Artificial neural network prediction of viruses in shellfish

A database was probed with artificial neural network (ANN) and multivariate logistic regression (MLR) models to investigate the efficacy of predicting PCR-identified human adenovirus (ADV), Norwalk-like virus (NLV), and enterovirus (EV) presence or absence in shellfish harvested from diverse countries in Europe (Spain, Sweden, Greece, and the United Kingdom). The relative importance of numerical and heuristic input variables to the ANN model for each country and for the combined data was analyzed with a newly defined relative strength effect, which illuminated the importance of bacteriophages as potential viral indicators. The results of this analysis showed that ANN models predicted all types of viral presence and absence in shellfish with better precision than MLR models for a multicountry database. For overall presence/absence classification accuracy, ANN modeling had a performance rate of 95.9%, 98.9%, and 95.7% versus 60.5%, 75.0%, and 64.6% for the MLR for ADV, NLV, and EV, respectively. The selectivity (prediction of viral negatives) was greater than the sensitivity (prediction of viral positives) for both models and with all virus types, with the ANN model performing with greater sensitivity than the MLR. ANN models were able to illuminate site-specific relationships between microbial indicators chosen as model inputs and human virus presence. A validation study on ADV demonstrated that the MLR and ANN models differed in sensitivity and selectivity, with the ANN model correctly identifying ADV presence with greater precision.     

Regulation of reactive oxygen species-induced endothelial cell-cell and cell-matrix contacts by focal adhesion kinase and adherens junction proteins

Oxidants, generated by activated neutrophils, have been implicated in the pathophysiology of vascular disorders and lung injury; however, mechanisms of oxidant-mediated endothelial barrier dysfunction are unclear. Here, we have investigated the role of focal adhesion kinase (FAK) in regulating hydrogen peroxide (H(2)O(2))-mediated tyrosine phosphorylation of intercellular adhesion proteins and barrier function in endothelium. Treatment of bovine pulmonary artery endothelial cells (BPAECs) with H(2)O(2) increased tyrosine phosphorylation of FAK, paxillin, beta-catenin, and vascular endothelial (VE)-cadherin and decreased transendothelial electrical resistance (TER), an index of cell-cell adhesion and/or cell-matrix adhesion. To study the role of FAK in H(2)O(2)-induced TER changes, BPAECs were transfected with vector or FAK wild-type or FAK-related non-kinase (FRNK) plasmids. Overexpression of FRNK reduced FAK expression and attenuated H(2)O(2)-mediated tyrosine phosphorylation of FAK, paxillin, beta-catenin, and VE-cadherin and cell-cell adhesion. Additionally, FRNK prevented H(2)O(2)-induced distribution of FAK, paxillin, beta-catenin, or VE-cadherin toward focal adhesions and cell-cell adhesions but not actin stress fiber formation. These results suggest that activation of FAK by H(2)O(2) is an important event in oxidant-mediated VE barrier function regulated by cell-cell and cell-matrix contacts.     

Identification of multiple RNA features that influence CCR4 deadenylation activity

The CCR4 family proteins are 3'-5'-deadenylases that function in the first step of the degradation of poly(A) mRNA. Here we report the purification to homogeneity of the yeast CCR4 protein and the analysis of its substrate specificities. CCR4 deadenylated a 7N+23A substrate (seven nucleotides followed by 23 A residues) in a distributive manner. Only small differences in CCR4 activity for different A length substrates were observed until only 1 A residue remained. Correspondingly, the K(m) for a 25N+2A substrate was found to be at least 20-fold lower than that for a 26N+1A substrate, although their V(max) values differed by only 2-fold. In addition, the total length of the RNA was found to contribute to CCR4 activity: up to 17 nucleotides (not necessarily poly(A)) could be recognized by CCR4. Poly(U), poly(C), and poly(G) were also found to be 12-30-fold better inhibitors of CCR4 compared with poly(A), supporting the observation that CCR4 contains a non-poly(A)-specific binding site. Surprisingly, even longer substrates (>/=45 nucleotides) stimulated CCR4 to become a processive enzyme, suggesting that CCR4 undergoes an additional transition in the presence of such substrates. CCR4 also displayed no difference in its activity with capped or uncapped RNA substrates. These results indicate that CCR4 recognition of its RNA substrates involves several features of the RNA that could be sites in vivo for controlling the rate of specific mRNA deadenylation.     

High-density microarray of small-subunit ribosomal DNA probes

Ribosomal DNA sequence analysis, originally conceived as a way to provide a universal phylogeny for life forms, has proven useful in many areas of biological research. Some of the most promising applications of this approach are presently limited by the rate at which sequences can be analyzed. As a step toward overcoming this limitation, we have investigated the use of photolithography chip technology to perform sequence analyses on amplified small-subunit rRNA genes. The GeneChip (Affymetrix Corporation) contained 31,179 20-mer oligonucleotides that were complementary to a subalignment of sequences in the Ribosomal Database Project (RDP) (B. L. Maidak et al., Nucleic Acids Res. 29:173-174, 2001). The chip and standard Affymetrix software were able to correctly match small-subunit ribosomal DNA amplicons with the corresponding sequences in the RDP database for 15 of 17 bacterial species grown in pure culture. When bacteria collected from an air sample were tested, the method compared favorably with cloning and sequencing amplicons in determining the presence of phylogenetic groups. However, the method could not resolve the individual sequences comprising a complex mixed sample. Given these results and the potential for future enhancement of this technology, it may become widely useful.     

