DSGAT: predicting frequencies of drug side effects by graph attention networks.

A critical issue of drug risk-benefit evaluation is to determine the frequencies of drug side effects. Randomized controlled trail is the conventional method for obtaining the frequencies of side effects, while it is laborious and slow. Therefore, it is necessary to guide the trail by computational methods. Existing methods for predicting the frequencies of drug side effects focus on modeling drug-side effect interaction graph. The inherent disadvantage of these approaches is that their performance is closely linked to the density of interactions but which is highly sparse. More importantly, for a cold start drug that does not appear in the training data, such methods cannot learn the preference embedding of the drug because there is no link to the drug in the interaction graph. In this work, we propose a new method for predicting the frequencies of drug side effects, DSGAT, by using the drug molecular graph instead of the commonly used interaction graph. This leads to the ability to learn embeddings for cold start drugs with graph attention networks. The proposed novel loss function, i.e. weighted $\varepsilon$-insensitive loss function, could alleviate the sparsity problem. Experimental results on one benchmark dataset demonstrate that DSGAT yields significant improvement for cold start drugs and outperforms the state-of-the-art performance in the warm start scenario. Source code and datasets are available at https://github.com/xxy45/DSGAT.

A neighborhood-regularization method leveraging multiview data for predicting the frequency of drug-side effects.

MOTIVATION: A critical issue in drug benefit-risk assessment is to determine the frequency of side effects, which is performed by randomized controlled trails. Computationally predicted frequencies of drug side effects can be used to effectively guide the randomized controlled trails. However, it is more challenging to predict drug side effect frequencies, and thus only a few studies cope with this problem. RESULTS: In this work, we propose a neighborhood-regularization method (NRFSE) that leverages multiview data on drugs and side effects to predict the frequency of side effects. First, we adopt a class-weighted non-negative matrix factorization to decompose the drug-side effect frequency matrix, in which Gaussian likelihood is used to model unknown drug-side effect pairs. Second, we design a multiview neighborhood regularization to integrate three drug attributes and two side effect attributes, respectively, which makes most similar drugs and most similar side effects have similar latent signatures. The regularization can adaptively determine the weights of different attributes. We conduct extensive experiments on one benchmark dataset, and NRFSE improves the prediction performance compared with five state-of-the-art approaches. Independent test set of post-marketing side effects further validate the effectiveness of NRFSE. AVAILABILITY AND IMPLEMENTATION: Source code and datasets are available at https://github.com/linwang1982/NRFSE or https://codeocean.com/capsule/4741497/tree/v1.

Silica nanoparticles induce ovarian granulosa cell apoptosis via activation of the PERK-ATF4-CHOP-ERO1α pathway-mediated IP3R1-dependent calcium mobilization.

Ambient particulate matters (PMs) have adverse effects in human and animal female reproductive health. Silica nanoparticles (SNPs), as a major component of PMs, can induce follicular atresia via the promotion of ovarian granulosa cell apoptosis. However, the molecular mechanisms of apoptosis induced by SNPs are not very clear. This work focuses on revealing the mechanisms of ER stress on SNP-induced apoptosis. Our results showed that spherical Stöber SNPs (110 nm, 25.0 mg/kg b.w.) induced follicular atresia via the promotion of granulosa cell apoptosis by intratracheal instillation in vivo; meanwhile, SNPs decreased the viability and increase apoptosis in granulosa cells in vitro. SNPs were taken up and accumulated in the vesicles of granulosa cells. Additionally, our results found that SNPs increased calcium ion (Ca2+) concentration in granulosa cell cytoplasm. Furthermore, SNPs activated ER stress via an increase in the PERK and ATF6 pathway-related protein levels and IP3R1-dependent calcium mobilization via an increase in IP3R1 level. In addition, 4-PBA restored IP3R1-dependent calcium mobilization and decreased apoptosis via the inhibition of ER stress. The ATF4-C/EBP homologous protein (CHOP)-ER oxidoreductase 1 alpha (ERO1α) pathway regulated SNP-induced IP3R1-dependent calcium mobilization and cell apoptosis via ATF4, CHOP, and ERO1α depletion in ovarian granulosa cells. Herein, we demonstrate that ER stress cooperated in SNP-induced ovarian toxicity via activation of IP3R1-mediated calcium mobilization, leading to apoptosis, in which the PERK-ATF4-CHOP-ERO1α pathway plays an essential role in ovarian granulosa cells.

Silica Nanoparticles Promote Apoptosis in Ovarian Granulosa Cells via Autophagy Dysfunction.

