Integrative analysis of multiple gene expression profiles applied to liver cancer study

A statistical method for combining multiple microarray studies has been previously developed by the authors. Here, we present the application of the method to our hepatocellular carcinoma (HCC) data and report new findings on gene expression changes accompanying HCC. From the cross-verification result of our studies and that of published studies, we found that single microarray analysis might lead to false findings. To avoid those pitfalls of single-set analyses, we employed our effect size method to integrate multiple datasets. Of 9982 genes analyzed, 477 significant genes were identified with a false discovery rate of 10%. Gene ontology (GO) terms associated with these genes were explored to validate our method in the biological context with respect to HCC. Furthermore, it was demonstrated that the data integration process increases the sensitivity of analysis and allows small but consistent expression changes to be detected. These integration-driven discoveries contained meaningful and interesting genes not reported in previous expression profiling studies, such as growth hormone receptor, erythropoietin receptor, tissue factor pathway inhibitor-2, etc. Our findings support the use of meta-analysis for a variety of microarray data beyond the scope of this specific application.     

The p21-activated protein kinase-related kinase Cla4 is a coincidence detector of signaling by Cdc42 and phosphatidylinositol 4-phosphate

Signal transduction pathways that co-regulate a given biological process often are organized into networks by molecules that act as coincidence detectors. Phosphoinositides and the Rho-type GTPase Cdc42 regulate overlapping processes in all eukaryotic cells. However, the coincidence detectors that link these pathways into networks remain unknown. Here we show that the p21-activated protein kinase-related kinase Cla4 of yeast integrates signaling by Cdc42 and phosphatidylinositol 4-phosphate (PI4P). We found that the Cla4 pleckstrin homology (PH) domain binds in vitro to several phosphoinositide species. To determine which phosphoinositides regulate Cla4 in vivo, we analyzed phosphatidylinositol kinase mutants (stt4, mss4, and pik1). This indicated that the plasma membrane pool of PI4P, but not phosphatidylinositol 4,5-bisphosphate or the Golgi pool of PI4P, is required for localization of Cla4 to sites of polarized growth. A combination of the Cdc42-binding and PH domains of Cla4 was necessary and sufficient for localization to sites of polarized growth. Point mutations affecting either domain impaired the ability of Cla4 to regulate cell morphogenesis and the mitotic exit network (localization of Lte1). Therefore, Cla4 must retain the ability to bind both Cdc42 and phosphoinositides, the hallmark of a coincidence detector. PI4P may recruit Cla4 to the plasma membrane where Cdc42 activates its kinase activity and refines its localization to cortical sites of polarized growth. In mammalian cells, the myotonic dystrophy-related Cdc42-binding kinase possesses p21-binding and PH domains, suggesting that this kinase may be a coincidence detector of signaling by Cdc42 and phosphoinositides.     

Saturated fatty acid activates but polyunsaturated fatty acid inhibits Toll-like receptor 2 dimerized with Toll-like receptor 6 or 1

Toll-like receptor 4 (TLR4) and TLR2 agonists from bacterial origin require acylated saturated fatty acids in their molecules. Previously, we reported that TLR4 activation is reciprocally modulated by saturated and polyunsaturated fatty acids in macrophages. However, it is not known whether fatty acids can modulate the activation of TLR2 or other TLRs for which respective ligands do not require acylated fatty acids. A saturated fatty acid, lauric acid, induced NFkappaB activation when TLR2 was co-transfected with TLR1 or TLR6 in 293T cells, but not when TLR1, 2, 3, 5, 6, or 9 was transfected individually. An n-3 polyunsaturated fatty acid (docosahexaenoic acid (DHA)) suppressed NFkappaB activation and cyclooxygenase-2 expression induced by the agonist for TLR2, 3, 4, 5, or 9 in a macrophage cell line (RAW264.7). Because dimerization is considered one of the potential mechanisms for the activation of TLR2 and TLR4, we determined whether the fatty acids modulate the dimerization. However, neither lauric acid nor DHA affected the heterodimerization of TLR2 with TLR6 as well as the homodimerization of TLR4 as determined by co-immunoprecipitation assays in 293T cells in which these TLRs were transiently overexpressed. Together, these results demonstrate that lauric acid activates TLR2 dimers as well as TLR4 for which respective bacterial agonists require acylated fatty acids, whereas DHA inhibits the activation of all TLRs tested. Thus, responsiveness of different cell types and tissues to saturated fatty acids would depend on the expression of TLR4 or TLR2 with either TLR1 or TLR6. These results also suggest that inflammatory responses induced by the activation of TLRs can be differentially modulated by types of dietary fatty acids.     

