NJGCG: A node-based joint Gaussian copula graphical model for gene networks inference across multiple states

<div>Abstract<p>Different kinases are expressed in different clinical subsets of breast cancer. In this study, we assessed kinase expression patterns in different clinical subtypes of breast cancer, evaluated the prognostic and predictive values of kinase metagenes, and investigated their functions <i>in vitro</i>. Four hundred twenty-eight protein kinases in gene expression data were examined from 684 cases of breast cancer and 51 breast cancer cell lines to identify kinase expression patterns. We tested the prognostic value of kinase metagenes in 684 node-negative patients who received no adjuvant therapy and the predictive value in 233 patients who received uniform neoadjuvant chemotherapy. Twelve kinases were overexpressed in estrogen receptor (ER)–positive/human epidermal growth factor receptor 2 (HER2)–negative, 7 in HER2<sup>+</sup>, and 28 in ER<sup>−</sup>/HER2<sup>−</sup> cancers, respectively. We examined the functional role of 22 kinases overexpressed in ER<sup>−</sup>/HER2<sup>−</sup> cancers using siRNA. Downregulation of these kinases caused significant subtype-specific inhibition of cell growth <i>in vitro</i>. Two robust kinase clusters, including an immune kinase cluster and a mitosis kinase cluster, were present in all clinical subgroups. High mitosis kinase score was associated with worse prognosis but higher pathologic complete response (pCR) in ER<sup>+</sup>/HER2<sup>−</sup> cancers, but not in ER<sup>−</sup>/HER2<sup>−</sup> or HER2<sup>+</sup> cancers, in univariate and multivariate analyses including other genomic predictors (MammaPrint, genomic grade index, and the 76-gene signature). Conversely, higher immune kinase score was associated with better survival in ER<sup>+</sup>/HER2<sup>−</sup> and HER2<sup>+</sup> tumors and also predicted higher probability of pCR in HER2<sup>+</sup> cancers. Taken together, our results indicate that kinases regulating mitosis and immune functions convey distinct prognostic information that varies by clinical subtype. Cancer Res; 70(21); 8852–62. ©2010 AACR.</p></div>

The histone lysine demethylase KDM4 subfamily, comprised of four members (A, B, C, and D), play critical roles in controlling transcription, chromatin architecture and cellular differentiation. Although KDM4 family members have a high degree of homology, they may play different roles in various types of breast cancer. The goal of this study is to analyze the genomic and transcriptomic alterations of the KDM4 subfamily in different subtypes of breast cancer, and explore the therapeutic potential of a novel KDM4 inhibitor in aggressive basal breast cancer. We conducted a large-scale meta-analysis of KDM4A, B, C, and D in breast cancer and identified associations among recurrent copy number alterations, gene expression and breast cancer subtypes. We examined KDM4 expression in a panel of non-tumorigenic and cancerous breast epithelial cell lines using quantitative RT-PCR and Western blot assays, and assessed global histone 3 methylation levels by Western blot in a panel of breast cancer cell lines. Next, we tested a novel inhibitor of the KDM4 demethylases, small molecule NCDM-32B, for its ability to attenuate basal breast cancer cell growth and impair metastatic potential. Finally, to examine gene expression changes induced by NCDM-32B, we conducted expression profiling and pathway analyses. We found that the KDM4 members show different expression patterns in subtypes of breast cancer. Our findings suggest that H3 global methylation levels vary among different breast cancer cell lines. We demonstrated that NCDM-32B significantly impaired viability, cellular growth and transforming phenotypes of basal breast cancer in vitro. Furthermore, NCDM-32B impaired several critical pathways and classical oncogenes, including MET, CDC26, and CDK6, that drive cellular proliferation and transformation in breast cancer. In summary, our findings add layers of information to the genomic and transcriptomic profiles of the KDM4 subfamily in different subtypes of breast cancer. We provide the first evidence that a novel KDM4 demethylase inhibitor, the small molecule NCDM-32B, led to significant inhibition of cellular growth of basal breast cancer in vitro. These findings lay the foundation for evaluating KDM4 inhibitors as therapeutic approaches against aggressive breast cancer. Citation Format: Andreana N. Holowatyj, Qin Ye, Jack Wu, Hui Liu, Lihong Zhang, Takayoshi Suzuki, Zeng-Quan Yang. Genetic alterations of KDM4 subfamily and therapeutic effect of novel demethylase inhibitor in breast cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 98. doi:10.1158/1538-7445.AM2015-98

