maSigPro: a method to identify significantly differential expression profiles in time-course microarray experiments

MOTIVATION: Multi-series time-course microarray experiments are useful approaches for exploring biological processes. In this type of experiments, the researcher is frequently interested in studying gene expression changes along time and in evaluating trend differences between the various experimental groups. The large amount of data, multiplicity of experimental conditions and the dynamic nature of the experiments poses great challenges to data analysis. RESULTS: In this work, we propose a statistical procedure to identify genes that show different gene expression profiles across analytical groups in time-course experiments. The method is a two-regression step approach where the experimental groups are identified by dummy variables. The procedure first adjusts a global regression model with all the defined variables to identify differentially expressed genes, and in second a variable selection strategy is applied to study differences between groups and to find statistically significant different profiles. The methodology is illustrated on both a real and a simulated microarray dataset.     

Recursive regularization for inferring gene networks from time-course gene expression profiles

Background  

Inferring gene networks from time-course microarray experiments with vector autoregressive (VAR) model is the process of identifying functional associations between genes through multivariate time series. This problem can be cast as a variable selection problem in Statistics. One of the promising methods for variable selection is the elastic net proposed by Zou and Hastie (2005). However, VAR modeling with the elastic net succeeds in increasing the number of true positives while it also results in increasing the number of false positives.     

Information criterion-based clustering with order-restricted candidate profiles in short time-course microarray experiments

Background  

Time-course microarray experiments produce vector gene expression profiles across a series of time points. Clustering genes based on these profiles is important in discovering functional related and co-regulated genes. Early developed clustering algorithms do not take advantage of the ordering in a time-course study, explicit use of which should allow more sensitive detection of genes that display a consistent pattern over time. Peddada et al. [1] proposed a clustering algorithm that can incorporate the temporal ordering using order-restricted statistical inference. This algorithm is, however, very time-consuming and hence inapplicable to most microarray experiments that contain a large number of genes. Its computational burden also imposes difficulty to assess the clustering reliability, which is a very important measure when clustering noisy microarray data.     

A response to information criterion-based clustering with order-restricted candidate profiles in short time-course microarray experiments

Background  

For gene expression data obtained from a time-course microarray experiment, Liu et al. [1] developed a new algorithm for clustering genes with similar expression profiles over time. Performance of their proposal was compared with three other methods including the order-restricted inference based methodology of Peddada et al. [2, 3]. In this note we point out several inaccuracies in Liu et al. [1] and conclude that the order-restricted inference based methodology of Peddada et al. (programmed in the software ORIOGEN) indeed operates at the desired nominal Type 1 error level, an important feature of a statistical decision rule, while being computationally substantially faster than indicated by Liu et al. [1].     

Anthraquinone cytotoxicity and apoptosis in primary cultures of rat hepatocytes

We compared three different anthraquinones, rhein (4,5-dihydroxy-anthraquinone-2-carboxylic acid), danthron (1,8-dihydroxy-anthraquinone) and chrysophanol (1,8-dihydroxy-3-methylanthraquinone), with respect to their toxicity and ability to induce apoptosis in primary cultures of rat hepatocytes. Rhein was the most effective in producing free radicals, and was the only one of the tested anthraquinones that could induce apoptosis. Addition of 50μM rhein to hepatocyte cultures led to depletion of intracellular reduced glutathione (GSH) and ATP and accumulation of lipid peroxidation products. The substances N,N'-diphenyl-p-phenylenediamine (DPPD), dithiothreitol (DTT), nifedipine and desferal all protected the hepatocytes, i.e. prevented viability loss and ATP depletion, and decreased the GSH depletion.

Cultures exposed to rhein for 15min and subsequently rinsed and incubated for 16h under normal culture conditions (complete medium) exhibited apoptosis, as shown by DNA fragmentation, nuclear condensation and positive TUNEL reaction. Pretreatment with the antioxidant DPPD and the iron-chelator desferal gave complete protection against apoptosis.

No signs of oxidative cell damage were detected when the cultures were exposed to danthron or chrysophanol. All three anthraquinones did, however, cause an immediate increase in the intracellular Ca²⁺ concentration.

We conclude that rhein, which contains one carboxyl group, is a suitable substrate for one-electron-reducing enzymes and an effective redox cycler, which leads to the production of oxygen-derived free radicals that eventually induce apoptotic cell death.     

