Search

1. Basic characteristics of endometrial hyperplasia and EC patients. Supplementary Table 2. Basic information of clinical samples for Immunohistochemistry. Supplementary Table 3. Primers used.

Supplementary Materials and Methods and Supplementay Figure legends

Persistent unopposed estrogen stimulation is a central oncogenic mechanism driving the formation of type I endometrial cancer. Recent epidemiologic and clinical studies of endometrial cancer have also revealed a role for insulin resistance, clinically manifested by chronic inflammation. However, the role of inflammation in estrogen-driven endometrial cancer is not well characterized. In this study, we investigated the association between infiltrating macrophages and estrogen sensitivity in endometrial cancer. Evaluating tissue samples and serum from patients with precancerous lesions or endometrial cancer, we found that tissue macrophage infiltration, but not serum estradiol levels, correlated positively with endometrial cancer development. Furthermore, IL4/IL13-induced CD68+CD163+ macrophages enhanced the proliferative effects of estradiol in endometrial cancer cells by upregulating estrogen receptor alpha (ERα), but not ERβ. Mechanistic investigations revealed that CD68+CD163+ macrophages secreted cytokines, such as IL17A, that upregulated ERα expression through TET1-mediated epigenetic modulation of the ERα gene. Overall, our findings show how cytokines produced by infiltrating macrophages in the endometrial microenvironment can induce epigenetic upregulation of ERα expression, which in turn sensitizes endometrial cells to estrogen stimulation. The concept that inflammation-induced estrogen sensitivity in the endometrium acts as a driver of type I endometrial cancer has implications for infiltrating macrophages as a prognostic biomarker of progression in this disease setting. Cancer Res; 76(6); 1354–66. ©2016 AACR.

IL-17A or overexpression of ERα promoted estradiol-driven endometrial cancer cell proliferation

Induction of CD68+CD163+ macrophages from THP-1 cells

IL-17 receptor antibody compromises the effect of IL-17A on ERα transcription up-regulation

Serum estradiol levels were not elevated in patients with endometrioid adenocarcinoma after stratification by pathologic grading

Semi-quantitative evaluation of the immunohistochemical staining of IL-17A, TET1, 5-hmC, and ERα expression in endometrial hyperplasia and cancer lesions

The expression pattern of ERα in different cell lines

BackgroundProgestin resistance is the main obstacle for the conservative therapy to maintain fertility in women with endometrial hyperplasia and cancer. Up to 30% of endometrial cancers fail to respond to progestin, a rate that has not significantly changed due to the lack of a detailed understanding of progestin resistance.MethodsCell Counting Kit-8 (CCK-8) assays were used to detect the synergistic effects of brusatol in combination with progestin. Using commercial kits, the conversion of progestin to 20α-dihydroxyprogesterone following btusatol treatment or AKR1C1 silence was investigated. The correlation betwwen AKR1C1 expression profile and prgestin response was further analyzed in paired endometrial hyperplasia and cancer samples from the same individuals before and after progestin therapy. The effects of brusatol-mediated reversal of progestin resistance was explored in both mouse xenograft and human organoid models. DNA dot blot, HMeDIP, and dural-luciferase reporter assays were performed to uncover the mechanism through which brusatol inhibits AKR1C1 and sensitizes endometrial hyperplasia and cancer to progestin.ResultsBrusatol sensitizes endometrial cancer cell to progestin by downregulating the expression of Nrf2 and its target AKR1C1. Increased AKR1C1 facilitated production of 20-α-dihydroxyprogeserone and was associated with declined progesterone. Suppression of AKR1C1 by brusatol resulted in decreased progesterone catabolism and maintained potent progesterone to inhibit endometrial cancer growth. Aberrant overexpression of AKR1C1 was found in paired endometrial hyperplasia and cancer samples from the same individuals with progestin resistance, whereas attenuated or loss of AKR1C1 was observed in post-treatment samples with well progestin response as compared with paired pre-treatment tissues. Tet1 was identified as a novel Nrf2 target gene. It in turn upregulated AKR1C1 expression by increasing hydroxymethylation levels in its promoter regions. ConclusionsWe found that Nrf2-Tet1-AKR1C1 axis plays an essential role in progestin resistance, and brusatol sensitizes endometrial hyperplasia and cancer to progestin by suppressing Nrf2-Tet1-AKR1C1-mediated progestin metabolism. Our findings suggest that AKR1C1 expression pattern may serve as an important biomarker of progestin resistance in endometrial hyperplasia and cancer .