A nontoxic chemopreventive intervention efficacious against different subtypes of breast cancer is still a clinically unmet need. The present study was undertaken to determine the efficacy of an Ayurvedic medicine phytochemical (Withaferin A, [WA]) for chemoprevention of breast cancer and to elucidate its mode of action. Chemopreventive efficacy of WA (4 and 8 mg/kg body weight) was determined using a rat model of breast cancer induced by N-methyl-N-nitrosourea (MNU; n = 14 for control group, n = 15 for 4 mg/kg group, and n = 18 for 8 mg/kg group). The mechanisms underlying breast cancer chemoprevention by WA were elucidated by immunoblotting, biochemical assays, immunohistochemistry, and cytokine profiling using plasma and tumors from the MNU-rat (n = 8-12 for control group, n = 7-11 for 4 mg/kg group, and n = 8-12 for 8 mg/kg group) and/or mouse mammary tumor virus-neu (MMTV-neu) models (n = 4-11 for control group and n = 4-21 for 4 mg/kg group). Inhibitory effect of WA on exit from mitosis and leptin-induced oncogenic signaling was determined using MCF-7 and/or MDA-MB-231 cells. All statistical tests were two-sided. Incidence, multiplicity, and burden of breast cancer in rats were decreased by WA administration. For example, the tumor weight in the 8 mg/kg group was lower by about 68% compared with controls (8 mg/kg vs control, mean = 2.76 vs 8.59, difference = -5.83, 95% confidence interval of difference = -9.89 to -1.76, P = .004). Mitotic arrest and apoptosis induction were some common determinants of breast cancer chemoprevention by WA in the MNU-rat and MMTV-neu models. Cytokine profiling showed suppression of plasma leptin levels by WA in rats. WA inhibited leptin-induced oncogenic signaling in cultured breast cancer cells. WA is a promising chemopreventative phytochemical with the ability to inhibit at least two different subtypes of breast cancer.

Supplementary Figure Legends 1-5, Table Legends 1-11 from Prognostic and Therapeutic Implications of Distinct Kinase Expression Patterns in Different Subtypes of Breast Cancer

Supplementary Tables 1-11 from Prognostic and Therapeutic Implications of Distinct Kinase Expression Patterns in Different Subtypes of Breast Cancer

Joint learning of multiple gene networks from single-cell gene expression data

With a limiting dilution technique, clones have been established from the human pancreatic adenocarcinoma cell line, HPAF. Phenotypic analysis by a panel of murine monoclonal antibodies demonstrated distinct profiles of antigenic expression between the clones. However, identical isozyme patterns of different clones indicated their common origin from the parental HPAF cells. Two clones, CD11 and CD18, appeared to be arrested at different stages of secretory epithelial cell differentiation. CD11 cells demonstrated many characteristics of a well-differentiated state, including the formation of ductal structures with polarized, long columnar-shaped cells, the presence of secretory granules in the cytoplasm, high DU-PAN-2 antigen expression in nude mouse xenografts, and a longer doubling time (42 h) in tissue culture. In contrast, CD18 cells exhibited characteristics of a poorly differentiated state, including solid nests of isoprismatic cells without luminal spaces and cellular polarization, absence of secretory granules and DU-PAN-2 antigen expression in xenografts, and a shorter doubling time (26 h) in tissue culture. Since no culture systems of normal pancreatic ductal cells are currently available, these two pancreatic adenocarcinoma clones may provide a unique system to study genes and antigens related to pancreatic ductal cell differentiation.

