Antiphospholipid antibodies (aPL) have been suggested to play a role in causing cognitive and behavioral impairments. In the present study we investigated the pathogenic potential of aPL by intracerebro-ventricular (ICV) administration of immunoglobulins (IgG) from patients with antiphospholipid syndrome (APS). IgG, purified from the sera of four APS patients, was tested for binding to normal mouse brain by immunohistological staining. These IgG (7.5 microg) were injected ICV unilaterally to male C3H mice. Mice injected with IgG purified from pooled sera derived from healthy subjects served as controls. The mice were examined neurologically for motor function and coordination, and cognitively in a Morris water maze. The cognitive tests were performed with the experimenter blinded to the treatment. The performance of the mice in four separate experiments was compared by analysis of variance with repeated measures. IgG from one APS patient was found to bind best to neuronal structures in the hippocampus and cerebral cortex. Mice (n = 43) injected with this IgG performed worse in the water maze compared to the controls (n = 45) with significant effects of the aPL IgG on the overall performance of the mice (treatment, P < 0.03), on learning throughout the experiment (treatment x day, P < 0.02) and on short term memory (treatment x day xtrial, P < 0.002). IgG injected from two of the three other patients also bound specifically to mouse brain neurons and produced an impairment in performance of the water maze. These results support the hypothesis that aPL that gain access to the central nervous system may play a direct role in the pathogenesis of neurological manifestations of APS.

Inflammatory demyelinating diseases of peripheral nerves are associated with altered nerve conduction and with activation of the coagulation pathway. Thrombin mediates many of its effects through protease-activated receptor 1 (PAR-1). We examined the possibility that thrombin may mediate conduction abnormalities through PAR-1 on rat sciatic nerve. PAR-1 was found to be present by both RT-PCR and Western blot analysis of the sciatic nerve. Activation of PAR-1 by a specific peptide agonist caused a 3-fold increase in phosphorylated extracellular signal-regulated kinase (ERK) in the sciatic nerve indicating the existence of functional receptors in the nerve. By confocal immunofluoresence microscopy of the sciatic nerve using anti-PAR-1 antibody and double staining for the paranodal marker contactin-associated protein 1 (Caspr1) or the nodal markers gliomedin and ezrin, the receptor was localized predominantly to myelin microvilli at the node of Ranvier. Thrombin and the PAR-1-specific agonist were applied to exposed rat sciatic nerve and their effects on nerve conduction were measured. Thrombin at concentrations of 100 and 200 U/ml and PAR-1 agonists 150 and 300 muM produced a conduction block within 30 min of application. This effect was maintained for at least 1 h and was reversible by washing. The function of the nodal non-compacted myelin is not well known. The current results implicate this structure and PAR-1 activation in the pathogenesis of conduction block in inflammatory and thrombotic nerve diseases.

No abstract

Multiple sclerosis (MS) and its animal model, experimental autoimmune encephalomyelitis (EAE), are inflammatory diseases of the central nervous system (CNS). Activated coagulation factors are associated with inflammation and are elevated in the plasma of animals with EAE. Thrombin is a key coagulation factor and its major endogenous inhibitors are antithrombin III (ATIII) in the plasma and protease nexin 1 (PN-1) in the brain. We measured the capacity of brain homogenates to inhibit exogenous thrombin and the CNS levels of ATIII and PN-1 during the course of EAE. Acute EAE was induced in SJL/J mice by immunization with mouse spinal cord homogenates. On Days 8, 13, and 22 post-immunization, inhibition of exogenous thrombin activity was measured by a recently developed fluorimetric assay. PN-1 and ATIII were assayed both by immunohistochemistry and by immunoblots in the brain and spinal cord. Total brain thrombin inhibitory activity increased (32%) in EAE mice at the peak of clinical disease (Day 13, P=0.04 compared to controls). Brain ATIII also increased at the peak of disease (2.5-fold higher than controls, P=0.0001), and correlated significantly with clinical scores at all stages of disease (r=0.72, P=0.0068). In contrast, PN-1 elevations were more pronounced at the preclinical stage on Day 8 (3-fold higher than controls, P=0.01) than on Day 13 (1.4-fold higher, P=0.005). Increased brain thrombin inhibition at the clinical peak of EAE probably reflects increased influx of plasma thrombin inhibitors. Early PN-1 changes represent a potential target for thrombin modulating drugs in EAE and MS.

Thrombin-like enzymatic activity was measured in mouse brain homogenates and slices by cleavage of a peptide substrate, N-p-Tosyl-Gly-Pro-Arg-7-amido-4-methylcoumarin. The activity was localized mainly to white matter. However, it was not affected by specific thrombin inhibitors, and was found to represent the sum of at least two enzyme activities, a prolyl endopeptidase and an aminopeptidase. By specifically inhibiting this endogenous activity in combination with exogenously added thrombin, mouse brain tissue was shown to express a capacity of thrombin inhibitory activity equivalent to 0.2 mU thrombin/mg brain tissue. The present study offers a simple and reliable method for measuring total thrombin inhibitory activity in brain.