promise as therapy for non-immune CNS injury. In this study, we have microinjected IL-1beta or lipopolysaccharide (LPS) into the rat brain as a simple model of brain injury and have systemically administered the TNF-alpha antagonist etanercept to discover whether hepatic TNF-alpha, produced as part of the acute-phase response to CNS injury, modulates the inflammatory response in the brain. We report a significant reduction in neutrophil numbers recruited to the IL-1beta- or LPS-challenged brain as a result of TNF-alpha inhibition. We also show an attenuation in the levels of hepatic mRNA including TNF-alpha mRNA and of TNF-alpha-induced genes, such as the chemokines CCL-2, CXCL-5, and CXCL-10, although other chemokines elevated by the injury were not significantly changed. The reduction in hepatic chemokine synthesis results in reduced numbers of circulating neutrophils, and also a reduction in the numbers recruited to the liver as a consequence of brain injury. These findings suggest that TNF-alpha inhibitors may reduce CNS inflammatory responses by targeting the hepatic acute-phase response, and thus therapies for brain injury need not cross the blood-brain barrier to be effective.

Post-conditioning with lipopolysaccharide reduces the inflammatory infiltrate to the injured brain and spinal cord: a potential neuroprotective treatment.

Systemic infection often accompanies or precedes acute brain injury, but it remains unclear how the systemic response contributes to outcome. To examine this problem we have microinjected recombinant interleukin-1beta (IL-1beta), a cytokine associated with acute brain injury, into the rat brain parenchyma and either preceded or followed this challenge with the intravenous injection of lipopolysaccharide (LPS), which mimics systemic inflammatory response syndrome. The microinjection of IL-1beta alone into the brain parenchyma gives rise to leukocyte mobilization in the blood, and to the delayed recruitment of neutrophils and monocytes to the brain with no evidence of blood-brain barrier breakdown or overt neuronal cell death. Systemic LPS pre-conditioning resulted in a dose-dependent reduction both in the number of circulating leukocytes and in the number of leukocytes recruited to the brain parenchyma after 12 h. Surprisingly, LPS given two hours after injury was equally effective in reducing the recruitment of leukocytes to the brain, which is more relevant to the management of clinical disease. In a more clinically relevant model of spinal cord injury, intravenous LPS post-conditioning also reduced the numbers of leukocytes mobilized in the blood and recruited to the spinal cord and thus limited the breakdown of the blood-spinal cord barrier. The effects appear to be specific to LPS, as they were not observed after intravenous IL-1beta pre-conditioning. Our studies suggest that individual pro-inflammatory conditioning strategies may protect the injured central nervous system from the damaging consequences of leukocyte recruitment and may provide scope for novel therapeutic intervention.