Inhibitors of Rho-kinase modulate amyloid- beta (A beta ) secretion but lack selectivity for A beta 42

• Published in: Journal of neurochemistry 2006-01, Vol.96 (2), p.355-365
• Main Authors: Leuchtenberger, Stefanie, Kummer, Markus P, Kukar, Thomas, Czirr, Eva, Teusch, Nicole, Sagi, Sarah A, Berdeaux, Rebecca, Pietrzik, Claus U, Ladd, Thomas B, Golde, Todd E, Koo, Edward H, Weggen, Sascha
• Format: Article
• Language: English
• ISSN: 0022-3042; 1471-4159
• DOI: 10.1111/j.1471-4159.2005.03553.x

Certain non-steroidal anti-inflammatory drugs (NSAIDs) preferentially inhibit production of the amyloidogenic A beta 42 peptide, presumably by direct modulation of gamma -secretase activity. A recent report indicated that NSAIDs could reduce A beta 42 by inhibition of the small GTPase Rho, and a single inhibitor of Rho kinase (ROCK) mimicked the effects of A beta 42-lowering NSAIDs. To investigate whether A beta 42 reduction is a common property of ROCK inhibitors, we tested commercially available compounds in cell lines that were previously used to demonstrate the A beta 42-lowering activity of NSAIDs. Surprisingly, we found that two ROCK inhibitors reduced total A beta secretion in a dose-dependent manner but showed no selectivity for A beta 42. In addition, ROCK inhibitors did not increase A beta 38 secretion in cell-based assays or reduce A beta production in gamma -secretase in vitro assays, which are critical characteristics of A beta 42-lowering NSAIDs. The reduction in total A beta levels by ROCK inhibitors was not accompanied by overall-changes in amyloid precursor protein processing. Targeting ROCK by expression of dominant-negative or constitutively active ROCK mutants failed to modulate A beta secretion, indicating that ROCK inhibition may either be redundant or insufficient for A beta reduction by ROCK inhibitors. Taken together, these results seem to exclude a mechanistic involvement of ROCK in the A beta 42-lowering activity of NSAIDs.