Although silica nanoparticles (SNPs) are generally thought to be biocompatible and safe, the adverse effects of SNPs were also reported in previous studies. SNPs cause follicular atresia via the induction of ovarian granulosa cell apoptosis. However, the mechanisms for this phenomenon are not well understood. This study focuses on exploring the relationship between autophagy and apoptosis induced by SNPs in ovarian granulosa cells. Our results showed that 25.0 mg/kg body weight (b.w.)/intratracheal instillation of 110 nm in diameter spherical Stöber SNPs caused ovarian granulosa cell apoptosis in follicles in vivo. We also found that SNPs mainly internalized into the lumens of the lysosomes in primary cultured ovarian granulosa cells in vitro. SNPs induced cytotoxicity via a decrease in viability and an increase in apoptosis in a dose-dependent manner. SNPs increased BECLIN-1 and LC3-II levels, leading to the activation of autophagy and increased P62 level, resulting in the blockage of autophagic flux. SNPs increased the BAX/BCL-2 ratio and cleaved the caspase-3 level, resulting in the activation of the mitochondrial-mediated caspase-dependent apoptotic signaling pathway. SNPs enlarged the LysoTracker Red-positive compartments, decreased the CTSD level, and increased the acidity of lysosomes, leading to lysosomal impairment. Our results reveal that SNPs cause autophagy dysfunction via lysosomal impairment, resulting in follicular atresia via the enhancement of apoptosis in ovarian granulosa cells.

Activation of Autophagy by Low-Dose Silica Nanoparticles Enhances Testosterone Secretion in Leydig Cells.

Silica nanoparticles (SNPs) can cause abnormal spermatogenesis in male reproductive toxicity. However, the toxicity and toxicological mechanisms of SNPs in testosterone synthesis and secretion in Leydig cells are not well known. Therefore, this study aimed to determine the effect and molecular mechanism of low doses of SNPs in testosterone production in Leydig cells. For this, mouse primary Leydig cells (PLCs) were exposed to 100 nm Stöber nonporous spherical SNPs. We observed significant accumulation of SNPs in the cytoplasm of PLCs via transmission electron microscopy (TEM). CCK-8 and flow cytometry assays confirmed that low doses (50 and 100 µg/mL) of SNPs had no significant effect on cell viability and apoptosis, whereas high doses (more than 200 µg/mL) decreased cell viability and increased cell apoptosis in PLCs. Monodansylcadaverine (MDC) staining showed that SNPs caused the significant accumulation of autophagosomes in the cytoplasm of PLCs. SNPs activated autophagy by upregulating microtubule-associated protein light chain 3 (LC3-II) and BCL-2-interacting protein (BECLIN-1) levels, in addition to downregulating sequestosome 1 (SQSTM1/P62) level at low doses. In addition, low doses of SNPs enhanced testosterone secretion and increased steroidogenic acute regulatory protein (StAR) expression. SNPs combined with rapamycin (RAP), an autophagy activator, enhanced testosterone production and increased StAR expression, whereas SNPs combined with 3-methyladenine (3-MA) and chloroquine (CQ), autophagy inhibitors, had an opposite effect. Furthermore, BECLIN-1 depletion inhibited testosterone production and StAR expression. Altogether, our results demonstrate that low doses of SNPs enhanced testosterone secretion via the activation of autophagy in PLCs.

BECLIN-1-Mediated Autophagy Suppresses Silica Nanoparticle-Induced Testicular Toxicity via the Inhibition of Caspase 8-Mediated Cell Apoptosis in Leydig Cells.

Accumulation of silica nanoparticles (SNPs) in the testes leads to male reproductive toxicity. However, little is known about the effect and mechanistic insights of SNP-induced autophagy on apoptosis in Leydig cells. In this study, we aimed to verify the role of SNP-induced autophagy in apoptosis and explore the possible underlying mechanism in mouse primary Leydig cells (PLCs). H&E staining showed that SNPs changed the histological structures of the testes, including a reduction in the Leydig cell populations in vivo. CCK-8 assay showed that SNPs decreased cell viability, and flow cytometry showed that SNPs increased cell apoptosis, both in a dose-dependent manner in vitro. Additionally, Western blotting further found that SNPs activated autophagy by an increase in BECLIN-1, ATG16L, and LC3-II levels and promoted the intrinsic pathway of apoptosis by an increase in the BAX/BCL-2 ratio, cleaved the caspase 8 and caspase 3 levels. Furthermore, autophagy decreased SNP-induced apoptosis via regulation of the caspase 8 level combined with rapamycin, 3-methyladenine, and chloroquine. BECLIN-1 depletion increased the caspase 8 level, leading to an increase in SNP-induced cell apoptosis. Collectively, this evidence demonstrates that SNPs activated BECLIN-1-mediated autophagy, which prevented SNP-induced testicular toxicity via the inhibition of caspase 8-mediated cell apoptosis in Leydig cells.