Cyclosporine A blocks autophagic flux in tubular epithelial cells by impairing TFEB-mediated lysosomal function.

Cyclosporine A (CsA) is an immunosuppressor widely used for the prevention of acute rejection during solid organ transplantation. However, severe nephrotoxicity has substantially limited its long-term usage. Recently, an impaired autophagy pathway was suggested to be involved in the pathogenesis of chronic CsA nephrotoxicity. However, the underlying mechanisms of CsA-induced autophagy blockade in tubular cells remain unclear. In the present study, we observed that CsA suppressed the activation and expression of transcription factor EB (TFEB) by increasing the activation of mTOR, in turn promoting lysosomal dysfunction and autophagy flux blockade in tubular epithelial cells (TECs) in vivo and in vitro. Restoration of TFEB activation by Torin1-mediated mTOR inhibition significantly improved lysosomal function and rescued autophagy pathway activity, suppressing TEC injury. In summary, targeting TFEB-mediated autophagy flux represents a potential therapeutic strategy for CsA-induced nephrotoxicity.

Accelerated Glomerular Cell Senescence in Experimental Lupus Nephritis.

BACKGROUND The aim of this study was to determine whether senescence in renal glomeruli is involved in lupus nephritis (LN); the expression of senescence-associated ß-galactosidase (SA-ß-Gal) and its association with glomerular lesions were investigated in a mouse model of LN. MATERIAL AND METHODS Eighteen MRL/lpr mice with severe proteinuria were randomly divided into 2 equal groups and intraperitoneally injected with dexamethasone (DEX) or saline; 4 age-matched mice with mild proteinuria served as controls. Serum creatinine and urinary protein levels were analyzed, and kidney histological changes were observed by periodic acid-Schiff and Sirius Red staining. SA-ß-Gal was detected via histochemistry. Glomerular expression of collagen IV, α-SMA, and nephrin was analyzed by immunohistochemistry, and glomerular complement C3 deposition was tested by immunofluorescence. The relationships between SA-ß-Gal expression and renal function or glomerular lesion markers were determined by Spearman's correlation analysis. RESULTS Mice with severe proteinuria exhibited glomerular segmental sclerosis and endothelial cell proliferation. DEX administration suppressed these lesions but had no significant effect on 24-hour urinary protein levels. The elevated glomerular expression of SA-ß-Gal in proteinuric mice was attenuated by DEX treatment. In addition, DEX treatment markedly downregulated glomerular C3 deposition and collagen IV and α-SMA expression, while significantly increasing nephrin expression. Furthermore, SA-ß-Gal expression was positively correlated with urinary protein levels and expression of α-SMA. CONCLUSIONS Accelerated senescence of glomerular cells may contribute to glomerular injury in LN.

Joint stereo 3D object detection and implicit surface reconstruction.

We present a new learning-based framework S-3D-RCNN that can recover accurate object orientation in SO(3) and simultaneously predict implicit rigid shapes from stereo RGB images. For orientation estimation, in contrast to previous studies that map local appearance to observation angles, we propose a progressive approach by extracting meaningful Intermediate Geometrical Representations (IGRs). This approach features a deep model that transforms perceived intensities from one or two views to object part coordinates to achieve direct egocentric object orientation estimation in the camera coordinate system. To further achieve finer description inside 3D bounding boxes, we investigate the implicit shape estimation problem from stereo images. We model visible object surfaces by designing a point-based representation, augmenting IGRs to explicitly address the unseen surface hallucination problem. Extensive experiments validate the effectiveness of the proposed IGRs, and S-3D-RCNN achieves superior 3D scene understanding performance. We also designed new metrics on the KITTI benchmark for our evaluation of implicit shape estimation.

FedMix: Mixed Supervised Federated Learning for Medical Image Segmentation.