With a younger tendency in morbidity age, endometrial cancer (EC) incidence has grown year after year. Worse, even more commonly occurring is endometrial hyperplasia (EH), which is a precancerous endometrial proliferation. For young women with early EC and EH who want to preserve fertility, progestin therapy has been utilized as a routine fertility-preserving treatment approach. Nevertheless, progestin medication failure in some patients is mostly due to progestin resistance and side effects. In order to further analyze the potential mechanisms of progestin resistance in EH and EC, to provide theoretical support for effective therapeutic strategies, and to lay the groundwork for searching novel treatment approaches, this article reviews the current therapeutic effects of progestin in EH and EC, as well as the mechanisms and molecular biomarkers of progestin resistance, and systematically expounds on the potential therapeutic methods to overcome progestin resistance.

With the rapid development of mobile sensor, network infrastructure and cloud computing, the scale of AR application scenario is expanding from small or medium scale to large-scale environments. Localization in the large-scale environment is a critical demand for the AR applications. Most of the commonly used localization techniques require quite a number of data with groundtruth localization for algorithm benchmarking or model training. The existed groundtruth collection methods can only be used in the outdoors, or require quite expensive equipments or special deployments in the environment, thus are not scalable to large-scale environments or to massively produce a large amount of groundtruth data. In this work, we propose LSFB, a novel low-cost and scalable frame-work to build localization benchmark in large-scale environments with groundtruth poses. The key is to build an accurate HD map of the environment. For each visual-inertial sequence captured in it, the groundtruth poses are obtained by joint optimization taking both the HD map and visual-inertial constraints. The experiments demonstrate the obtained groundtruth poses are accurate enough for AR applications. We use the proposed method to collect a dataset of both mobile phones and AR glass exploring in large-scale environments, and will release the dataset as a new localization benchmark for AR.

PURPOSE: Patients with Klinefelter syndrome (KS) who receive assisted reproductive technology (ART) treatment often experience poor pregnancy rates due to decreased fertilization, cleavage, and implantation rates and even an increased miscarriage rate. Mounting evidence from recent studies has shown that various technological advances and approaches could facilitate the success of ART treatment for KS patients. In this review, we summarize the methods for guiding KS patients during ART and for developing optimal strategies for preserving fertility, improving pregnancy rate and live birth rate, and avoiding the birth of KS infants. METHODS: We searched PubMed and Google Scholar publications related to KS patients on topics of controlled ovarian stimulation protocols, sperm extraction, fertility preservation, gamete artificial activation, round spermatid injection (ROSI), and non-invasive prenatal screening (PGD) methods. RESULTS: This review outlines the different ovulation-inducing treatments for female partners according to the individual sperm status in the KS patient. We further summarize the methods of retrieving sperm, storing, and freezing rare sperm. We reviewed different methods of gamete artificial activation and discussed the feasibility of ROSI for sterile KS patients who absolutely lack sperm. The activation of eggs in the process of intracytoplasmic sperm injection and non-invasive PGD are urgently needed to prevent the birth of KS infants. CONCLUSION: The integrated strategies will pave the way for the establishment of ART treatment approaches and improve the clinical outcome for KS patients.

As a classical machine learning method, multi-task learning (MTL) has been widely applied in computer vision technology. Due to deep convolutional neural network (D-CNN) having strong ability of feature representation, the combination of MTL and D-CNN has attracted much attention from researchers recently. However, this kind of combination has rarely been explored in the field of vehicle analysis. The authors propose a D-CNN enhanced with weighted multi-attribute strategy for extensive exploration of comprehensive vehicle attributes over surveillance images. Specifically, regarding to recognising vehicle model and make/manufacturer, several related attributes as auxiliary tasks are incorporated in the training process of D-CNN structure. The proposed strategy focuses more on the main task compared with traditional MTL methods, which has assigned different weights for the main task and auxiliary tasks rather than treating all involved tasks equally. To the extent of their knowledge, this is the first report relating to the combination of D-CNN and weighted MTL for exploration of comprehensive vehicle attributes. The following experiments will show that the proposed approach outperforms the state-of-the-art method for the vehicle recognition and improves the accuracy rate by about 10% for the analysis of other vehicle attributes on the recently public CompCars dataset.