Anthraquinone cytotoxicity and apoptosis in primary cultures of rat hepatocytes

We compared three different anthraquinones, rhein (4,5-dihydroxy-anthraquinone-2-carboxylic acid), danthron (1,8-dihydroxy-anthraquinone) and chrysophanol (1,8-dihydroxy-3-methylanthraquinone), with respect to their toxicity and ability to induce apoptosis in primary cultures of rat hepatocytes. Rhein was the most effective in producing free radicals, and was the only one of the tested anthraquinones that could induce apoptosis. Addition of 50μM rhein to hepatocyte cultures led to depletion of intracellular reduced glutathione (GSH) and ATP and accumulation of lipid peroxidation products. The substances N,N′-diphenyl-p-phenylenediamine (DPPD), dithiothreitol (DTT), nifedipine and desferal all protected the hepatocytes, i.e. prevented viability loss and ATP depletion, and decreased the GSH depletion.

Cultures exposed to rhein for 15min and subsequently rinsed and incubated for 16h under normal culture conditions (complete medium) exhibited apoptosis, as shown by DNA fragmentation, nuclear condensation and positive TUNEL reaction. Pretreatment with the antioxidant DPPD and the iron-chelator desferal gave complete protection against apoptosis.

No signs of oxidative cell damage were detected when the cultures were exposed to danthron or chrysophanol. All three anthraquinones did, however, cause an immediate increase in the intracellular Ca²⁺ concentration.

We conclude that rhein, which contains one carboxyl group, is a suitable substrate for one-electron-reducing enzymes and an effective redox cycler, which leads to the production of oxygen-derived free radicals that eventually induce apoptotic cell death.     

Functional hierarchical models for identifying genes with different time-course expression profiles

Time-course studies of gene expression are essential in biomedical research to understand biological phenomena that evolve in a temporal fashion. We introduce a functional hierarchical model for detecting temporally differentially expressed (TDE) genes between two experimental conditions for cross-sectional designs, where the gene expression profiles are treated as functional data and modeled by basis function expansions. A Monte Carlo EM algorithm was developed for estimating both the gene-specific parameters and the hyperparameters in the second level of modeling. We use a direct posterior probability approach to bound the rate of false discovery at a pre-specified level and evaluate the methods by simulations and application to microarray time-course gene expression data on Caenorhabditis elegans developmental processes. Simulation results suggested that the procedure performs better than the two-way ANOVA in identifying TDE genes, resulting in both higher sensitivity and specificity. Genes identified from the C. elegans developmental data set show clear patterns of changes between the two experimental conditions.     

Increased metallothionein gene expression in 5-aza-2'-deoxycytidine-induced resistance to cadmium cytotoxicity

The pyrimidine analog, 5-azacytidine (AZA-CR), has been shown to increase the expression of the metallothionein (MT) gene and to induce tolerance to cadmium toxicity. Since incorporation into DNA of AZA-CR appears to be required for this effect, the deoxynucleoside of AZA-CR should also be effective. Therefore, this study was undertaken to assess the effect of 5-aza-2'-deoxycytidine (AZA-CdR) pretreatment on cadmium-induced cytotoxicity and MT expression in cultured cells. TRL 1215 cells in log phase of growth were exposed to AZA-CdR (0.4, 0.8, 4.0, 8.0 microM) followed 48 h later by the addition of cadmium (10 microM). MT concentrations were measured 24 h after the addition of cadmium. AZA-CdR alone caused modest, dose-related increases in MT levels (2.3-fold maximum), while cadmium alone resulted in a 9.5-fold increase. Pretreatment with AZA-CdR in combination with cadmium caused a 19--24-fold increase in cellular MT at all doses of AZA-CdR. Addition of the DNA synthesis inhibitor, hydroxyurea (HU), to the incubation medium during AZA-CdR exposure prevented the enhancing effect of the analog on cadmium induction of MT accumulation. Time course studies revealed that AZA-CdR pretreatment reduced the time required for cadmium to induce MT levels from 4--8 h to 0--2 h. AZA-CdR pretreated cells placed in suspension with cadmium (125 microM) showed a marked reduction in cadmium-induced cytotoxicity as reflected by reduced glutamic-oxaloacetic transaminase (GOT) loss. Uptake studies showed that AZA-CdR pretreatment had no effect on cadmium transport during the initial phases of exposure, indicating that an alteration in the toxicokinetics of the metal did not account for the reduction in toxicity. AZA-CdR did, however, cause hypomethylation of the MT-I gene. These results suggest that AZA-CdR pretreatment induces tolerance to cadmium toxicity by increasing the genetic expression of MT possibly through hypomethylation of the MT gene.     