The purpose of federated learning is to enable multiple clients to jointly train a machine learning model without sharing data. However, the existing methods for training an image segmentation model have been based on an unrealistic assumption that the training set for each local client is annotated in a similar fashion and thus follows the same image supervision level. To relax this assumption, in this work, we propose a label-agnostic unified federated learning framework, named FedMix, for medical image segmentation based on mixed image labels. In FedMix, each client updates the federated model by integrating and effectively making use of all available labeled data ranging from strong pixel-level labels, weak bounding box labels, to weakest image-level class labels. Based on these local models, we further propose an adaptive weight assignment procedure across local clients, where each client learns an aggregation weight during the global model update. Compared to the existing methods, FedMix not only breaks through the constraint of a single level of image supervision but also can dynamically adjust the aggregation weight of each local client, achieving rich yet discriminative feature representations. Experimental results on multiple publicly-available datasets validate that the proposed FedMix outperforms the state-of-the-art methods by a large margin. In addition, we demonstrate through experiments that FedMix is extendable to multi-class medical image segmentation and much more feasible in clinical scenarios. The code is available at: https://github.com/Jwicaksana/FedMix.

VideoPro: A Visual Analytics Approach for Interactive Video Programming.

Constructing supervised machine learning models for real-world video analysis require substantial labeled data, which is costly to acquire due to scarce domain expertise and laborious manual inspection. While data programming shows promise in generating labeled data at scale with user-defined labeling functions, the high dimensional and complex temporal information in videos poses additional challenges for effectively composing and evaluating labeling functions. In this paper, we propose VideoPro, a visual analytics approach to support flexible and scalable video data programming for model steering with reduced human effort. We first extract human-understandable events from videos using computer vision techniques and treat them as atomic components of labeling functions. We further propose a two-stage template mining algorithm that characterizes the sequential patterns of these events to serve as labeling function templates for efficient data labeling. The visual interface of VideoPro facilitates multifaceted exploration, examination, and application of the labeling templates, allowing for effective programming of video data at scale. Moreover, users can monitor the impact of programming on model performance and make informed adjustments during the iterative programming process. We demonstrate the efficiency and effectiveness of our approach with two case studies and expert interviews.

Transferable Interactiveness Knowledge for Human-Object Interaction Detection.

Human-object interaction (HOI) Detection is an important problem to understand how humans interact with objects. In this paper, we explore Interactiveness Knowledge which indicates whether human and object interact with each other or not. We found that interactiveness knowledge can be learned across HOI datasets and alleviate the gap between diverse HOI category settings. Our core idea is to exploit an Interactiveness Network to learn the general interactiveness knowledge from multiple HOI datasets and perform Non-Interaction Suppression before HOI classification in inference. On account of the generalization of interactiveness, interactiveness network is a transferable knowledge learner and can be cooperated with any HOI detection models to achieve desirable results. We utilize the human instance and body part features together to learn the interactiveness in hierarchical paradigm, i.e., instance-level and body part-level interactivenesses. Thereafter, a consistency task is proposed to guide the learning and extract deeper interactive visual clues. We extensively evaluate the proposed method on HICO-DET, V-COCO, and a newly constructed HAKE-HOI dataset. With the learned interactiveness, our method outperforms state-of-the-art HOI detection methods, verifying its efficacy and flexibility. Code is available at https://github.com/DirtyHarryLYL/Transferable-Interactiveness-Network.

Hydroxychloroquine administration exacerbates acute kidney injury complicated by lupus nephritis.

