Predicting Novel Human Gene Ontology Annotations Using Semantic Analysis

The correct interpretation of many molecular biology experiments depends in an essential way on the accuracy and consistency of the existing annotation databases. Such databases are meant to act as repositories for our biological knowledge as we acquire and refine it. Hence, by definition, they are incomplete at any given time. In this paper, we describe a technique that improves our previous method for predicting novel GO annotations by extracting implicit semantic relationships between genes and functions. In this work, we use a vector space model and a number of weighting schemes in addition to our previous latent semantic indexing approach. The technique described here is able to take into consideration the hierarchical structure of the Gene Ontology (GO) and can weight differently GO terms situated at different depths. The prediction abilities of 15 different weighting schemes are compared and evaluated. Nine such schemes were previously used in other problem domains, while six of them are introduced in this paper. The best weighting scheme was a novel scheme, n2tn. Out of the top 50 functional annotations predicted using this weighting scheme, we found support in the literature for 84 percent of them, while 6 percent of the predictions were contradicted by the existing literature. For the remaining 10 percent, we did not find any relevant publications to confirm or contradict the predictions. The n2tn weighting scheme also outperformed the simple binary scheme used in our previous approach.     

Circadian and seasonal variations of physiological and biochemical determinants of acute myocardial infarction

Almost all cardiovascular events occur according to a circadian rhythm with a greater frequency in the morning on waking and when resuming activity, the mechanism and precise triggering events for myocardial infarction (MI) are not yet fully known. Multiple biologic functions show a diurnal and/or seasonal variation that may contribute to adverse cardiac outomes. Exogenous factors may also modulates these variations. The MI peak usually occurs between 07:00 and 12:00 h. This timing corresponds to the concurrent increase in platelet aggregability, blood concentration of cortisol, catecholamines, angiotensin II, myocardial oxygen demand and coagulation activity, while fibrinolytic activity is decreased. In this review paper we will point out the biological rhythms of a number of functions involved in acute myocardial infarction e.g. blood pressure, hormonal determinants, cholesterol, among others.     

The chemical form of mitochondrial iron in Friedreich's ataxia

Friedreich's ataxia (FRDA) results from cellular damage caused by a deficiency in the mitochondrial matrix protein frataxin. To address the effect of frataxin deficiency on mitochondrial iron chemistry, the heavy mitochondrial fraction (HMF) was isolated from primary fibroblasts from FRDA affected and unaffected individuals. X-ray absorption spectroscopy was used to characterize the chemical form of iron. Near K-edge spectra were fitted with a series of model iron compounds to determine the proportion of each iron species. Most of the iron in both affected and unaffected fibroblasts was ferrihydrite. The iron K-edge from unaffected HMFs were best fitted with poorly organized ferrihydrite modeled by frataxin whereas HMFs from affected cells were best fitted with highly organized ferrihydrite modeled by ferritin. Both had several minor iron species but these did not differ consistently with disease. Since the iron K-edge spectra of ferritin and frataxin are very similar, we present additional evidence for the presence of ferritin-bound iron in HMF. The predominant ferritin subunit in HMFs from affected cells resembled mitochondrial ferritin (MtFt) in size and antigenicity. Western blotting of native gels showed that HMF from affected cells had 3-fold more holoferritin containing stainable iron. We conclude that most of the iron in fibroblast HMF from both affected and unaffected cells is ferrihydrite but only FRDA affected cells mineralize significant iron in mitochondrial ferritin.     

RRM proteins interacting with the cap region of topoisomerase I

RNA recognition motif (RRM) domains bind both nucleic acids and proteins. Several proteins that contain two closely spaced RRM domains were previously found in protein complexes formed by the cap region of human topoisomerase I, a nuclear enzyme responsible for DNA relaxation or phosphorylation of SR splicing proteins. To obtain molecular insight into specific interactions between the RRM proteins and the cap region of topo I we examined their binary interactions using the yeast two-hybrid system. The interactions were established for hnRNP A1, p54(nrb) and SF2/ASF, but not for hnRNP L or HuR. To identify the amino acid pattern responsible for binding, experimental mutagenesis was employed and computational modelling of these processes was carried out. These studies revealed that two RRM domains and six residues of the consensus sequence are required for the binding to the cap region. On the basis of the above data, a structural model for the hnRNP A1-topoisomerase I complex was proposed. The main component of the hnRNP A1 binding site is a hydrophobic pocket on the beta-surface of the first RRM domain, similar to that described for Y14 protein interacting with Mago. We demonstrated that the interaction between RRM domains and the cap region was important for the kinase reaction catalyzed by topoisomerase I. Together with the previously described inhibitory effect of RRM domains of SF2/ASF on DNA cleavage, the above suggests that the binding of RRM proteins could regulate the activity of topoisomerase I.     

