A key problem in salient object detection is how to effectively model the semantic properties of salient objects in a data-driven manner. In this paper, we propose a multi-task deep saliency model based on a fully convolutional neural network with global input (whole raw images) and global output (whole saliency maps). In principle, the proposed saliency model takes a data-driven strategy for encoding the underlying saliency prior information, and then sets up a multi-task learning scheme for exploring the intrinsic correlations between saliency detection and semantic image segmentation. Through collaborative feature learning from such two correlated tasks, the shared fully convolutional layers produce effective features for object perception. Moreover, it is capable of capturing the semantic information on salient objects across different levels using the fully convolutional layers, which investigate the feature-sharing properties of salient object detection with a great reduction of feature redundancy. Finally, we present a graph Laplacian regularized nonlinear regression model for saliency refinement. Experimental results demonstrate the effectiveness of our approach in comparison with the state-of-the-art approaches.
The androgen receptor (AR) is a key driver of prostate cancer (PC), even in the state of castration-resistant PC (CRPC) and frequently even after treatment with second-line hormonal therapies such as abiraterone and enzalutamide. The persistence of AR activity via both ligand-dependent and ligand-independent mechanisms (including constitutively active AR splice variants) highlights the unmet need for alternative approaches to block AR signaling in CRPC. We investigated the transcription factor GATA-binding protein 2 (GATA2) as a regulator of AR signaling and an actionable therapeutic target in PC. We demonstrate that GATA2 directly promotes expression of both full-length and splice-variant AR, resulting in a strong positive correlation between GATA2 and AR expression in both PC cell lines and patient specimens. Conversely, GATA2 expression is repressed by androgen and AR, suggesting a negative feedback regulatory loop that, upon androgen deprivation, derepresses GATA2 to contribute to AR overexpression in CRPC. Simultaneously, GATA2 is necessary for optimal transcriptional activity of both fulllength and splice-variant AR. GATA2 colocalizes with AR and Forkhead box protein A1 on chromatin to enhance recruitment of steroid receptor coactivators and formation of the transcriptional holocomplex. In agreement with these important functions, high GATA2 expression and transcriptional activity predicted worse clinical outcome in PC patients. A GATA2 small molecule inhibitor suppressed the expression and transcriptional function of both full-length and splice-variant AR and exerted potent anticancer activity against PC cell lines. We propose pharmacological inhibition of GATA2 as a firstin-field approach to target AR expression and function and improve outcomes in CRPC.prostate cancer | small molecule inhibitor | AR signaling | GATA2 | steroid receptor coactivator
BackgroundMixed lineage leukemia (MLL) gene translocations are found in ~75 % infant and 10 % adult acute leukemia, showing a poor prognosis. Lysine-specific demethylase 1 (LSD1) has recently been implicated to be a drug target for this subtype of leukemia. More studies using potent LSD1 inhibitors against MLL-rearranged leukemia are needed.MethodsLSD1 inhibitors were examined for their biochemical and biological activities against LSD1 and MLL-rearranged leukemia as well as other cancer cells.ResultsPotent LSD1 inhibitors with biochemical IC50 values of 9.8–77 nM were found to strongly inhibit proliferation of MLL-rearranged leukemia cells with EC50 of 10–320 nM, while these compounds are generally non-cytotoxic to several other tumor cells. LSD1 inhibition increased histone H3 lysine 4 (H3K4) methylation, downregulated expression of several leukemia-relevant genes, induced apoptosis and differentiation, and inhibited self-renewal of stem-like leukemia cells. Moreover, LSD1 inhibitors worked synergistically with inhibition of DOT1L, a histone H3 lysine 79 (H3K79) methyltransferase, against MLL-rearranged leukemia. The most potent LSD1 inhibitor showed significant in vivo activity in a systemic mouse model of MLL-rearranged leukemia without overt toxicities. Mechanistically, LSD1 inhibitors caused significant upregulation of several pathways that promote hematopoietic differentiation and apoptosis.ConclusionsLSD1 is a drug target for MLL-rearranged leukemia, and LSD1 inhibitors are potential therapeutics for the malignancy.Electronic supplementary materialThe online version of this article (doi:10.1186/s13045-016-0252-7) contains supplementary material, which is available to authorized users.
Histone lysine methylation regulates gene expression and cancer initiation. Bioinformatics analysis suggested that DOT1L, a histone H3-lysine79 (H3K79) methyltransferase, plays a potentially important role in breast cancer. DOT1L inhibition selectively inhibited proliferation, self-renewal, metastatic potential of breast cancer cells and induced cell differentiation. In addition, inhibitors of S-adenosylhomocysteine hydrolase (SAHH), such as neplanocin and 3-deazaneplanocin, also inhibited both H3K79 methylation and proliferation of breast cancer cells in vitro and in vivo. The activity of SAHH inhibitors was previously attributed to inhibition of H3K27 methyltransferase EZH2. However, inhibition of EZH2 by a specific inhibitor did not contribute to cell death. SAHH inhibitors had only weak activity against H3K27 methylation and their activity is therefore mainly due to DOT1L/H3K79 methylation inhibition. Overall, we showed that DOT1L is a potential drug target for breast cancer.
Methylation of histone lysine residues plays important roles in gene expression regulation as well as cancer initiation. Lysine specific demethylase 1 (LSD1) is responsible for maintaining balanced methylation levels at histone H3 lysine 4 (H3K4). LSD1 is a drug target for certain cancers, due to important functions of methylated H3K4 or LSD1 overexpression. We report the design, synthesis and structure activity relationships of 3-(piperidin-4-ylmethoxy)pyridine containing compounds as potent LSD1 inhibitors with Ki values as low as 29 nM. These compounds exhibited high selectivity (>160×) against related monoamine oxidase A and B. Enzyme kinetics and docking studies suggested they are competitive inhibitors against a dimethylated H3K4 substrate and provided a possible binding mode. The potent LSD1 inhibitors can increase cellular H3K4 methylation and strongly inhibit proliferation of several leukemia and solid tumor cells with EC50 values as low as 280 nM, while they had negligible effects on normal cells.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.