Hans A. Kestler

MYC stimulates EZH2 expression by repression of its negative regulator miR-26a

The MYC oncogene, which is commonly mutated/amplified in tumors, represents an important regulator of cell growth because of its ability to induce both proliferation and apoptosis. Recent evidence links MYC to altered miRNA expression, thereby suggesting that MYC-regulated miRNAs might contribute to tumorigenesis. To further analyze the impact of MYC-regulated miRNAs, we investigated a murine lymphoma model harboring the MYC transgene in a Tet-off system to control its expression. Microarray-based miRNA expression profiling revealed both known and novel MYC targets. Among the miRNAs repressed by MYC, we identified the potential tumor suppressor miR-26a, which possessed the ability to attenuate proliferation in MYC-dependent cells. Interestingly, miR-26a was also found to be deregulated in primary human Burkitt lymphoma samples, thereby probably being of clinical relevance. Although today only few miRNA targets have been identified in human disease, we could show that ectopic expression of miR-26a influenced cell cycle progression by targeting the bona fide oncogene EZH2, a Polycomb protein and global regulator of gene expression yet unknown to be regulated by miRNAs. Thus, in addition to directly targeting protein-coding genes, MYC modulates genes important to oncogenesis via deregulation of miRNAs, thereby vitally contributing to MYC-induced lymphomagenesis.

Three learning phases for radial-basis-function networks

In this paper, learning algorithms for radial basis function (RBF) networks are discussed. Whereas multilayer perceptrons (MLP) are typically trained with backpropagation algorithms, starting the training procedure with a random initialization of the MLPs parameters, an RBF network may be trained in many different ways. We categorize these RBF training methods into one-, two-, and three-phase learning schemes. Two-phase RBF learning is a very common learning scheme. The two layers of an RBF network are learnt separately; first the RBF layer is trained, including the adaptation of centers and scaling parameters, and then the weights of the output layer are adapted. RBF centers may be trained by clustering, vector quantization and classification tree algorithms, and the output layer by supervised learning (through gradient descent or pseudo inverse solution). Results from numerical experiments of RBF classifiers trained by two-phase learning are presented in three completely different pattern recognition applications: (a) the classification of 3D visual objects; (b) the recognition hand-written digits (2D objects); and (c) the categorization of high-resolution electrocardiograms given as a time series (ID objects) and as a set of features extracted from these time series. In these applications, it can be observed that the performance of RBF classifiers trained with two-phase learning can be improved through a third backpropagation-like training phase of the RBF network, adapting the whole set of parameters (RBF centers, scaling parameters, and output layer weights) simultaneously. This, we call three-phase learning in RBF networks. A practical advantage of two- and three-phase learning in RBF networks is the possibility to use unlabeled training data for the first training phase. Support vector (SV) learning in RBF networks is a different learning approach. SV learning can be considered, in this context of learning, as a special type of one-phase learning, where only the output layer weights of the RBF network are calculated, and the RBF centers are restricted to be a subset of the training data. Numerical experiments with several classifier schemes including k-nearest-neighbor, learning vector quantization and RBF classifiers trained through two-phase, three-phase and support vector learning are given. The performance of the RBF classifiers trained through SV learning and three-phase learning are superior to the results of two-phase learning, but SV learning often leads to complex network structures, since the number of support vectors is not a small fraction of the total number of data points.

A Differentiation Checkpoint Limits Hematopoietic Stem Cell Self-Renewal in Response to DNA Damage

Checkpoints that limit stem cell self-renewal in response to DNA damage can contribute to cancer protection but may also promote tissue aging. Molecular components that control stem cell responses to DNA damage remain to be delineated. Using in vivo RNAi screens, we identified basic leucine zipper transcription factor, ATF-like (BATF) as a major component limiting self-renewal of hematopoietic stem cells (HSCs) in response to telomere dysfunction and γ-irradiation. DNA damage induces BATF in a G-CSF/STAT3-dependent manner resulting in lymphoid differentiation of HSCs. BATF deletion improves HSC self-renewal and function in response to γ-irradiation or telomere shortening but results in accumulation of DNA damage in HSCs. Analysis of bone marrow from patients with myelodysplastic syndrome supports the conclusion that DNA damage-dependent induction of BATF is conserved in human HSCs. Together, these results provide experimental evidence that a BATF-dependent differentiation checkpoint limits self-renewal of HSCs in response to DNA damage.

BoolNet--an R package for generation, reconstruction and analysis of Boolean networks.