A visual simultaneous localization and mapping (SLAM) system usually contains a relocalization module to recover the camera pose after tracking failure. The core of this module is to establish correspondences between map points and key points in the image, which is typically achieved by local image feature matching. Since recently emerged binary features have orders of magnitudes higher extraction speed than traditional features such as scale invariant feature transform, they can be applied to develop a real-time relocalization module once an efficient method of binary feature matching is provided. In this paper, we propose such a method by indexing binary features with hashing. Being different from the popular locality sensitive hashing, the proposed method constructs the hash keys by an online learning process instead of pure randomness. Specifically, the hash keys are trained with the aim of attaining uniform hash buckets and high collision rates of matched feature pairs, which makes the method more efficient on approximate nearest neighbor search. By distributing the online learning into the simultaneous localization and mapping process, we successfully apply the method to SLAM relocalization. Experiments show that camera poses can be recovered in real time even when there are tens of thousands of landmarks in the map.

Golgi phosphoprotein 3 (GOLPH3), a newly recognized oncogene, is associated with tumor growth, metastasis, and poor prognosis in several types of cancer. However, its biological role and underlying mechanism in epithelial ovarian cancer (EOC) remain poorly understood. Here, we found that GOLPH3 was overexpressed in EOC tissues and cell lines. This overexpression promoted the migration and invasion of EOC cells. Moreover, GOLPH3 upregulated the expression of epithelial–mesenchymal transition (EMT) markers, such as N‐cadherin and Snail, and the Wnt/β‐catenin‐related genes cyclin‐D1 and c‐Myc, which were restored via silencing of GOLPH3 expression. Furthermore, the inhibitor and activator of the Wnt/β‐catenin pathway, XAV939 and LiCl, enhanced or decreased, respectively, the effect of GOLPH3 on EMT, which further confirmed that GOLPH3 promoted EMT progression via activation of Wnt/β‐catenin signaling. In addition, we found that EDD, the human hyperplastic discs gene, was consistent with GOLPH3 expression and also promoted the EMT process and activated Wnt/β‐catenin signaling. These findings demonstrate that EDD might be a downstream factor of GOLPH3. Taken together, our findings demonstrate the existence of a GOLPH3–Wnt/β‐catenin–EMT axis in EOC and provide a new therapeutic target to treat EOC.

Large amount of clinical evidence has demonstrated that insulin resistance is closely related to oncogenesis of endometrial cancer (EC). Despite recent studies showed the up-regulatory role of insulin in G protein-coupled estrogen receptor (GPER/GPR30) expression, GPER expression was not decreased compared to control when insulin receptor was blocked even in insulin treatment. The purpose of this study was to explore the possible mechanism by which insulin up-regulates GPER that drives EC cell proliferation. For this purpose, we first investigated the GPER expression in tissues of endometrial lesions, further explored the effect of GPER on EC cell proliferation in insulin resistance context. Then we analyzed the role of Ten-Eleven Translocation 1 (TET1) in insulin-induced GEPR expression and EC cell proliferation. The results showed that GPER was highly expressed in endometrial atypical hyperplasia and EC tissues. Mechanistically, insulin up-regulated TET1 expression and the latter played an important role in up-regulating GPER expression and activating PI3K/AKT signaling pathway. TET1 mediated GPER up-regulation was another mechanism that insulin promotes EC cell proliferation.

