Knowledge-assisted recognition of cluster boundaries in gene expression data

BACKGROUND AND MOTIVATION: DNA microarray technology has made it possible to determine the expression levels of thousands of genes in parallel under multiple experimental conditions. Genome-wide analyses using DNA microarrays make a great contribution to the exploration of the dynamic state of genetic networks, and further lead to the development of new disease diagnosis technologies. An important step in the analysis of gene expression data is to classify genes with similar expression patterns into the same groups. To this end, hierarchical clustering algorithms have been widely used. Major advantages of hierarchical clustering algorithms are that investigators do not need to specify the number of clusters in advance and results are presented visually in the form of a dendrogram. However, since traditional hierarchical clustering methods simply provide results on the statistical characteristics of expression data, biological interpretations of the resulting clusters are not easy, and it requires laborious tasks to unveil hidden biological processes regulated by members in the clusters. Therefore, it has been a very difficult routine for experts. OBJECTIVE: Here, we propose a novel algorithm in which cluster boundaries are determined by referring to functional annotations stored in genome databases. MATERIALS AND METHODS: The algorithm first performs hierarchical clustering of gene expression profiles. Then, the cluster boundaries are determined by the Variance Inflation Factor among the Gene Function Vectors, which represents distributions of gene functions in each cluster. Our algorithm automatically specifies a cutoff that leads to functionally independent agglomerations of genes on the dendrogram derived from similarities among gene expression patterns. Finally, each cluster is annotated according to dominant gene functions within the respective cluster. RESULTS AND CONCLUSIONS: In this paper, we apply our algorithm to two gene expression datasets related to cell cycle and cold stress response in budding yeast Saccharomyces cerevisiae. As a result, we show that the algorithm enables us to recognize cluster boundaries characterizing fundamental biological processes such as the Early G1, Late G1, S, G2 and M phases in cell cycles, and also provides novel annotation information that has not been obtained by traditional hierarchical clustering methods. In addition, using formal cluster validity indices, high validity of our algorithm is verified by the comparison through other popular clustering algorithms, K-means, self-organizing map and AutoClass.     

Dominant-negative hypoxia-inducible factor-1 alpha reduces tumorigenicity of pancreatic cancer cells through the suppression of glucose metabolism

In the tumor cells exposed to hypoxia, hypoxia-inducible factor-1 (HIF-1)-mediated adaptation responses such as angiogenesis and anaerobic metabolism are induced for their survival. We have recently reported that the constitutive expression of HIF-1 alpha renders pancreatic cancer cells resistant to apoptosis induced by hypoxia and glucose deprivation. We then established dominant-negative HIF-1 alpha (dnHIF-1 alpha) transfectants and examined their susceptibility to apoptosis and growth inhibition induced by hypoxia and glucose deprivation in vitro and their tumorigenicity in SCID mice. We further examined the expressions of aldolase A and Glut-1 in vitro and Glut-1 expression and glucose uptake in the tumor tissues and microvessel counts in the tumor tissues. As a result, dnHIF-1 alpha rendered the pancreatic cancer cells sensitive to apoptosis and growth inhibition induced by hypoxia and glucose deprivation. Also it abrogated the enhanced expression of Glut-1 and aldolase A mRNAs under hypoxia and reduced the expression of Glut-1 and the glucose uptake in the tumor tissues and consequently in vivo tumorigenicity. We found no significant difference in the microvessel counts among the tumor tissues. From these results, we suggest that the disruption of the HIF-1 pathway might be effective in the treatment of pancreatic cancers.     

Enterostatin (Val-Pro-Asp-Pro-Arg), the activation peptide of procolipase, selectively reduces fat intake

Valine-proline-aspartate-proline-arginine (VPDPR), the amino terminal pentapeptide of pancreatic procolipase, produced a dose-dependent reduction in food intake when injected intraperitoneally into Osborne-Mendel rats that had been starved overnight. This inhibition of feeding was observed when the rats were fed a high-fat diet but not in rats fed a high-carbohydrate, low-fat diet. At higher doses of VPDPR, the inhibition of feeding was maintained for over 6 hours. An equimolar mixture of the free amino acids had no effect on food intake. In rats adapted to a three-choice macronutrient diet, VPDPR inhibited fat intake but had no effect on carbohydrate or protein intake. This selective inhibition of fat intake was observed in both overnight-fasted rats presented with food and in ad-lib-fed rats at the beginning of the dark-onset feeding period. It is suggested that this peptide may be a feedback signal to regulate the intake of dietary fat.     

Excitatory effect of adenosine on neurotransmission is due to increase of transmitter release in the hippocampal slices.

Adenosine has dose-dependent biphasic excitatory and inhibitory effects on neurotransmission in the hippocampus. The mechanism of the excitatory action is not known although that of the inhibitory action has been well analyzed. Here we report on the mechanism of excitatory action of adenosine, using hippocampal slices. Studies of intracellular recordings of CA3 pyramidal neurons showed that the amplitude of EPSP was dramatically enhanced by application of adenosine at low concentration (0.1 microM) without changing resting membrane potentials, membrane conductance or the threshold for spike generation by injecting current pulses. On the other hand, the presence of adenosine at a concentration of 0.1 microM during electrical stimulation to the slices increased 1.7 times the release of glutamate, an excitatory neurotransmitter in the hippocampus. These results indicate that the excitatory action of adenosine at low doses is due to the increase of transmitter release.     

Enhancement of CD3-mediated thymocyte apoptosis by the cross-linkage of heat-stable antigen.

Heat-stable antigen (HSA) is a murine differentiating antigen that is expressed on both CD4-CD8- double-negative and CD4+CD8+ double-positive thymocytes but not CD4+ or CD8+ single-positive thymocytes. Effects of anti-HSA monoclonal antibody, R13, on thymocyte apoptosis induced by various stimulations were investigated by a single-cell suspension culture system. Immobilized R13 enhanced the CD3-mediated DNA fragmentation and killing of thymocytes but not the dexamethasone-induced or phorbol myristate acetate-induced killing of thymocytes. Immobilized R13 by itself could not induce thymocyte apoptosis. Soluble R13 enhanced CD3-mediated apoptosis when HSA and T-cell receptor (TCR)/CD3 were co-cross-linked by a cross-reactive secondary antibody. Even without the cross-reactive secondary antibody, soluble R13 enhanced CD3-mediated apoptosis, although a greater than 100-fold increase in the amount of R13 was needed to give a similar enhancement compared with immobilized R13. Neither R13 by itself nor R13 plus secondary antibody induced cytosolic calcium influx, whereas R13 enhanced CD3-mediated cytosolic calcium increase. These results suggest a functional role of HSA in promoting the activation-induced apoptosis of thymocytes and the involvement of HSA in negative selection.     

Biological activities of glycoproteins of HVJ (Sendai virus) studied by reconstitution of hybrid envelope and by concanavalin A-mediated binding: a new function of HANA protein and structural requirement of F protein in hemolysis.

When HANA protein of HVJ (Sendai virus) was denatured by dithiothreitol treatment, concanavalin A added to the system or influenza virus hemagglutinin reconstituted with treated HVJ-envelope to form hybrid by SM-2 dialysis method could not replace a role of HANA for hemolysis. Binding of the treated samples to erythrocytes was restored by them, however. Thus, a hypothesis is proposed that intactness of not only F but also HANA protein of HVJ is required for envelope-cell fusion, in other words, HANA protein of HVJ may play some positive role other than binding in the process of the virus-induced hemolysis.