BACKGROUND: Hydroxychloroquine (HCQ) has been recommended as a basic treatment for lupus nephritis (LN) during this decade based on its ability to improve LN-related renal immune-mediated inflammatory lesions. As a classical lysosomal inhibitor, HCQ may inhibit lysosomal degradation and disrupt protective autophagy in proximal tubular epithelial cells (PTECs). Therefore, the final renal effects of HCQ on LN need to be clarified. METHOD: HCQ was administered on spontaneous female MRL/lpr LN mice with severe proteinuria daily for 4 weeks. Moreover, the MRL/lpr mice with proteinuric LN were subjected to cisplatin-induced or unilateral ischemia/reperfusion (I/R)-induced acute kidney injury (AKI) after 2 weeks of HCQ preadministration. RESULTS: As expected, HCQ treatment increased the survival ratio and downregulated the levels of serum creatinine in the mice with LN, ameliorated renal lesions, and inhibited renal interstitial inflammation. Unexpectedly, HCQ preadministration significantly increased susceptibility to and delayed the recovery of AKI complicated by LN, as demonstrated by an increase in PTEC apoptosis and expression of the tubular injury marker KIM-1 as well as the retardation of PTEC replenishment. HCQ preadministration suppressed the proliferation of PTECs by arresting cells in G1/S phase and upregulated the expression of cell cycle inhibitors. Furthermore, HCQ preadministration disrupted the PTEC autophagy-lysosomal pathway and accelerated PTEC senescence. CONCLUSION: HCQ treatment may increase susceptibility and delay the recovery of AKI complicated by LN despite its ability to improve LN-related renal immune-mediated inflammatory lesions. The probable mechanism involves accelerated apoptosis and inhibited proliferation of PTECs via autophagy-lysosomal pathway disruption and senescence promotion.

Prospects of and limitations to the clinical applications of genistein.

Genistein is an isoflavone derived from soy-rich products, which is known to exhibit several beneficial biological effects, such as anti-tumor activity, improvement of glucose metabolism, and reduction of the frequency of peri-menopausal hot flashes, and thus has potential for clinical application. Certain limitations and side effects, such as low bioavailability, biological estrogenic activity, and detrimental effects on thyroid function, have restricted its clinical applications to some extent. Recently, it has been reported that fermentation, use of micromicelles, and modification of its chemical structure can enhance the bioavailability of genistein. Moreover, the modification of its molecular structure may also eliminate its biological estrogenic activity and adverse effects on thyroid function. In this review, we summarize the clinical application prospects and limitations of genistein, as well as the plausible solutions to overcome its low bioavailability, phytoestrogenic activity, and adverse effects on thyroid function.

Asiatic acid protects against cisplatin-induced acute kidney injury via anti-apoptosis and anti-inflammation.

Cisplatin is a well-known chemotherapeutic drug applied for the treatment of numerous human cancers. However, the use of cisplatin in clinic is limited by certain serious side effects, such as nephrotoxicity. Unfortunately, there is currently no effective therapeutic approach to prevent cisplatin-induced AKI. Increasing evidence suggests that apoptosis of tubular epithelial cells and renal inflammation mainly determine the progression and outcome of cisplatin-induced AKI. Asiatic acid (AA) has been reported have the functions of anti-inflammation and anti-apoptosis, etc. But the effects of AA on kidney injury induced by cisplatin are still not known. The current study aimed to determine the potential renoprotective effects of AA on kidney injury induced by cisplatin. Twenty-four C57BL/6 male mice were randomly divided into four groups: normal control (CON), cisplatin-induced AKI (CIS), AKI with 50 mg/kg AA pretreatment (CIS + AA50), and AKI with 100 mg/kg AA pretreatment (CIS + AA100). Mice were anesthetized and sacrificed at 72 h after the cisplatin injection. Blood and kidney samples were collected for analyses. Compared with CON mice, cisplatin-treated mice exhibited severe tubular necrosis and elevated serum creatinine level. However, AA pretreatment (50 mg/kg or 100 mg/kg) markedly suppressed the elevated serum creatinine, blood urea nitrogen and histological changes. Moreover, AA pretreatment notably downregulated tubular expression of kidney injury molecule-1 (KIM-1) and the number of apoptotic cells, and upregulated the expression of the apoptosis inhibitor survivin and promoted tubular proliferation as evidenced by an increase in the number of proliferating cell nuclear antigen-positive cells. In addition, AA suppressed the enhanced mRNA expression of proinflammatory cytokines IL-1ß, TNF-α, MCP-1 and caspase-1 in the kidneys. Furthermore, AA pretreatment inhibited NF-κB activation and the inflammatory response, which may result from Smad7 up-regulation. In conclusion, AA protects against cisplatin-induced AKI via anti-apoptosis and anti-inflammation.