Background: Insulin resistance (IR) has been well studied in the initiation and development of endometrial endometrioid carcinoma (EEC). As yet, it has been largely neglected for estrogen sensitivity in local endometrium in hyperinsulinemia-induced systemic microenvironment. The aim of this study was to investigate the role of insulin in regulating estrogen sensitivity and explore the potential mechanisms in insulin-driven inflammatory microenvironment. Methods: We first investigated the effect of insulin on estradiol-driven endometrial cancer cells proliferation in vitro to address the roles of insulin in modulating estrogen sensitivity. Then GPER, ERα and TET1 in EEC samples with or without insulin resistance were screened by immunohistochemistry to confirm whether insulin resistance regulates estrogen receptors. Further mechanism analysis was carried out to address whether TET1 was mediated epigenetic modulation of GPER in insulin-induced microenvironment. Results: Insulin enhanced estradiol-driven endometrial cancer cells proliferation by up-regulating G-protein-coupled estrogen receptor (GPER) expression, but not ERα or ERβ. Immunohistochemistry of EEC tissues showed that GPER expression was greatly increased in endometrial tissues from EEC subjects with insulin resistance and was positively correlated with Ten-eleven-translocation 1 (TET1) expression. Mechanistically, insulin up-regulates TET1 expression, and the latter, an important DNA hydroxymethylase, could up-regulate GPER expression through epigenetic modulation. Conclusion: This study identified TET1 as the upstream regulator of GPER expression and provides a possible mechanism that insulin-induced positive regulation of estrogen sensitivity in endometrial cancer cells. Increasing expression of GPER through TET1-mediated epigenetic modulation may emerge as the main regulator to enhance the response of endometrial cancer to estrogen in insulin-driven inflammatory microenvironment.

Progestins has been used widely for endometrial cancer (EC) patients. However, long term use of high dose progestin often lead to progestin resistance. Our previous studies have demonstrated that metformin reversed progestin resistance through the downregulation of the expression of glyoxalase I (GLOI) in type I endometrial cancer. Recent studies have demonstrated the role of Ten-eleven translocation 1 (TET1) in endometrial cancer, but the physiological role of TET1 in GLOI-mediated progestin resistance has been poorly addressed. Immunohistochemistry was used to detect the expression of TET1, GLOI and 5hmC in various endometrium. Western blot was carried out to analyses TET1 and GLOI expression with different treatment. Cell counting kit-8 was used to evaluate cell proliferation after various treatment. Dot blot assay and HMeDIP assay were performed to detect global hydroxmethylation levels and hydroxymethylation levels in GLOI gene respectively. In current study, we found that metformin effectively sensitized progestin in endometrial cancer cell lines through the down regulation of the expression of TET1 and GLOI. Interestingly, the exogenous increase of TET1 expression enhanced total 5hmC level and hydroxymethylation modification in glyoxalase I promoter region. This effect was abated by metformin treatment. Moreover, the expression profile of TET1 and glyoxalase I in various endometrial tissue parallelized with 5hmC level. Therefore, this finding suggests that metformin sensitized progestin in endometrial cancer through the TET1-5hmC-GLOI signaling pathway.

The aim of present study was to investigate the role of preoperative neutrophil to lymphocyte ratio (NLR) and platelet to lymphocyte ratio (PLR) used as prognostic markers for predicting chemotherapeutic response and survival outcomes in patients with epithelial ovarian cancer (EOC) who are receiving platinum-based chemotherapy. A total of 344 patients diagnosed with EOC who are receiving platinum-based chemotherapy from 2005 to 2010 in the hospital were enrolled. NLR and PLR were calculated from complete blood cell count taken before operation. The patients were divided into platinum-resistant (P-R) group and platinum-sensitive (P-S) group according to chemotherapeutic response. Clinicopathologic variables and outcomes were retrospectively collected and compared among groups. We used receiver operating characteristic (ROC) curves to calculate optimal cut-off values for NLR and PLR to predict chemotherapeutic response and prognosis. The AUC, sensitivity, specificity of NLR > 3.02 to predict platinum resistance were 0.819, 75.0% and 81.45%, respectively. The corresponding values of PLR > 207 were 0.727, 60.42% and 85.48%, respectively. Patients with lower value of NLR (NLR < 3.02) or PLR (PLR < 207) had a longer progression-free survival (PFS) and overall survival (OS). In multivariate analysis, NLR and PLR showed a significant association with PFS (hazard ratio [HR], 1.733; 95%CI, 1.225-2.453, P = 0.002 and HR, 1.952; 95%CI, 1.430-2.662, P < 0.001) and OS (HR, 1.616; 95%CI, 1.138-2.297, P = 0.007, and HR, 2.167; 95%CI, 1.565-3.000, P < 0.001). These results suggest that the assessment of NLR and PLR could assist the identification of patients with poor prognosis and had potential clinical value in predicting platinum resistance in patients with EOC.

