• cyclooxygenases
• demyelination
• lipoxygenases
• multiple sclerosis

SIGNIFICANCE OF THE STUDY
Multiple sclerosis (MS) is a chronic inflammatory demyelinating disease of the central nervous system (CNS) leading to permanent cognitive and motor disabilities. The etiology of MS is unknown and no effective cure is currently available. Four patterns of MS active demyelinating lesions have been described; patterns I and II are immune-mediated while patterns III and IV seem to be independent of the immune system activation and to be induced by a direct insult to oligodendrocytes. Animal models of MS have been developed to investigate and clarify pathological mechanisms of the disease, and to test novel therapeutic approaches. Patterns I and II lesions are reproduced with the murine model of experimental autoimmune encephalomyelitis (EAE), while patterns III and IV can be reproduced with the cuprizone model of demyelination. The EAE model has been extensively considered to investigate the role of immune-system activation in the disease pathogenesis and progression, and has lead to the development of three immune-modulatory therapies currently used in MS patients such as glatiramer acetate, natalizumab and mitoxantrone. However, EAE has been recently criticized because many therapeutic approaches that showed promising effects in this model, did not translated into effective therapies for MS patients. Besides, the cuprizone model is a model of direct insult to oligodendrocytes without immune-system activation, being suitable for therapeutical trials aimed to reduce demyelination and promote remyelination in types III and IV lesions of MS.
This study aimed to clarify the role of the arachidonic acid cascade in the pathogenesis of cuprizone-induced demyelination, independently by the immune system activation, and to individuate new therapies to reduce demyelination and improve motor disability in MS.
BACKGROUND
Arachidonic acid (AA) is a membrane lipid which is released upon inflammatory stimuli and then further metabolized by cyclooxygenases (COX-1 and -2) and by lipoxygenases (LO-5, -12 and -15) to eicosanoids and leukotriens (LTs), potent mediators of inflammation. In patients with MS, AA derived prostaglandins (PGs) such as PGE2, PGD2, PGF2α, PGI2, were found to be upregulated in active demyelinating plaques and in the cerebral spinal fluid (CSF). Moreover, COX-2 was upregulated and colocalized with macrophages/microglia and dying oligodendrocytes. Also, an increase in the expression of 5-LO, a key enzyme in the biosynthesis of LTs, and in LTs in the CSF of MS patients were reported. Our preliminary data demonstrated that in cuprizone-induced primary demyelination, COX-2 gene expression is upregulated prior to the detection of histological demyelination when oligodendrocytes begin undergoing caspase-3-mediated apoptosis, and that COX-2 is expressed by oligodendrocytes, suggesting that COX-2 could play a causative role in the initiation of demyelination. On the other hand, we found that COX-1 and 5-LO gene and protein expression are upregulated during the histological demyelination along with microglia and astrocyte activation and proliferation suggesting a role in the inflammatory response that accompanies demyelination.
HYPOTHESIS
Since COX-2 gene expression is upregulated and expressed by oligodendrocytes prior to the detection of histological demyelination, we hypothesized that COX-2 plays a causative role in oligodendrocytes apoptosis.
Since COX-1 and 5-LO enzymes are upregulated at the peak of demyelination and glial activation, we hypothesize that COX-1 and 5-LO are involved in the inflammatory response associated with demyelination.
MATERIAL AND METHODS
Cuprizone is a neurotoxicant that induces specific mature oligodendrocyte death and consequent demyelination accompanied by motor dysfunction. Mice treated up to 6 weeks to a 0.2% cuprizone diet develop brain demyelination. Upon cuprizone withdrawal, mice spontaneously remyelinate in 6 weeks.
COX-1 (COX-1-/-) and COX-2 (COX-2-/-) deficient mice and their wild type controls, or C57BL/6 mice treated with celecoxib (COX-2 selective inhibitor) 4000 ppm by the diet, or AH6809 (antagonist of the PGE2-EP2 receptor) 30 mg/kg i.p daily, were exposed to a 0.2% cuprizone diet for 5 weeks to cause demyelination. The MK-886 (5-LO inhibitor) 3 mg/kg was administered i.p. daily only during the pick of histological demyelination (from week 4 to week 5 of cuprizone exposure). Age-matched normal control and a group of animal treated only with cuprizone were used. The degree of demyelination was determined by Black Gold-II staining after cuprizone exposure and in age-matched controls. Inflammatory markers and apoptosis markers were assessed by gene (RT-PCR) and protein (immunohistochemistry, ELISA, western blotting) analysis. Motor coordination and balance were assessed using a rota-rod apparatus at 32 rpm.
RESULTS
We found that COX-1 genetic deletion did not change either demyelination or gene expression of inflammatory markers such as CD11b (microglia marker), GFAP (astrocyte marker) and tumor necrosis factor alpha (TNF-α), indicating that COX-1 does not play a role in the disease pathogenesis and progression or in the associated inflammatory response.
Similarly, 5-LO inhibition using MK886 did not attenuate cuprizone-induced demyelination; however, it reduced microglial activation, IL-6 production, axonal damage and ameliorated motor performance indicating that the 5-LO pathway is mainly involved in microglia activation and neuroinflammation independently of the demyelination process.
Finally, in the early stage of the disease, we found a concomitant increase in the gene expression of COX-2 and PGE2-EP2 receptor, which were expressed by oligodendrocytes, suggesting a causative role for the COX-2/EP2 pathway in the initiation of demyelination. Supporting this hypothesis, COX-2 gene deletion, or chronic treatment with the COX-2 selective inhibitor celecoxib or with the EP2 receptor antagonist AH6809 reduced oligodendrocyte apoptosis caspase-3-mediated, the degree of demyelination and motor dysfunction induced by cuprizone exposure.
CONCLUSION
Overall our data suggest that COX isoforms have different impact on the demyelination process with COX-2 being selectively involved in the initiation and progression of demyelination via the modulation of caspase-3-mediated apoptosis by a PGE2-EP2 receptor signaling mechanism. On the other hand, the 5-LO pathway contributes to microglial activation and neuroinflammation and to axonal damage resulting in motor dysfunction. Thus, COX-2 inhibitors or EP2 receptor antagonists may be useful as an early intervention therapeutic approach in MS to attenuate the degree of demyelination and the loss of motor function while 5-LO inhibition may be a useful symptomatic treatment.