To explore the prognostic value of applying a semi-automated detection system for detecting circulating tumor cells (CTC) for different stages and various subtypes of breast cancer. Immunomagnetic separation and immunofluorescent staining were employed to detect the expressions of CTC and CTC HER-2. Chi-square test, univariate and multivariate analyses were used to examine the correlations between CTC and clinical characteristics. CTC was detected in 57.5% (138/240) of metastatic breast cancer (MBC). CTC enumeration, HER-2 status, number of metastasis and bone metastasis were relevant (P < 0.05). CTC was detected in 23.1% (6/26) of early-stage breast cancer (EBC). CTC enumeration, pathological type, estrogen receptor (ER), progesterone receptor (PR), Ki-67, tumor size and lymph node metastasis were irrelevant (P > 0.05). HER-2-positive MBC detected less CTC than HER-2-negative (χ(2) = 12.296, P < 0.001) . CTC HER-2 expression was different between HER-2-positive and negative patients in 99 MBC cases (χ(2) = 8.082, P = 0.004). CTC enumeration is different for various stages and different subtypes of breast cancer. CTC enumeration and bone metastasis are relevant. CTC detection in EBC may predict tumor relapse. And CTC HER-2 expression is different between HER-2-positive and negative MBC patients.

Central nervous system (CNS) cancers are one of the leading cancer types in the world. Analysing gene hubs involved in CNS cancer is important for predicting therapies. We have used gene expression data from one of the common brain cancer type (Pheochromocytoma /paraganglioma (PCPG)) for hub gene identification. We have used density-based (MCODE and ClusterOne) algorithms and cytohubba tool to identify hub genes and, R and Cytoscape for data analysis. According to results, RPS4X and RPL37A are common hub genes in all networks. There are three hub gene networks (identified using MCODE and ClusterOne hub gene networks which can be used as a classification feature in future studies. HNRNPA2B1 gene is connecting major PPI complexes in PCPG cancers. Ribosomal protein S4 genes as hub genes play a significant role in PCPG protein interaction networks.

Background:Short sleep has been hypothesised to increase the risk of breast cancer. However, little is known about the association between sleep and different subtypes of breast cancer defined by hormone receptor status.Methods:Among 40 013 women in the Breast Cancer Detection Demonstration Project, including 1846 incident breast cancer cases, we prospectively examined self-reported weekday and weekend sleep duration in relation to breast cancer risk. We used multivariate Cox proportional hazards regression models to estimate relative risks (RRs) and 95% confidence intervals (CIs).Results:We found no association between sleep and overall breast cancer. However, we observed a decreased risk of ER+PR+ breast cancer (RR <6 vs 8 – 9 h (95% CI): 0.54 (0.31, 0.93), P for trend, 0.003) with shorter sleep duration.Conclusions:Our finding does not support an association between sleep duration and overall breast cancer risk. However, the effect of sleep on different subtypes of breast cancer deserves further investigation.

Joint reconstruction of multiple gene regulatory networks (GRNs) using gene expression data from multiple tissues/conditions is very important for understanding common and tissue/condition-specific regulation. However, there are currently no computational models and methods available for directly constructing such multiple GRNs that not only share some common hub genes but also possess tissue/condition-specific regulatory edges. In this paper, we proposed a new graphic Gaussian model for joint reconstruction of multiple gene regulatory networks (JRmGRN), which highlighted hub genes, using gene expression data from several tissues/conditions. Under the framework of Gaussian graphical model, JRmGRN method constructs the GRNs through maximizing a penalized log likelihood function. We formulated it as a convex optimization problem, and then solved it with an alternating direction method of multipliers (ADMM) algorithm. The performance of JRmGRN was first evaluated with synthetic data and the results showed that JRmGRN outperformed several other methods for reconstruction of GRNs. We also applied our method to real Arabidopsis thaliana RNA-seq data from two light regime conditions in comparison with other methods, and both common hub genes and some conditions-specific hub genes were identified with higher accuracy and precision. JRmGRN is available as a R program from: https://github.com/wenpingd. Supplementary data are available at Bioinformatics online.