Peripheral disruption of the Grb10 gene enhances insulin signaling and sensitivity in vivo

Grb10 is a pleckstrin homology and Src homology 2 domain-containing protein that interacts with a number of phosphorylated receptor tyrosine kinases, including the insulin receptor. In mice, Grb10 gene expression is imprinted with maternal expression in all tissues except the brain. While the interaction between Grb10 and the insulin receptor has been extensively investigated in cultured cells, whether this adaptor protein plays a positive or negative role in insulin signaling and action remains controversial. In order to investigate the in vivo role of Grb10 in insulin signaling and action in the periphery, we generated Grb10 knockout mice by the gene trap technique and analyzed mice with maternal inheritance of the knockout allele. Disruption of Grb10 gene expression in peripheral tissues had no significant effect on fasting glucose and insulin levels. On the other hand, peripheral-tissue-specific knockout of Grb10 led to significant overgrowth of the mice, consistent with a role for endogenous Grb10 as a growth suppressor. Loss of Grb10 expression in insulin target tissues, such as skeletal muscle and fat, resulted in enhanced insulin-stimulated Akt and mitogen-activated protein kinase phosphorylation. Hyperinsulinemic-euglycemic clamp studies revealed that disruption of Grb10 gene expression in peripheral tissues led to increased insulin sensitivity. Taken together, our results provide strong evidence that Grb10 is a negative regulator of insulin signaling and action in vivo.     

Aurora-A kinase phosphorylation of Aurora-A kinase interacting protein (AIP) and stabilization of the enzyme-substrate complex

Aurora-A is an oncogenic kinase that plays essential roles in mitosis as well as cell survival. Aurora-A interacting protein (AIP) was identified as a negative regulator of Aurora-A with its ectopic over expression inducing destabilization of Aurora-A protein. Here we present evidence that in human cells, contrary to the earlier report, AIP functions in stabilizing rather than destabilizing Aurora-A. Furthermore, AIP is phosphorylated on Serine 70 by Aurora-A but not Aurora-B and expression of phosphorylation mimic mutant of AIP results in prolonged protein stability compared to unphosphorylatable mutant. We observed that when co-expressed with AIP, protein levels of both Aurora-A and Aurora-B are markedly elevated regardless of their kinase activities and phosphorylation state of AIP. Interaction of Aurora kinases with AIP is necessary for this elevated stability. This phenomenon is commonly detected in several human cancer cell lines used in this study. Depletion of AIP by RNA interference decreased Aurora-A but not Aurora-B in two of the three cell lines analyzed, indicating that under physiological condition, AIP functions in stabilization of Aurora-A but not Aurora-B, though this regulation may be dependent on additional factors as well. Further, AIP siRNA induced cell cycle arrest at G2/M, which is consistent with anticipated loss of function of Aurora-A in these cells. Thus, our study provides the first evidence of a role for AIP in G2/M cell cycle progression by cooperatively regulating protein stabilization of its up-stream regulator, Aurora-A kinase through protein-protein interaction as well as protein phosphorylation.     

Differentially expressed genes in embryonic cardiac tissues of mice lacking <Emphasis Type="Italic">Folr1</Emphasis>gene activity

Background  

Heart anomalies are the most frequently observed among all human congenital defects. As with the situation for neural tube defects (NTDs), it has been demonstrated that women who use multivitamins containing folic acid peri-conceptionally have a reduced risk for delivering offspring with conotruncal heart defects [1–3]. Cellular folate transport is mediated by a receptor or binding protein and by an anionic transporter protein system. Defective function of the Folr1 (also known as Folbp1; homologue of human FRα) gene in mice results in inadequate transport, accumulation, or metabolism of folate during cardiovascular morphogenesis.     

Prins and C-PRINS: Promising tools for the physical mapping of the lupin genome

Two molecular cytogenetics methods, PRINS (primed in situ DNA labeling) and C-PRINS (cycling PRINS), were optimized for the physical mapping of several types of DNA sequences on the mitotic chromosomes of the narrow-leafed lupin (Lupinus angustifolius L.). The fragment of the FokI element from Vicia faba was localised by indirect PRINS reaction. Two other sequences, fragments of the coding sequences of L. luteus and of L. angustifolius, were localised by indirect C-PRINS. These techniques are faster and more sensitive than FISH, and they allowed the mapping of short DNA fragments. The data obtained shows that both types of PRINS are valuable tools for chromosome identification in lupin.