Ego4D: Around the World in 3,000 Hours of Egocentric Video.

We introduce Ego4D, a massive-scale egocentric video dataset and benchmark suite. It offers 3,670 hours of daily-life activity video spanning hundreds of scenarios (household, outdoor, workplace, leisure, etc.) captured by 931 unique camera wearers from 74 worldwide locations and 9 different countries. The approach to collection is designed to uphold rigorous privacy and ethics standards, with consenting participants and robust de-identification procedures where relevant. Ego4D dramatically expands the volume of diverse egocentric video footage publicly available to the research community. Portions of the video are accompanied by audio, 3D meshes of the environment, eye gaze, stereo, and/or synchronized videos from multiple egocentric cameras at the same event. Furthermore, we present a host of new benchmark challenges centered around understanding the first-person visual experience in the past (querying an episodic memory), present (analyzing hand-object manipulation, audio-visual conversation, and social interactions), and future (forecasting activities). By publicly sharing this massive annotated dataset and benchmark suite, we aim to push the frontier of first-person perception. Project page: https://ego4d-data.org/.

Si-Ni-San promotes liver regeneration by maintaining hepatic oxidative equilibrium and glucose/lipid metabolism homeostasis.

ETHNOPHARMACOLOGICAL RELEVANCE: The efficacy of clinical treatments for various liver diseases is intricately tied to the liver's regenerative capacity. Insufficient or failed liver regeneration is a direct cause of mortality following fulminant hepatic failure and extensive hepatectomy. Si-Ni-San (SNS), a renowned traditional Chinese medicine prescription for harmonizing liver and spleen functions, has shown clinical efficacy in the alleviation of liver injury for thousands of years. However, the precise molecular pharmacological mechanisms underlying its effects remain unclear. AIMS OF THE STUDY: This study aimed to investigate the effects of SNS on liver regeneration and elucidate the underlying mechanisms. MATERIALS AND METHODS: A mouse model of 70% partial hepatectomy (PHx) was used to analyze the effects of SNS on liver regeneration. Aquaporin-9 knockout mice (AQP9-/-) were used to demonstrate that SNS-mediated enhancement of liver regeneration was AQP9-targeted. A tandem dimer-Tomato-tagged AQP9 transgenic mouse line (AQP9-RFP) was utilized to determine the expression pattern of AQP9 protein in hepatocytes. Immunoblotting, quantitative real-time PCR, staining techniques, and biochemical assays were used to further explore the underlying mechanisms of SNS. RESULTS: SNS treatment significantly enhanced liver regeneration and increased AQP9 protein expression in hepatocytes of wild-type mice (AQP9+/+) post 70% PHx, but had no significant effects on AQP9-/- mice. Following 70% PHx, SNS helped maintain hepatic oxidative equilibrium by increasing the levels of reactive oxygen species scavengers glutathione and superoxide dismutase and reducing the levels of oxidative stress molecules H2O2 and malondialdehyde in liver tissues, thereby preserving this crucial process for hepatocyte proliferation. Simultaneously, SNS augmented glycerol uptake by hepatocytes, stimulated gluconeogenesis, and maintained glucose/lipid metabolism homeostasis, ensuring the energy supply required for liver regeneration. CONCLUSIONS: This study provides the first evidence that SNS maintains liver oxidative equilibrium and glucose/lipid metabolism homeostasis by upregulating AQP9 expression in hepatocytes, thereby promoting liver regeneration. These findings offer novel insights into the molecular pharmacological mechanisms of SNS in promoting liver regeneration and provide guidance for its clinical application and optimization in liver disease treatment.

Aquaporin-7 Facilitates Proliferation and Adipogenic Differentiation of Mouse Bone Marrow Mesenchymal Stem Cells by Regulating Hydrogen Peroxide Transport.

