br COX and LOX in post mortem AD brain Minghetti
COX and 5-LOX in post-mortem AD brain Minghetti (2004) reviewed the findings on COX-2 mRNA levels in AD brains, pointing out that the available evidence demonstrated either decreased or increased levels, possibly because of the short half-life of COX-2 transcripts or individual variability. Histological analyses of AD brains have also produced apparently conflicting results. For example, Kitamura et al. (1999) found that in AD brains, protein levels of COX-1 were increased in both cytosolic and particulate fractions and that COX-2 protein was also increased in the particulate fraction. In a study analyzing the post-mortem brains of 45 autopsy subjects without dementia and 25AD patients from the town of Hisayama, Japan, it was found that in AD patients, neurons of CA1 exhibited increased COX-2 immunoreactivity that correlated with the severity of AD pathology (Fujimi et al., 2007). Furthermore, it appears that in AD, changes in COX-1 and COX-2 expression depend on the stage of the disease and the type of cells expressing these enzymes (for review, see Choi et al., 2009, Hoozemans et al., 2008, Minghetti, 2004). Quantitative Western blot assays were used to investigate 5-LOX protein content in the hippocampus and frontal cortex of humans with AD and in matching controls. 5-LOX levels were significantly increased in AD brain samples (Firuzi et al., 2008). Another study investigated the distribution and cellular localization of 5-LOX in the medial temporal lobe from AD and control subjects (Ikonomovic et al., 2008) and found that in AD subjects, 5-LOX immunoreactivity is elevated relative to controls and that its localization is dependent on the antibody-targeted portion of the 5-LOX amino SKF 83566 hydrobromide sequence. Thus, carboxy terminus-directed antibodies detected 5-LOX in glial cells and neurons but less frequently in neurons with dystrophic morphology, whereas immunoreactivity observed using 5-LOX amino terminus-directed antibodies was absent in neurons and abundant in neurofibrillary tangles, neuritic plaques, and glia. Furthermore, double-labeling studies showed a close association of 5-LOX-immunoreactive processes and glial cells with amyloid-beta immunoreactive plaques and vasculature.
AD and transgenic COX and 5-LOX mouse models Various models of transgenic mice have been used to investigate the contribution of COX and 5-LOX to AD (Schulte et al., 2009). Typically, the intent of the models has been to demonstrate that the absence or the overexpression of COX/5-LOX influences AD-like phenotypes. A microarray analysis of gene expression in the cerebral cortex and hippocampus of mice deficient in either COX-1 or COX-2 revealed that the majority (>93%) of the differentially expressed genes in both the cortex and hippocampus were altered in one COX isoform knockout mouse but not the other (Toscano et al., 2007), suggesting that COX-1 and COX-2 differentially modulate brain gene expression and that this type of modulation could impact the development/progression of AD. For example, Choi and Bosetti (2009) investigated the effect of COX-1 genetic deletion on neurodegeneration induced by beta-amyloid peptide 1–42, which was centrally injected in the lateral ventricle of COX-1-deficient mice and their respective wild-type controls. These authors found that in the absence of COX-1 the beta-amyloid-induced damage was attenuated. On the other hand, Melnikova et al. (2006) showed that overexpression of COX-2 in APPswe-PS1dE9 mice (a model of AD pathology) leads to deficits in spatial working memory in female but not male mice. This COX-2-dependent sex-specific deficit was abolished by pharmacological inhibition of COX-2. Furthermore, these authors reported that the effects of COX-2 and amyloid-beta peptides on cognition occurred in a sex-specific manner in the absence of significant changes in amyloid burden. Firuzi et al. (2008) investigated the effect of 5-LOX deficiency on the amyloid-beta pathology of a transgenic mouse model of AD-like amyloidosis, the Tg2576 mice. These authors generated a double transgenic model crossing 5-LOX-deficient with Tg2576 mice and found that the genetic disruption of 5-LOX reduced amyloid-beta deposits and amyloid-beta 42 levels. Furthermore, these changes occurred in the absence of alteration in amyloid-beta precursor protein processing or amyloid-beta catabolism. Additional in-vitro studies showed that 5-LOX activation and its metabolites increased whereas 5-LOX inhibition decreased amyloid-beta formation by modulating the gamma-secretase complex activity (Firuzi et al., 2008). These authors suggested that 5-LOX could participate in AD pathobiology through a mechanism that involves the modulation of the gamma-secretase activity.