Hyperoxia-induced NAD(P)H oxidase activation and regulation by MAP kinases in human lung endothelial cells

Hyperoxia increases reactive oxygen species (ROS) production in vascular endothelium; however, the mechanisms involved in ROS generation are not well characterized. We determined the role and regulation of NAD(P)H oxidase in hyperoxia-induced ROS formation in human pulmonary artery endothelial cells (HPAECs). Exposure of HPAECs to hyperoxia for 1, 3, and 12 h increased the generation of superoxide anion, which was blocked by diphenyleneiodonium but not by rotenone or oxypurinol. Furthermore, hyperoxia enhanced NADPH- and NADH-dependent and superoxide dismutase- or diphenyleneiodonium-inhibitable ROS production in HPAECs. Immunohistocytochemistry and Western blotting revealed the presence of gp91, p67 phox, p22 phox, and p47 phox subcomponents of NADPH oxidase in HPAECs. Transfection of HPAECs with p22 phox antisense plasmid inhibited hyperoxia-induced ROS production. Exposure of HPAECs to hyperoxia activated p38 MAPK and ERK, and inhibition of p38 MAPK and MEK1/2 attenuated the hyperoxia-induced ROS generation. These results suggest a role for MAPK in regulating hyperoxia-induced NAD(P)H oxidase activation in HPAECs.     

Role of mitogen-activated protein kinases in 4-hydroxy-2-nonenal-induced actin remodeling and barrier function in endothelial cells

In vivo and in vitro studies indicate that 4-hydroxy-2-nonenal (4-HNE), generated by cellular lipid peroxidation or after oxidative stress, affects endothelial permeability and vascular tone. However, the mechanism(s) of 4-HNE-induced endothelial barrier function is not well defined. Here we provide evidence for the first time on the involvement of mitogen-activated protein kinases (MAPKs) in 4-HNE-mediated actin stress fiber formation and barrier function in lung endothelial cells. Treatment of bovine lung microvascular endothelial cells with hydrogen peroxide (H(2)O(2)), as a model oxidant, resulted in accumulation of 4-HNE as evidenced by the formation of 4-HNE-Michael protein adducts. Exposure of cells to 4-HNE, in a dose- and time-dependent manner, decreased endothelial cell permeability measured as transendothelial electrical resistance. The 4-HNE-induced permeability changes were not because of cytotoxicity or endothelial cell apoptosis, which occurred after prolonged treatment and at higher concentrations of 4-HNE. 4-HNE-induced changes in transendothelial electrical resistance were calcium independent, as 4-HNE did not alter intracellular free calcium levels as compared with H(2)O(2) or diperoxovanadate. Stimulation of quiescent cells with 4-HNE (1-100 microm) resulted in phosphorylation of ERK1/2, JNK, and p38 MAPKs, and actin cytoskeleton remodeling. Furthermore, pretreatment of bovine lung microvascular endothelial cells with PD 98059 (25 microm), an inhibitor of MEK1/2, or SP 600125 (25 microm), an inhibitor of JNK, or SB 202190 (25 microm), an inhibitor of p38 MAPK, partially attenuated 4-HNE-mediated barrier function and cytoskeletal remodeling. These results suggest that the activation of ERK, JNK, and p38 MAP kinases is involved in 4-HNE-mediated actin remodeling and endothelial barrier function.     

Role of a PDZ1 domain of NHERF1 in the binding of airway epithelial RACK1 to NHERF1

In past studies, we demonstrated regulation of CFTR Cl channel function by protein kinase C (PKC)- through the binding of PKC- to RACK1 (a receptor for activated C-kinase) and of RACK1 to human Na⁺/H⁺ exchanger regulatory factor (NHERF1). In this study, we investigated the site of RACK1 binding on NHERF1 using solid-phase and solution binding assays and pulldown, immunoprecipitation, and ³⁶Cl efflux experiments. Recombinant RACK1 binding to glutathione S-transferase (GST)-tagged PDZ1 domain of NHERF1 was 10-fold higher than its binding to GST-tagged PDZ2 domain of NHERF1. PDZ1 binds to RACK1 in a dose-dependent manner and vice versa, with similar binding constants of 1.67 and 1.26 µg, respectively. Interaction of the PDZ1 domain with RACK1 was not blocked by binding of activated PKC- to RACK1. A GST-tagged PDZ1 domain pulled down endogenous RACK1 from Calu-3 cell lysate. An internal 11-amino acid motif embedding the GYGF carboxylate binding loop of PDZ1 binds to RACK1, inhibits binding of recombinant NHERF1 and RACK1, pulls down endogenous RACK1 from Calu-3 cell lysate, and blocks coimmunoprecipitation of endogenous RACK1 with endogenous NHERF1 but does not affect cAMP-dependent activation of CFTR. A similar amino acid sequence in the PDZ2 domain did not bind RACK1. Our results indicate binding of Calu-3 RACK1 predominantly to the PDZ1 domain of NHERF1 at a site encompassing the GYGF loop of the PDZ1 domain and a site on RACK1 distinct from a PKC- binding site. CFTR activation by cAMP-generating agent is not affected by loss of RACK1-NHERF1 interaction. cystic fibrosis; cystic fibrosis transmembrane conductance regulator; protein-protein interaction; slot blot assay; pulldown; PDZ domain; chloride efflux; immunoprecipitation