Hydrogen peroxide (H2O2) is a major form of reactive oxygen species that play an important role in the survival, proliferation and differentiation of bone marrow mesenchymal stem cells (BMSCs). The regulatory mechanisms of H2O2 homeostasis in BMSCs are not fully understood. Here we demonstrate for the first time that aquaglyceroporin AQP7 is a functional peroxiporin expressed in BMSCs and remarkably upregulated upon adipodenic induction. The proliferation ability of BMSCs from AQP7-/- mice was significantly decreased, as indicated by fewer clonal formation and cell cycle arrest compared with wildtype BMSCs. AQP7 deficiency caused accumulation of intracellular generated H2O2 during BMSCs proliferation, leading to oxidative stress and inhibition of PI3K/AKT and STAT3 signaling pathways. After adipogenic induction, however, the AQP7-/- BMSCs exhibited greatly reduced adipogenic differentiation with fewer lipid droplets formation and lower cellular triglycerides content than wildtype BMSCs. In such case AQP7 deficiency was found to diminish import of extracellular H2O2 produced by plasma membrane NADPH Oxidases, resulting in altered AMPK and MAPK signaling pathways and reduced expression of lipogenic genes C/EBPα and PPARγ. Our data revealed a novel regulatory mechanism of BMSCs function through AQP7-mediated H2O2 transport across plasma membrane. AQP7 is a peroxiporin mediating H2O2 transport across the plasma membrane of BMSCs. During proliferation, AQP7 deficiency results in accumulation of intracellular generated H2O2 due to reduced export, which inhibited STAT3 and PI3K/AKT/insulin receptor signaling pathways and cell proliferation. During adipogenic differentiation, however, AQP7 deficiency blocked the uptake of extracelluar H2O2 generated through plasma membrane NOX enzymes. The reduced intracellular H2O2 level causes decreased expression of lipogenic genes C/EBPα and PPARγ due to altered AMPK and MAPK signaling pathways, leading to impaired adipogenic differentiation.

The function and mechanism of microRNA-92a-3p in lipopolysaccharide-induced acute lung injury.

OBJECTIVES: Sepsis-associated acute lung injury (ALI) is a clinically severe respiratory disorder and remains the leading cause of multiple organ failure and mortality. Herein, we used lipopolysaccharide (LPS) to generate sepsis-induced ALI and try to explore the role and mechanism of microRNA-92a-3p (miR-92a-3p) in this process. METHODS: Mice were intravenously injected with miR-92a-3p agomir, antagomir and negative controls for 3 consecutive days and then were intratracheally instillated by LPS (5 mg/kg) for 12 h. To knock down the endogenous A-kinase anchoring protein 1 (AKAP1), mice were intratracheally injected with recombinant adenovirus carrying the short hairpin RNA targeting AKAP1 (shAkap1) at 1 week before LPS administration. RESULTS: miR-92a-3p level was significantly upregulated in the lungs by LPS injection. miR-92a-3p antagomir reduced LPS-induced intrapulmonary inflammation and oxidative stress, thereby preventing pulmonary injury and dysfunction. In contrast, miR-92a-3p agomir aggravated LPS-induced intrapulmonary inflammation, oxidative stress, pulmonary injury and dysfunction. Moreover, we reported that AKAP1 upregulation was required for the beneficial effects of miR-92a-3p antagomir, and that AKAP1 knockdown completely abolished the anti-inflammatory and antioxidant capacities of miR-92a-3p antagomir. CONCLUSION: Our data identify that miR-92a-3p modulates LPS-induced intrapulmonary inflammation, oxidative stress and ALI via AKAP1 in mice.

Modality Compensation Network: Cross-Modal Adaptation for Action Recognition.

With the prevalence of RGB-D cameras, multimodal video data have become more available for human action recognition. One main challenge for this task lies in how to effectively leverage their complementary information. In this work, we propose a Modality Compensation Network (MCN) to explore the relationships of different modalities, and boost the representations for human action recognition. We regard RGB/ optical flow videos as source modalities, skeletons as auxiliary modality. Our goal is to extract more discriminative features from source modalities, with the help of auxiliary modality. Built on deep Convolutional Neural Networks (CNN) and Long Short Term Memory (LSTM) networks, our model bridges data from source and auxiliary modalities by a modality adaptation block to achieve adaptive representation learning, that the network learns to compensate for the loss of skeletons at test time and even at training time. We explore multiple adaptation schemes to narrow the distance between source and auxiliary modal distributions from different levels, according to the alignment of source and auxiliary data in training. In addition, skeletons are only required in the training phase. Our model is able to improve the recognition performance with source data when testing. Experimental results reveal that MCN outperforms stateof- the-art approaches on four widely-used action recognition benchmarks.