The aim of this study was to investigate the relationship between HIF-1α and SNAIL gene expression in the epithelial ovarian cancer (EOC) cell line. EOC cells were treated with hypoxia, hypoxia combined with rapamycin, and control. The expression of HIF-1α and E-cad were assessed by reverse transcription–polymerase chain reaction (RT-PCR) and western blotting. The gene expression of SNAIL was studied by RT-PCR and real-time PCR. RNA interference technology was used to determine the relationship between HIF-1α and SNAIL. The present study indicated that the HIF-1α protein was expressed and increased in EOC cell line. SNAIL mRNA was found to increase and E-cad expression decreased with the time of hypoxia prolonged. Hypoxia increased invasion abilities of EOC cell line, but compared with cells exposed to hypoxia, the change of invasive ability of cells with rapamycin had no effect. The expression of HIF-1α protein and SNAIL mRNA could be inhibited gradually by rapamycin. siRNA of HIF-1α could suppress the expression of SNAIL while siRNA of SNAIL had no influence on HIF-1α protein expression. HIF-1α may be the upstream of the SNAIL gene in EOC. Our data suggested that HIF-1α might be an upregulator of the SNAIL gene and HIF-1α-SNAIL-E-cad pathway may play an important role in EOC invasion and metastasis.

FKBP14 belongs to the family of FK506-binding proteins (FKBPs). Altered expression of FKBPs has been reported in several malignancies. This study aimed to reveal the expression profile of FKBP14 in ovarian cancer and evaluate whether FKBP14 is a molecular target for cancer therapy. We found that the FKBP14 mRNA level was significantly higher in ovarian cancer tissues than in normal tissues. FKBP14 expression was then knocked down in two ovarian cancer cell lines, SKOV3 and HO8910 cells, by a lentiviral short hairpin RNA (shRNA) delivery system. Reduced expression of FKBP14 markedly impaired the proliferative ability of ovarian cancer cells. Additionally, ovarian cancer cells infected with FKBP14 shRNA lentivirus tended to arrest in the G0/G1 phase and undergo apoptosis. Moreover, knockdown of FKBP14 induced cell apoptosis via increasing the ratio of Bax to Bcl-2. These results indicated that FKBP14 might be a diagnostic marker for ovarian cancer and could be a potential molecular target for the therapy of ovarian cancer.

The multi-center, randomized, open-label, phase III trial discussed in the present study was performed to compare the clinical outcomes of nedaplatin (NDP) plus paclitaxel, and carboplatin (CBP) plus paclitaxel for the treatment of epithelial ovarian cancer (EOC). In the current study, 182 patients with International Federation of Gynecology and Obstetrics (FIGO) stage II-IV EOC were randomly assigned to receive NDP plus paclitaxel or CBP plus paclitaxel at 3-week intervals for a total of six courses. The primary endpoints were progression-free survival rate (PFS) and overall survival rate (OS). The secondary endpoints were toxicity profiles. The median follow-up was 44.63 months [95% confidence interval (CI) 33.67-46.47 months] for the NDP group and 47.63 months (95% CI 45.13-49.07 months) for the CBP group. Overall, there was no significant difference in PFS or OS between the two groups (P=0.09 for PFS, and P=0.65 for OS). For the patients with FIGO stage III-IV EOC, the NDP plus paclitaxel regimen significantly prolonged PFS (P=0.02) but did not result in improved OS (P=0.53) when compared with the CBP group. The patients in the NDP plus paclitaxel group also exhibited a lower incidence rate of grade 3 or 4 leucopenia (P=0.03). Other hematological and non-hematological toxicity profiles were similar between the two groups. Compared with CBP plus paclitaxel regimens, NDP plus paclitaxel regimens achieved comparable survival outcomes and similar toxicity profiles. However, patients of FIGO stage III-IV EOC may experience more clinical benefits from NDP plus paclitaxel treatment, including a prolonged PFS and a lower incidence rate of leucopenia. Therefore, an NDP-based regimen may be an alternative choice when using platinum-based agents to treat EOC.

