Optimization, characterization of Astragalus polysaccharides, and evaluation of anti-inflammation effect in primary cultured astrocytes via HMGB1/RAGE/NF-κB/NLRP3 signal pathway

The random forest approach combined with spherically symmetric design was applied to optimize the water extraction and alcohol precipitation extraction process of Astragalus polysaccharides (APs). And the evaluation index is calculated by the entropy weight method. The optimal extraction condition was as follows: ratio of water-to-raw material (61:1 mL/g), extraction temperature at 95 ℃, extraction time of 125 min, and the optimum precipitation condition was ethanol concentration of 68%. Two purified polysaccharides (AP1 and AP2) were successfully obtained from APs using ion exchange and gel filtration chromatography. AP1 is composed of arabinose, glucose, galacturonic acid, glucuronic acid; AP2 is made up of arabinose, galactose, glucose, galacturonic acid, and glucuronic acid. The molecular weight of AP1 and AP2 was 7.8 × 103 Da and 7.5 × 103 Da, respectively. The micro-structures analysis showed that AP1 and AP2 had different morphologies structures. The α- or β-configuration of AP1 and AP2 was confirmed by infrared and NMR spectroscopies. In addition, pharmacological research showed that AP1 and AP2 could protected astrocytes against OGD/R-induced neuroinflammation by inhibiting the HMGB1/RAGE/NF-κB/NLRP3 signaling pathway. This research was expected to lay the groundwork for the process optimization based on random forest and clinical application of APs.