MOTIVATION As the study of information processing in living cells moves from individual pathways to complex regulatory networks, mathematical models and simulation become indispensable tools for analyzing the complex behavior of such networks and can provide deep insights into the functioning of cells. The dynamics of gene expression, for example, can be modeled with Boolean networks (BNs). These are mathematical models of low complexity, but have the advantage of being able to capture essential properties of gene-regulatory networks. However, current implementations of BNs only focus on different sub-aspects of this model and do not allow for a seamless integration into existing preprocessing pipelines. RESULTS BoolNet efficiently integrates methods for synchronous, asynchronous and probabilistic BNs. This includes reconstructing networks from time series, generating random networks, robustness analysis via perturbation, Markov chain simulations, and identification and visualization of attractors. AVAILABILITY The package BoolNet is freely available from the R project at http://cran.r-project.org/ or http://www.informatik.uni-ulm.de/ni/mitarbeiter/HKestler/boolnet/ under Artistic License 2.0. CONTACT hans.kestler@uni-ulm.de SUPPLEMENTARY INFORMATION Supplementary data are available at Bioinformatics online.

NF-κB controls the global pro-inflammatory response in endothelial cells: evidence for the regulation of a pro-atherogenic program

Activation of the transcription factor NF-κB is critical for the tumor necrosis factor-α (TNF-α)-induced inflammatory response. Here we report the complete gene expression profile from activated microvascular endothelial cells emphasizing the direct contribution of the NF-κB pathway. Human microvascular endothelial cell line-1 (HMEC-1) cells were modified to express dominant interfering mutants of the IKK/NF-κB signaling module and expression profiles were determined. Our results provide compelling evidence for the virtually absolute dependence of TNF-α-regulated genes on NF-κB. A constitutively active IKK2 was sufficient for maximal induction of most target genes, whereas a transdominant IκBα suppressed gene expression. Several genes with a critical role in atherogenesis were identified. The endothelial lipase (EL) gene, a key enzyme involved in lipoprotein metabolism, was investigated more in detail. Binding sites interacting with NF-κB in vitro and in vivo were identified and co-transfection experiments demonstrated the direct regulation of the EL promoter by NF-κB. We conclude that targeting the IKK/NF-κB pathway or specific genes downstream may be effective for the control or prevention of chronic inflammatory diseases such as atherosclerosis.

Disclosure of Candidate Genes in Acute Myeloid Leukemia With Complex Karyotypes Using Microarray-Based Molecular Characterization

PURPOSE To identify novel genomic regions of interest in acute myeloid leukemia (AML) with complex karyotypes, we applied comparative genomic hybridization to microarrays (array-CGH), allowing high-resolution genome-wide screening of genomic imbalances. PATIENTS AND METHODS Sixty AML cases with complex karyotypes were analyzed using array-CGH; parallel analysis of gene expression was performed in a subset of cases. RESULTS Genomic losses were found more frequently than gains. The most frequent losses affected 5q (77%), 17p (55%), and 7q (45%), and the most frequent genomic gains 11q (40%) and 8q (38%). Critical segments could be delineated to genomic fragments of only 0.8 to a few megabase-pairs of DNA. In lost/gained regions, gene expression profiling detected a gene dosage effect with significant lower/higher average gene expression levels across the genes located in the respective regions. Furthermore, high-level DNA amplifications were identified in several regions: 11q23.3-q24.1 (n = 7), 21q22 (n = 6), 11q23.3 (n = 5), 13q12 (n = 3), 8q24 (n = 3), 9p24 (n = 2), 12p13 (n = 2), and 20q11 (n = 2). Parallel analysis of gene expression in critical amplicons displayed overexpressed candidate genes (eg, C8FW and MYC in 8q24). CONCLUSION In conclusion, a large spectrum of genomic imbalances, including novel recurring changes in AML with complex karyotypes, was identified using array-CGH. In addition, the combined analysis of array-CGH data with gene expression profiles allowed the detection of candidate genes involved in the pathogenesis of AML.

A canonical to non-canonical Wnt signalling switch in haematopoietic stem-cell ageing

Many organs with a high cell turnover (for example, skin, intestine and blood) are composed of short-lived cells that require continuous replenishment by somatic stem cells. Ageing results in the inability of these tissues to maintain homeostasis and it is believed that somatic stem-cell ageing is one underlying cause of tissue attrition with age or age-related diseases. Ageing of haematopoietic stem cells (HSCs) is associated with impaired haematopoiesis in the elderly. Despite a large amount of data describing the decline of HSC function on ageing, the molecular mechanisms of this process remain largely unknown, which precludes rational approaches to attenuate stem-cell ageing. Here we report an unexpected shift from canonical to non-canonical Wnt signalling in mice due to elevated expression of Wnt5a in aged HSCs, which causes stem-cell ageing. Wnt5a treatment of young HSCs induces ageing-associated stem-cell apolarity, reduction of regenerative capacity and an ageing-like myeloid–lymphoid differentiation skewing via activation of the small Rho GTPase Cdc42. Conversely, Wnt5a haploinsufficiency attenuates HSC ageing, whereas stem-cell-intrinsic reduction of Wnt5a expression results in functionally rejuvenated aged HSCs. Our data demonstrate a critical role for stem-cell-intrinsic non-canonical Wnt5a signalling in HSC ageing.