Our previous study showed that Octamer-binding transcription factor 4 (OCT4) expression was upregulated and significantly associated with histological grade through the analysis of OCT4 expression in 159 ovarian cancer tissue samples, and OCT4 mediated follicle-stimulating hormone (FSH)-induced anti-apoptosis in epithelial ovarian cancer. Nevertheless, whether OCT4 participates in FSH-induced invasion in ovarian cancer is still unknown. Therefore, the present study aimed to define whether FSH-induced ovarian cancer invasion is mediated by OCT4. In present study, we showed that FSH induced not only the epithelial-mesenchymal transition (EMT) and invasive phenotype but also the upregulation of OCT4 expression in a dose- and time-dependent manner in epithelial ovarian cancer cells. In addition, the expression of FSH receptor (FSHR) was upregulated by FSH induction, and knockdown of FSHR inhibited FSH-stimulated OCT4 expression. ERK1/2 signaling pathway participated in the enhanced expression of OCT4 and Snail induced by FSH. We further showed that the activated expression of Snail and N-cadherin, the suppressed expression of E-cadherin and the morphological change of the cells stimulated by FSH were blocked by OCT4-specific small interfering RNA. Moreover, our results showed that OCT4 mediated the increase in invasive capacity induced by FSH in ovarian cancer cells. Taken together, our work reveals that OCT4 is an essential mediator in FSH-induced EMT and invasion in epithelial ovarian cancer and may act as a potential therapeutic target.

Semi-supervised visual domain adaptation is devoted to adapting a model learned in source domain to target domain where there are only a few labeled samples. In this paper, we propose a semi-supervised cross-domain image recognition method which unifies the feature learning and recognition model training into a convolutional neural network framework. Based on a few labeled samples and massive unlabeled samples in the source and target domains, we specially design three branches for class label, domain label and similarity label prediction which simultaneously optimizes the network to generate image features that are domain invariance and inter-class discriminative. Experimental results demonstrate that our method is effective for learning robust cross-domain image recognition model, and achieves the state-of-the-art performance on the widely used visual domain adaptation benchmark.

Supervised deep learning models like convolutional neural network (CNN) have shown very promising results for the face recognition problem, which often require a huge number of labeled face images. Since manually labeling a large training set is a very difficult and time-consuming task, it is very beneficial if the deep model can be trained from face samples with only weak annotations. In this paper, we propose a general framework to train a deep CNN model with weakly labeled facial images that are available on the Internet. Specifically, we first design a deep Multiscale-Inception CNN (MICNN) architecture to exploit the multi-scale information for face recognition. Then, we train an initial MICNN model with only a limited number of labeled samples. After that, we propose a dual-level sample selection strategy to further fine-tune the MICNN model with the weakly labeled samples from both the sample level and class level, which aims to skip outliers and select more samples from confusing class pairs during training. Extensive experimental results on the LFW and YTF benchmarks demonstrate the effectiveness of the proposed method.

Chemoresistance is the major obstacle to cure endometrial cancer, whereas metformin has demonstrated sensitization to chemotherapy in endometrial cancer. A novel finding states that isocitrate dehydrogenase 1 (IDH1) involves in cancer chemoresistance. Recent studies have revealed that epigenetic modifications facilitate chemoresistance. However, whether IDH1 play a role in metformin-induced endometrial cancer chemosensitivity through epigenetic modification is incompletely understood. Immunohistochemistry and Elisa assays were used to evaluate the expression pattern of IDH1 in endometrial tissue and serum, respectively. Western blot was performed to determine changes in expression of key molecules in the IDH1-ɑ-KG-TET1-Nrf2 signaling pathway after various treatments. Dot blot assays were used to assess global hydroxymethylation levels after metformin administration or plasmid transfection. Antioxidant response element (ARE) activity in the IDH1 promoter region was monitored by luciferase assay. Cancer cell sensitivity to chemotherapy was detected by SRB assay. We found that activation of the IDH1 signaling pathway in endometrial cancer tissue resulting from aberrant expression of IDH1 and its downstream mediators conferred chemoresistance. We found that this effect was abated by metformin treatment. Dot blot and HMeDIP assays revealed that metformin blocked IDH1-ɑ-KG-TET1-mediated enhancement of Nrf2 hydroxymethylation levels, eliminating chemoresistance. Moreover, we observed that chemoresistance was enhanced via a regulatory loop in which Nrf2 activated IDH1-ɑ-KG-TET1-Nrf2 signaling via binding to the ARE sites in the IDH1 promoter region. Our findings highlight a critical role of IDH1-ɑ-KG-TET1-Nrf2 signaling in chemoresistance and suggest that rational combination therapy with metformin and chemotherapeutics has the potential to suppress chemoresistance.

