30 Mar 2008

No, not road rage, but biological RAGE, aka receptor for advanced glycation end products. RAGE is a cell surface receptor of the immunoglobulin family and it binds Aβ. In this week’s Journal of Neuroscience, researchers led by Luciano Domenici at the Institute of Neuroscience, Pisa, Italy, and Shi Du Yan at Columbia University, New York, report that RAGE mediates the toxic effects of Aβ oligomers in entorhinal cortex neurons. The receptor does so by setting off p38 mitogen activated protein kinase (MAPK), a protein already implicated in Aβ toxicity. The signal cascade somehow limits long-term potentiation, a neuronal response that is essential for learning and memory. The finding may help unravel the myriad of cellular changes that have been attributed to Aβ.

This is not the first time RAGE has been linked to Aβ toxicity. Yan and colleagues previously showed that RAGE binds to Aβ (see Yan et al., 1996) and that the cell surface protein may contribute to pathology in AD transgenic mice (see Arancio et al., 2004). Now, first author Nicola Origlia and colleagues demonstrate that the Aβ/RAGE liaison has some specific consequences. As has been shown by numerous other studies, Origlia and colleagues found that Aβ suppresses LTP in brain slices. While other labs have focused on the hippocampus (see ARF related news story), in this work the researchers chose to focus on the entorhinal cortex (ERC), an area of the brain that is intimately connected to the hippocampus and which is also decimated by AD pathology. Origlia found that adding a preparation of oligomeric Aβ to ERC slices completely suppressed LTP, even at concentrations (200 nM) that do not cause cell death. In slices taken from RAGE-deficient mice, however, Aβ had no effect on LTP and neither did the peptide have any effect on ERC slices pre-treated with an anti-RAGE antibody. The findings suggest that in this particular area of the brain, RAGE can mediate some of the toxic effects of Aβ. These effects seem specific to this type of potentiation since the researchers found that Aβ had no effect on ERC long-term depression or paired-pulse facilitation.

Aβ has previously been linked to LTP through effects on glutamate receptor subtypes (see ARF related news story), but it is not clear if RAGE contributes to glutamate receptor upheaval in the ERC. What does appear fairly certain is that p38 MAPK is involved. Origlia and colleagues found that suppression of LTP by Aβ could be prevented if ERC slices were first treated with the p38 MAPK inhibitor SB203580. A Jun N-terminal kinase (JNK) inhibitor had no effect. They also found that p38 MAPK activation is increased by Aβ, but that this can be suppressed if cultured neurons are treated with anti-RAGE antibodies. Together, the data suggest that Aβ, via RAGE, activates the kinase, setting off some downstream signaling events that prevent LTP.

This particular kinase has been implicated in Aβ toxicity before. Roger Anwyl and colleagues at Trinity College, Dublin, Ireland, reported that a p38 MAPK inhibitor can prevent LTP blockage by Aβ in hippocampal slices (see Wang et al., 2004). The kinase was also found to facilitate Aβ cytotoxicity in primary cortical neurons (see Zhu et al., 2005). Interestingly, Wang and colleagues also found that a JNK inhibitor can prevent LTP suppression by Aβ. That a JNK inhibitor had no effect on Aβ toxicity in the current study suggests that hippocampal and ERC responses to Aβ differ somewhat. Whether this difference reflects susceptibility to AD pathology remains to be determined.

What happens downstream of p38 MAPK activation in response to Aβ is unclear. It is likewise not clear if binding to RAGE is the only upstream event involving Aβ. RAGE, of course, is also found in glia, where it may contribute to inflammatory responses (Aβ may even be involved—see Bodles et al., 2005), but the authors put the blame for LTP deficit squarely on neuronal RAGE. Using a mouse line that expresses a dominant-negative RAGE in neurons alone, the researchers were able to block Aβ suppression of LTP in ERC slices. As the authors write, “…our results indicate that neuronal RAGE is involved in mechanisms underlying Aβ-induced inhibition of LTP, at least in part, through activation of p38 MAPK in mouse cortex.” The finding also gives new weight to the appearance of RAGE around Aβ deposits in AD brain.—Tom Fagan