Although the specific pathogenesis of Alzheimer's disease (AD) is not clear, studies from genetics and biochemistry show that amyloid beta precursor protein (APP) and gamma- secretase are the only two mutant genes found in familial AD patients. Additionally, gamma-secretase could also proteolyze the APP substrate to produce Abeta. Therefore, it is important to study the shear process of APP on the membrane in order to develop new treatment of AD and explore its mechanism.
Protein glycosylation was found to regulate the shedding of membrane proteins. Studies have found that the mucin type-O-glycosylation of Notch (another substrate of gamma-secretase) catalyzed by GalNac transferase 11 (GALNT11) is necessary for the shedding and maturation of Notch receptor (Boskovski et al., Nature 2013). Lack of this protein post-modification could cause developmental defects.
Since 2012, we construct a high throughput inhibitor screening assay for GALNT, which could convert the UDP into NADH, a common readout for high-throughput assay, by a tandem-enzyme coupling assay. By screening of about 5,000 compounds, nine GALNT inhibitors were identified, of which luteolin was found to specifically inhibit the activity of GALNT2. Using the newly-discovered inhibitor tool for O-glycosylation, we found that the glycosylation of the APP was controlled by GALNT2, indicating that the APP is a physiological substrate of GALNT2. Moreover, we found that luteolin could reduce the production of Abeta in cells and in APP/PS mice. Luteolin is an effective component of traditional Chinese medicine Xiao Chun Hua Oral Liquid (also known as Xiao Fu Oral Liquid), which may provide a potential remedy for the treatment of AD.
In addition, we also studied the preliminarily structure and activity relationship of luteolin and its biochemical mechanism, and elucidate that the flavonoid core C ring of luteolin could be optimized for better inhibitors. To our knowledge, luteolin is the first substrate competitive inhibitor for GALNT2.
In this study, a new high-throughput screening model for glycosyltransferase was constructed, and the first substrate competitive inhibitor of GALNT2 was discovered. Moreover, GALNT2 was identified as a new target for the natural product of flavonoid, which provides a new type of flavonoid skeleton for further development of flavonoid drugs in addition to the two known flavonoid drug, i.e. Dimefline (a respiratory stimulant) and Flavoxate (smooth muscle relaxant).
The article was published in the Journal of Biological Chemistry (https://www.ncbi.nlm.nih.gov/pubmed/29061849). M.sc Xu Kai (co-first author), teacher Wang Congrong, Zhou Yuanyang and Wu Fang (last correspondence) contribute the above part of the study.