This paper introduces our submission to the 2nd Facial Landmark Localisation Competition. We present a deep architecture to directly detect facial landmarks without using face detection as an initialization. The architecture consists of two stages, a Basic Landmark Prediction Stage and a Whole Landmark Regression Stage. At the former stage, given an input image, the basic landmarks of all faces are detected by a sub-network of landmark heatmap and affinity field prediction. At the latter stage, the coarse canonical face and the pose can be generated by a Pose Splitting Layer based on the visible basic landmarks. According to its pose, each canonical state is distributed to the corresponding branch of the shape regression sub-networks for the whole landmark detection. Experimental results show that our method obtains promising results on the 300-W dataset, and achieves superior performances over the baselines of the semi-frontal and the profile categories in this competition.

Pyruvate kinase (PK) is a key enzyme in the process of glycolysis, catalyzing phosphoenolpyruvate (PEP) into pyruvate. Currently, PK isozyme type M2 (PKM2), one subtype of PK, has been proposed as a new tumor marker with high expression in various tumor tissues. Here we aimed to explore the effects of siRNA-PKM2 on ovarian carcinoma (OC) cell lines SKOV3 and OVCAR3, in which PKM2 was notably expressed. PKM2 gene interference lentivirus vectors were built by miRNA transfection assay. siRNA-PKM2-transfected SKOV3 and OVCAR3 cells were evaluated for cell proliferation, cell cycle distribution, cell apoptosis, cell migration, and invasion in this study. In addition, the expression levels of several tumor-related genes were measured using real-time PCR and Western blot. Results showed that siRNA-PKM2 markedly inhibited cell proliferation, induced apoptosis, and caused cell cycle arrest at the G0/G1 phase. Cell migration and invasion were significantly suppressed by siRNA-PKM2. Furthermore, the tumor-related genes caspase 7, Bad, and E-cadherin were upregulated, while MMP2, HIF1α, VEGF, and MMP9 were depressed by siRNA-PKM2. The function of siRNA-PKM2 on the biological behavior of OC cells indicated that PKM2 may also be a target for treatment of OC.

Purpose It has previously been shown that follicle-stimulating hormone (FSH) and its receptor contribute to epithelial ovarian cancer (EOC) development. Epithelial-mesenchymal transition (EMT) is the early event of metastasis in cancer. Therefore, the aim of this study was to investigate the roles of FSH and the FSH receptor (FSHR) in EMT of EOC. Methods Ovarian cancer cells treated with various doses of FSH were used to investigate the effect of FSH on EMT. Small interfering RNA–mediated FSHR depletion or reexpression of FSHR by acute transfecting pcDNA-hFSHR plasmid was performed to determine the role of FSHR in FSH-induced EMT. Moreover, LY294002, a potent and specific cell-permeable inhibitor of phosphatidylinositol 3-kinases (PI3K), was selected to pretreat ovarian cancer cells to confirm whether PI3K/Akt signaling is involved in this event. Results In the current study, FSH was found to induce the phenotypes of EMT including migration and invasion in EOC cells. Elevated FSHR levels promoted EMT, migration, and invasion, whereas small interfering RNA–mediated FSHR knockdown inhibited these processes. Moreover, the inhibition of FSH-induced PI3K/Akt signaling pathway attenuated Snail expression and the EMT process. Conclusions Collectively, the findings of the current study indicate that FSH induced the EMT of ovarian cancer cells through the FSHR-PI3K/Akt-Snail signaling pathway.