Oligonucleotide microarray analysis of age-related gene expression profiles in miniature pigs

Miniature pigs are useful model animals for humans because they have similar anatomy and digestive physiology to humans and are easy to breed and handle. In this study, whole blood microarray analyses were conducted to evaluate variations of correlation among individuals and ages using specific pathogen-free (SPF) Clawn miniature pigs. Whole blood RNA is easy to handle compared to isolated white blood cell RNA and can be used for health and disease monitoring and animal control. In addition, whole blood is a heterogeneous mixture of subpopulation cells. Once a great change occurs in composition and expressing condition of subpopulations, their associated change will be reflected on whole blood RNA. From 12 to 30 weeks of age, fractions of lymphocytes, monocytes, neutrophils, eosinophils, and basophils in white blood cells showed insignificant differences with age as a result of ANOVA analysis. This study attempted to identify characteristics of age-related gene expression by taking into account the change in the number of expressed genes by age and similarities of gene expression intensity between individuals. As a result, the number of expressed genes was less in fetal stage and infancy period but increased with age, reaching a steady state of gene expression after 20 weeks of age. Variation in gene expression intensity within the same age was great in fetal stage and infancy period, but converged with age. The variation between 20 and 30 weeks of age was comparable to that among 30 weeks individuals. These results indicate that uniformity of laboratory animals is expected for miniature pigs after 20 weeks of age. Furthermore, a possibility was shown that whole blood RNA analysis is applicable to evaluation of physiological state.     

DGEM--a microarray gene expression database for primary human disease tissues

Gene expression patterns can reflect gene regulations in human tissues under normal or pathologic conditions. Gene expression profiling data from studies of primary human disease samples are particularly valuable since these studies often span many years in order to collect patient clinical information and achieve a large sample size. Disease-to-Gene Expression Mapper (DGEM) provides a beneficial community resource to access and analyze these data; it currently includes Affymetrix oligonucleotide array datasets for more than 40 human diseases and 1400 samples. The data are normalized to the same scale and stored in a relational database. A statistical-analysis pipeline was implemented to identify genes abnormally expressed in disease tissues or genes whose expressions are associated with clinical parameters such as cancer patient survival. Data-mining results can be queried through a web-based interface at http://dgem.dhcp.iupui.edu/. The query tool enables dynamic generation of graphs and tables that are further linked to major gene and pathway resources that connect the data to relevant biology, including Entrez Gene and Kyoto Encyclopedia of Genes and Genomes (KEGG). In summary, DGEM provides scientists and physicians a valuable tool to study disease mechanisms, to discover potential disease biomarkers for diagnosis and prognosis, and to identify novel gene targets for drug discovery. The source code is freely available for non-profit use, on request to the authors.     

Age-related decline of inducible nitric oxide synthase gene expression in primary cultured rat hepatocytes

Hepatocytes exhibit a non-specific immune response by expressing the enzyme inducible nitric oxide (NO) synthase (iNOS, NOS2) through the stimulation of a mixture of cytokines, or a single cytokine such as interleukin-1beta. We examined the age-dependent inducibility of the iNOS gene expression and the capacity of NO production in response to lipopolysaccharide (LPS) or interleukin-1beta (IL-1beta) in primary cultured rat hepatocytes that were isolated from the livers of rats, 3 (young) and 24 (aging) months of age. NO production (NO2-), indicating iNOS activity, was much higher in the young rat hepatocytes following stimulation with LPS or IL-1beta. Likewise, in the young hepatocytes, Western blot analyses showed much higher protein levels in the iNOS expression; it was also a little higher in mRNA levels that were analyzed by RT-PCR. Furthermore, after stimulation with IL-1beta, the levels of transactivation of nuclear factor-KB (NF-kappaB) that were involved in the induction of the iNOS gene were reduced without a significant difference in the aged cells. Therefore, the decrease of NO formation in the aged hepatocytes was due to the belated and incomplete inducibility of the iNOS protein expression, together with a minor contribution of the reduced-transactivation of NF-kappaB. These results suggest that the age-related decline of the iNOS gene expression in primary rat hepatocytes may be associated with the increased incidence of many infective diseases with aging.     