From individual Wnt pathways towards a Wnt signalling network

Wnt proteins play important roles during vertebrate and invertebrate development. They obviously have the ability to activate different intracellular signalling pathways. Based on the characteristic intracellular mediators used, these are commonly described as the Wnt/β-catenin, the Wnt/calcium and the Wnt/Jun N-terminal kinase pathways (also called planar cell polarity pathway). In the past, these different signalling events were mainly described as individual and independent signalling branches. Here, we discuss the possibility that Wnt proteins activate a complex intracellular signalling network rather than individual pathways and suggest a graph representation of this network. Furthermore, we discuss different ways of how to predict the specific outcome of an activation of this network in a particular cell type, which will require the use of mathematical models. We point out that the use of deterministic approaches via the application of differential equations is suitable to model only small aspects of the whole network and that more qualitative approaches are possibly a suitable starting point for the prediction of the global behaviour of such large protein interaction networks.

Transcriptome analysis of microdissected pancreatic intraepithelial neoplastic lesions

Pancreatic ductal adenocarcinoma (PDAC) carries the most dismal prognosis of all solid tumours. Both the late clinical presentation of patients, due to lack of early symptoms, as well as the rapid and aggressive course of the disease contribute to the extremely high mortality of this malignancy. Recently, a multistep progression model for PDAC integrating morphological, clinical and molecular evidence has been proposed. Putative precursor lesions, termed pancreatic intraepithelial neoplasia (PanIN), are classified into three different grades (PanIN-1 through -3) based on the degree of cellular atypia they display. We have conducted large-scale expression profiling analyses of microdissected cells from normal pancreatic ducts, PanINs of different grades and PDACs using whole-genome oligonucleotide microarrays. Verification of hybridisation results for selected genes was performed using quantitative real-time PCR and immunohistochemical analyses on PanIN tissue microarrays. Comparison of the expression profiles demonstrated that the greatest changes in gene expression occur between PanIN stages 1B and 2, suggesting that PanIN-2 may represent the first truly preneoplastic stage in PDAC progression. Our results identify a large number of potential target genes for the development of novel molecular diagnostic and therapeutic tools for the prevention and early diagnosis of PDAC and provide novel insights into the pathophysiological mechanisms involved in tumour progression in the pancreas.

Genomic DNA-Chip Hybridization Reveals a Higher Incidence of Genomic Amplifications in Pancreatic Cancer than Conventional Comparative Genomic Hybridization and Leads to the Identification of Novel Candidate Genes

Genomic analyses aimed at the detection of high-level DNA amplifications were performed on 13 widely used pancreatic cancer cell lines and 6 pancreatic tumor specimens. For these analyses, array-based comparative genomic hybridization (Matrix-CGH) onto dedicated microarrays was used. In comparison with chromosomal CGH (eight amplifications), a >3-fold number of DNA amplifications was detected (n = 29). The most frequent amplifications mapped to 7p12.3 (three pancreatic cancer cell lines and three pancreatic tumor specimens), 8q24 (four pancreatic cancer cell lines and one pancreatic tumor specimen), 11q13 (three pancreatic cancer cell lines and three pancreatic tumor specimens), and 20q13 (four pancreatic cancer cell lines and three pancreatic tumor specimens). Genes contained in the consensus regions were MYC (8q24), EGFR (7p12.3), and FGF3 (11q13). In six of seven pancreatic cancer cell lines and pancreatic tumor specimens with 20q13 amplifications, the novel candidate gene NFAT C2, which plays a role in the activation of cytokines, was amplified. Other amplifications also affected genes for which a pathogenetic role in pancreatic carcinoma has not been described, such as BCL10 and BCL6, two members of the BCL family. A subset of amplified genes was checked for overexpression by means of real-time PCR, revealing the highest expression levels for BCL6 and BCL10. Thus, Matrix-CGH allows the detection of a high number of amplifications, resulting in the identification of novel candidate genes in pancreatic cancer.