B-cell lymphoma-2 phosphorylation at Ser70 site-related autophagy mediates puerarin-inhibited the apoptosis of MC3T3-E1 cells during osteoblastogenesis.

OBJECTIVE: To explore the relationship between autophagy and apoptosis regulated by puerarin during osteoblastogenesis. METHODS: In this study, the effects of puerarin on the autophagic activity and apoptosis level of osteoblast precursors (MC3T3-E1 cells) was observed. Subsequently, the roles of puerarin on B-cell lymphoma-2 (Bcl-2) phosphorylation at different sites in osteoblast precursors were observed. The effect of puerarin on the interaction between Bcl-2 and autophagy regulatory molecule or pro-apoptotic molecule was also investigated using Co-immunoprecipitation assays. In addition, the effect of puerarin on mitochondrial membrane potential of osteoblast precursors was also identified by mitochondrial membrane potential fluorescence probe assays. RESULTS: Our results showed that puerarin can promote the autophagic activity and apoptosis level of MC3T3-E1 cells. In addition, puerarin promoted Bcl-2 phosphorylation at Ser70 site, and the dissociation of Bcl-2-Beclin1 complex. Moreover, puerarin could enhance the binding of Bcl-2-Bcl-2-Associated X (Bax) complex in MC3T3-E1 cells. Furthermore, puerarin increased the mitochondrial membrane potential of MC3T3-E1 cells. CONCLUSIONS: Therefore, puerarin promotes Beclin1 into autophagy flux through Bcl-2 phosphorylation at Ser70, thereby enhancing autophagy of osteoblast precursors, which mediates its anti-apoptotic role during osteoblastogenesis. Furthermore, the dissociation of Bcl-2-Beclin1 complex is conducive to the binding of Bcl-2-Bax complex, which resists the apoptosis of osteoblast precursors viathe increased mitochondrial membrane potential.

Poison Ink: Robust and Invisible Backdoor Attack.

Recent research shows deep neural networks are vulnerable to different types of attacks, such as adversarial attacks, data poisoning attacks, and backdoor attacks. Among them, backdoor attacks are the most cunning and can occur in almost every stage of the deep learning pipeline. Backdoor attacks have attracted lots of interest from both academia and industry. However, most existing backdoor attack methods are visible or fragile to some effortless pre-processing such as common data transformations. To address these limitations, we propose a robust and invisible backdoor attack called "Poison Ink". Concretely, we first leverage the image structures as target poisoning areas and fill them with poison ink (information) to generate the trigger pattern. As the image structure can keep its semantic meaning during the data transformation, such a trigger pattern is inherently robust to data transformations. Then we leverage a deep injection network to embed such input-aware trigger pattern into the cover image to achieve stealthiness. Compared to existing popular backdoor attack methods, Poison Ink outperforms both in stealthiness and robustness. Through extensive experiments, we demonstrate that Poison Ink is not only general to different datasets and network architectures but also flexible for different attack scenarios. Besides, it also has very strong resistance against many state-of-the-art defense techniques.