Recent studies have demonstrated that specific miRNAs, such as miR-221/222, may be responsible for tamoxifen resistance in breast cancer. Secreted miRNAs enclosed in exosomes can act as intercellular bio-messengers. Our objective is to investigate the role of secreted miR-221/222 in tamoxifen resistance of ER-positive breast cancer cells. Transmission electron microscopy analysis and nanoparticle tracking analysis were performed to determine the exosomes difference between MCF-7(TamR) (tamoxifen resistant) and MCF-7(wt) (tamoxifen sensitive) cells. PKH67 fluorescent labeling assay was used to detect exosomes derived from MCF-7(TamR) cells entering into MCF-7(wt) cells. The potential function of exosomes on tamoxifen resistance transmission was analyzed with cell viability, apoptosis ,and colony formation. MiRNA microarrays and qPCR were used to detect and compare the miRNAs expression levels in the two cells and exosomes. As the targets of miR-221/222, p27 and ERα were analyzed with western blot and qPCR. Compared with the MCF-7(wt) exosomes, there were significant differences in the concentration and size distribution of MCF-7(TamR) exosomes. MCF-7(wt) cells had an increased amount of exosomal RNA and proteins compared with MCF-7(TamR) cells. MCF-7(TamR) exosomes could enter into MCF-7(wt) cells, and then released miR-221/222. And the elevated miR-221/222 effectively reduced the target genes expression of P27 and ERα, which enhanced tamoxifen resistance in recipient cells. Our results are the first to show that secreted miR-221/222 serves as signaling molecules to mediate communication of tamoxifen resistance.

No abstract

Machine learning models, especially neural networks (NNs), have achieved outstanding performance on diverse and complex applications. However, recent work has found that they are vulnerable to Trojan attacks where an adversary trains a corrupted model with poisoned data or directly manipulates its parameters in a stealthy way. Such Trojaned models can obtain good performance on normal data during test time while predict incorrectly on the adversarially manipulated data samples. This paper aims to develop ways to detect Trojaned models. We mainly explore the idea of meta neural analysis, a technique involving training a meta NN model that can be used to predict whether or not a target NN model has certain properties. We develop a novel pipeline Meta Neural Trojaned model Detection (MNTD) system to predict if a given NN is Trojaned via meta neural analysis on a set of trained shadow models.We propose two ways to train the meta-classifier without knowing the Trojan attacker's strategies. The first one, oneclass learning, will fit a novel detection meta-classifier using only benign neural networks. The second one, called jumbo learning, will approximate a general distribution of Trojaned models and sample a "jumbo" set of Trojaned models to train the metaclassifier and evaluate on the unseen Trojan strategies. Extensive experiments demonstrate the effectiveness of MNTD in detecting different Trojan attacks in diverse areas such as vision, speech, tabular data, and natural language processing. We show that MNTD reaches an average of 97% detection AUC (Area Under the ROC Curve) score and outperforms existing approaches. Furthermore, we design and evaluate MNTD system to defend against strong adaptive attackers who have exactly the knowledge of the detection, which demonstrates the robustness of MNTD.

No abstract