Single-molecule three-color FRET

Fluorescence resonance energy transfer (FRET) measured at the single-molecule level can reveal conformational changes of biomolecules and intermolecular interactions in physiologically relevant conditions. Thus far single-molecule FRET has been measured only between two fluorophores. However, for many complex systems, the ability to observe changes in more than one distance is desired and FRET measured between three spectrally distinct fluorophores can provide a more complete picture. We have extended the single-molecule FRET technique to three colors, using the DNA four-way (Holliday) junction as a model system that undergoes two-state conformational fluctuations. By labeling three arms of the junction with Cy3 (donor), Cy5 (acceptor 1), and Cy5.5 (acceptor 2), distance changes between the donor and acceptor 1, and between the donor and acceptor 2, can be measured simultaneously. Thus we are able to show that the acceptor 1 arm moves away from the donor arm at the same time as the acceptor 2 arm approaches the donor arm, and vice versa, marking the first example of observing correlated movements of two different segments of a single molecule. Our data further suggest that Holliday junction does not spend measurable time with any of the helices unstacked, and that the parallel conformations are not populated to a detectable degree.     

Identification of chicken nebulin isoforms of the 31-residue motifs and non-muscle nebulin

Nebulin is a very large (M(r) 600-900kDa) actin-binding protein that is specific to skeletal muscle, and which is thought to act as a molecular template that regulates the length of sarcomere thin filaments. The 31-residue motif of nebulin contains a unique PEhXRVKXNQ consensus sequence. We have previously identified 11 different human nebulin isoforms of these 31-residue motifs. Here we present the identification of seven different isoforms (types II, III, IVa, IVb, V, VI, and X) of the 31-residue motifs in 15-day-old chicken embryo breast muscle. Isoform types II and III are also expressed in the brain, and type III is also detected in the heart, stomach, and liver. Chicken nebulin contains 11 copies of the 31-residue motif (R1a/b, R2, R3, R4, R5, R6, R7, R8, R9, R10, and R11), whereas human nebulin contains 13 copies. We confirmed the expression of nebulin in the heart, stomach, and brain in 15-day-old chicken embryos by immunofluorescence microscopy. The presence of nebulin in brain was further confirmed by in situ hybridization. These data suggest that there is even more diversity in nebulin isoforms than was previously known; this diversity likely contributes to the distinct actin filament architecture of different tissues.     

Identification of formaldehyde-induced modifications in proteins: reactions with model peptides

Formaldehyde is a well known cross-linking agent that can inactivate, stabilize, or immobilize proteins. The purpose of this study was to map the chemical modifications occurring on each natural amino acid residue caused by formaldehyde. Therefore, model peptides were treated with excess formaldehyde, and the reaction products were analyzed by liquid chromatography-mass spectrometry. Formaldehyde was shown to react with the amino group of the N-terminal amino acid residue and the side-chains of arginine, cysteine, histidine, and lysine residues. Depending on the peptide sequence, methylol groups, Schiff-bases, and methylene bridges were formed. To study intermolecular cross-linking in more detail, cyanoborohydride or glycine was added to the reaction solution. The use of cyanoborohydride could easily distinguish between peptides containing a Schiff-base or a methylene bridge. Formaldehyde and glycine formed a Schiff-base adduct, which was rapidly attached to primary N-terminal amino groups, arginine and tyrosine residues, and, to a lesser degree, asparagine, glutamine, histidine, and tryptophan residues. Unexpected modifications were found in peptides containing a free N-terminal amino group or an arginine residue. Formaldehyde-glycine adducts reacted with the N terminus by means of two steps: the N terminus formed an imidazolidinone, and then the glycine was attached via a methylene bridge. Two covalent modifications occurred on an arginine-containing peptide: (i) the attachment of one glycine molecule to the arginine residue via two methylene bridges, and (ii) the coupling of two glycine molecules via four methylene bridges. Remarkably, formaldehyde did not generate intermolecular cross-links between two primary amino groups. In conclusion, the use of model peptides enabled us to determine the reactivity of each particular cross-link reaction as a function of the reaction conditions and to identify new reaction products after incubation with formaldehyde.