Context-specific infinite mixtures for clustering gene expression profiles across diverse microarray dataset

MOTIVATION: Identifying groups of co-regulated genes by monitoring their expression over various experimental conditions is complicated by the fact that such co-regulation is condition-specific. Ignoring the context-specific nature of co-regulation significantly reduces the ability of clustering procedures to detect co-expressed genes due to additional 'noise' introduced by non-informative measurements. RESULTS: We have developed a novel Bayesian hierarchical model and corresponding computational algorithms for clustering gene expression profiles across diverse experimental conditions and studies that accounts for context-specificity of gene expression patterns. The model is based on the Bayesian infinite mixtures framework and does not require a priori specification of the number of clusters. We demonstrate that explicit modeling of context-specificity results in increased accuracy of the cluster analysis by examining the specificity and sensitivity of clusters in microarray data. We also demonstrate that probabilities of co-expression derived from the posterior distribution of clusterings are valid estimates of statistical significance of created clusters. AVAILABILITY: The open-source package gimm is available at http://eh3.uc.edu/gimm.     

Intracellular oxidative stress and cytotoxicity in rat primary cortical neurons exposed to cholesterol secoaldehyde

Cholesterol secoaldehyde (ChSeco or 3β-hydroxy-5-oxo-5,6-secocholestan-6-al) has been shown to induce Aβ aggregation and apoptosis in GT1-7 hypothalamic neurons. The present study was undertaken to evaluate the effects of ChSeco on rat primary cortical neuronal cells. ChSeco was cytotoxic at concentrations ranging from 5 to 20 μM, while cholesterol of comparable concentrations showed little or no toxicity. In ChSeco-exposed neuronal cells, there was an increased formation of intracellular peroxide or peroxide-like substance(s), the levels of which were comparable to those found in typical menadione exposures. There was a loss in the mitochondrial transmembrane potential, the extent of which was dependent on concentration of ChSeco employed. Pre-treatment with N-acetyl-l-cysteine (5 mM; 1 h) offered protection against the cytotoxicity and the generation of intracellular oxidants. Cytotoxicity of ChSeco was evidenced by the loss of axonal branches and also condensed apoptotic nuclei in these cells. Immunohistochemical analysis revealed a decreased intracellular Aβ42 staining proportional to the loss in the axonal out growth and dendritic branches. The observed decrease in Aβ42 has been suggested to be due to loss of integrity of dendrites and the plasma membrane, possibly resulting from increased production of reactive oxygen species.     

NF-kappaB activates fibronectin gene expression in rat hepatocytes

Resveratrol (RSVL), an edible polyphenolic stilbene, claims a myriad of cardiovascular benefits. However, the molecular underpinnings of such actions are poorly understood. Currently, in sheep coronary arteries (SCA), RSVL markedly (threefold) enhanced cGMP formation (t(1/2): 6.5 min; EC(50): 3 microM). This response was not abrogated by the phosphodiesterase inhibitor (IBMX, 0.5 mM), but was partly sensitive (20-30%) to either removal of the endothelium, treatment with the nitric oxide synthase-inhibitor (L-NMMA, 10 microM), or with the soluble GC (sGC)-inhibitor (ODQ, 10 microM). In membrane preparations from denuded SCA, either RSVL or the pGC agonist atrial natriuretic peptide (ANP, 0.1-1 microM) activated GC in the particulate, but not in the soluble, membrane fraction. By contrast, the nitric oxide donor, sodium nitroprusside (SNP, 1-10 microM), stimulated GC only in the soluble fraction. Further, pretreatment with RSVL partly desensitized the ANP response, but was additive to that of SNP. In arterial tension studies, RSVL relaxed PGF(2alpha)-precontracted denuded rings in a concentration-dependent manner, a response that was markedly enhanced (approximately 18 fold) in the presence of IBMX. Conversely, precontraction with phorbol ester, which also desensitizes pGC, blunted relaxations to RSVL but not to forskolin or SNP. These findings demonstrate that RSVL increases cGMP in coronary arteries, mostly by activation of pGC. This pathway triggers vasorelaxant responses that remain effective in endothelium-disrupted arteries.