Breast cancer is the leading cause of female cancer-related deaths worldwide. New technologies with enhanced sensitivity and specificity for early diagnosis and monitoring of postoperative recurrence are in critical demand. Automatic breast full volume scanning system (ABVS) is an emerging technology used as an alternative imaging method for breast cancer screening. Despite its improved detection rate of malignant tumors, ABVS cannot accurately stage breast cancer preoperatively in 30–40% of cases. As a major hallmark of breast cancer, the characteristic metabolic reprogramming may provide potential biomarkers as an auxiliary method for ABVS.ObjectiveThe objective of this study was to identify differential metabolomic signatures between benign and malignant breast tumors and among different subtypes of breast cancer patients based on untargeted metabolomics and improve breast cancer detection rate by combining key metabolites and ABVS.MethodsUntargeted metabolomics approach was used to profile serum samples from 70 patients with different subtypes of breast cancer and benign breast tumor to determine specific metabolomic profiles through univariate and multivariate statistical data analysis.ResultsMetabolic profiles correctly distinguished benign and malignant breast tumors patients, and a total of 791 metabolites were identified. There were 54 different metabolites between benign and malignant breast tumors and 17 different metabolites between invasive and non-invasive breast cancer. Notably, the missed diagnosis rate of ABVS could be reduced by differential metabolite analysis. Moreover, the diagnostic performance analyses of combined metabolites (pelargonic acid, N-acetylasparagine, and cysteine-S-sulfate) with ABVS performance gave a ROC area under the curve of 0.967 (95% CI: 0.926, 0.993).ConclusionsOur study identified metabolic features both in benign and malignant breast tumors and in invasive and non-invasive breast cancer. Combined ultrasound ABVS and a panel of differential serum metabolites could further improve the accuracy of preoperative diagnosis of breast cancer and guide surgical therapy.

Non-negative matrix factorization and differential expression analyses identify hub genes linked to progression and prognosis of glioblastoma multiforme

The aim of this analysis is to evaluate the relative weight of different epidemiological risk factors on the development of different breast cancer subtypes (i.e. luminal, Her2+ overexpressed or triple negative). De-identified datasets of female participants recruited within the Prostate, Lung, Colorectal, and Ovarian (PLCO) trial were accessed. Multivariate Cox regression analysis was utilized to assess factors affecting the development of breast cancer (regardless of subtype). Additional multivariate analyses were conducted to assess factors affecting the development of the three principal subtypes of breast cancer (ER+/Her2- breast cancer; Her2 overexpressed breast cancer and ER-/Her2- breast cancer). A total of 73,570 eligible participants were evaluated in the current analysis of which 2370 participants subsequently developed breast cancer. The following factors were associated with a higher risk of ER+/Her2- breast cancer: white race (P < 0.001), nulliparity (P < 0.001), higher body mass index (P = 0.003), prior exposure to hormone treatment (P = 0.004) and breast cancer in first-degree female relatives (P < 0.001). The following factors were associated with a higher risk of Her2 overexpressed breast cancer: prior exposure to hormone treatment (P = 0.002) and breast cancer in first-degree female relatives (P = 0.001). The following factors were associated with a higher risk of ER-/Her2- breast cancer: black race (P = 0.013), younger age (P = 0.017) and breast cancer in first-degree female relatives (P 0.023). There is considerable heterogeneity in risk factors among patients with different subtypes of breast cancer. In particular, factors associated with high estrogen levels seem to be associated with luminal breast cancer rather than other breast cancer subtypes.

Breast cancer has one of the highest incidences in females worldwide. Girdin is a novel actin‑binding protein, that induces cell migration and angiogenesis. However, a systematic study of Girdin function in distinct subtypes of breast cancer has not been reported to date. Therefore, the present study aimed to investigate the role of Girdin on cell proliferation, migration and angiogenesis in different subtypes of breast cancer. For this purpose, the breast epithelial MCF‑7, breast ductal T47D and breast metastatic MDA‑MB‑231 cancer cell lines were selected. Girdin small interfering RNA (siRNA) was transfected into MCF‑7, T47D and MDA‑MB‑231 cells. Girdin knockdown suppressed cell viability and migration in the different cancer cells tested. Girdin knockdown also suppressed mRNA expression of vascular endothelial growth factor (VEGF), and activation of phosphatidyl inositol 3‑kinase (PI3K) and RAC‑α serine/threonine‑protein kinase (Akt) in the subtypes tested. In conclusion, these data indicate that Girdin knockdown suppressed cell viability and migration and may suppress angiogenesis via the PI3K/Akt signaling pathway, in various breast cancers subtypes. The present study therefore suggests a role for Girdin as a novel therapeutic target for breast cancer, independent of subtype.