GPNN: Power studies and applications of a neural network method for detecting gene-gene interactions in studies of human disease

Background The identification and characterization of genes that influence the risk of common, complex multifactorial disease primarily through interactions with other genes and environmental factors remains a statistical and computational challenge in genetic epidemiology. We have previously introduced a genetic programming optimized neural network (GPNN) as a method for optimizing the architecture of a neural network to improve the identification of gene combinations associated with disease risk. The goal of this study was to evaluate the power of GPNN for identifying high-order gene-gene interactions. We were also interested in applying GPNN to a real data analysis in Parkinson's disease. Results We show that GPNN has high power to detect even relatively small genetic effects (2–3% heritability) in simulated data models involving two and three locus interactions. The limits of detection were reached under conditions with very small heritability (

Real-time PCR assays for detection of bocavirus in human specimens

The recently discovered human bocavirus (HBoV) is the first member of the family Parpoviridae, genus Bocavirus, to be potentially associated with human disease. Several studies have identified HBoV in respiratory specimens from children with acute respiratory disease, but the full spectrum of clinical disease and the epidemiology of HBoV infection remain unclear. The availability of rapid and reliable molecular diagnostics would therefore aid future studies of this novel virus. To address this, we developed two sensitive and specific real-time TaqMan PCR assays that target the HBoV NS1 and NP-1 genes. Both assays could reproducibly detect 10 copies of a recombinant DNA plasmid containing a partial region of the HBoV genome, with a dynamic range of 8 log units (10(1) to 10(8) copies). Eight blinded clinical specimen extracts positive for HBoV by an independent PCR assay were positive by both real-time assays. Among 1,178 NP swabs collected from hospitalized pneumonia patients in Sa Kaeo Province, Thailand, 53 (4.5%) were reproducibly positive for HBoV by one or both targets. Our data confirm the possible association of HBoV infection with pneumonia and demonstrate the utility of these real-time PCR assays for HBoV detection.

GPNN: Power studies and applications of a neural network method for detecting gene-gene interactions in studies of human disease

Background: The identification and characterization of genes that influence the risk of common, complex multifactorial disease primarily through interactions with other genes and environmental factors remains a statistical and computational challenge in genetic epidemiology. We have previously introduced a genetic programming optimized neural network (GPNN) as a method for optimizing the architecture of a neural network to improve the identification of gene combinations associated with disease risk. The goal of this study was to evaluate the power of GPNN for identifying high-order gene-gene interactions. We were also interested in applying GPNN to a real data analysis in Parkinson's disease. Results: We show that GPNN has high power to detect even relatively small genetic effects (2-3% heritability) in simulated data models involving two and three locus interactions. The limits of detection were reached under conditions with very small heritability (

Localization of N-acetyltransferases NAT1 and NAT2 in human tissues

Human acetyl coenzyme A-dependent N-acetyltransferase (EC 2.3.1.5) (NAT) catalyzes the biotransformation of a number of arylamine and hydrazine compounds. NAT isozymes are encoded at 2 loci; one encodes NAT1, formerly known as the monomorphic form of the enzyme, while the other encodes the polymorphic NAT2, which is responsible for individual differences in the ability to acetylate certain compounds. Human epidemiological studies have suggested an association between the acetylator phenotype and particular cancers such as those of the bladder and colon. In the present study, NAT1- and NAT2-specific riboprobes were used in hybridization histochemistry studies to localize NAT1 and NAT2 mRNA sequences in formalin-fixed, paraffin-embedded human tissue sections. Expression of both NAT1 and NAT2 mRNA was observed in liver, gastrointestinal tract tissues (esophagus, stomach, small intestine, and colon), ureter, bladder, and lung. In extrahepatic tissues, NAT1 and NAT2 mRNA expression was localized to intestinal epithelial cells, urothelial cells, and the epithelial cells of the respiratory bronchioles. The observed heterogeneity of NAT1 and NAT2 mRNA expression between human tissue types may be of significance in assessing their contribution to known organ-specific toxicities of various arylamine drugs and carcinogens.

Differential localization and regulation of death and decoy receptors for TNF-related apoptosis-inducing ligand (TRAIL) in human melanoma cells

Induction of apoptosis in cells by TNF-related apoptosis-inducing ligand (TRAIL), a member of the TNF family, is believed to be regulated by expression of two death-inducing and two inhibitory (decoy) receptors on the cell surface. In previous studies we found no correlation between expression of decoy receptors and susceptibility of human melanoma cells to TRAIL-induced apoptosis, In view of this, we studied the localization of the receptors in melanoma cells by confocal microscopy to better understand their function. We show that the death receptors TRAIL-R1 and R2 are located in the trans-Golgi network, whereas the inhibitory receptors TRAIL-R3 and -R4 are located in the nucleus. After exposure to TRAIL, TRAIL-R1 and -R2 are internalized into endosomes, whereas TRAIL-R3 and -R4 undergo relocation from the nucleus to the cytoplasm and cell membranes. This movement of decoy receptors was dependent on signals from TRAIL-R1 and -R2, as shown by blocking experiments with Abs to TRAIL-R1 and -R2, The location of TRAIL-R1, -R3, and -R4 in melanoma cells transfected with cDNA for these receptors was similar to that in nontransfected cells, Transfection of TRAIL-R3 and -R4 increased resistance of the melanoma lines to TRAIL-induced apoptosis even in melanoma lines that naturally expressed these receptors. These results indicate that abnormalities in decoy receptor location or function may contribute to sensitivity of melanoma to TRAIL-induced apoptosis and suggest that further studies are needed on the functional significance of their nuclear location and TRAIL-induced movement within cell.

Glucocorticoid receptors in human preadipocytes: regional and gender differences

Glucocorticoid excess causes visceral obesity and its accompanying insulin resistance, dyslipidemia and hypertension. Glucocorticoids enhance preadipocyte (PA) differentiation and increase their aromatase activity (oestrogen production) and there is regional variability in these PA processes. Therefore, we studied human PAs for the presence of, and any regional or gender differences in, glucocorticoid receptors (GRs). Confluent subcultured human subcutaneous (Sc) and visceral (Vis) PAs from both genders contained GRs as assessed by GR gene expression and specific glucocorticoid (dexamethasone) binding. The dissociation constant was similar to that of other human cells and there was no difference between Sc and Vis sites or between males and females. There was significantly less GR mRNA in Vis PAs compared with Sc PAs in females (P=0.008) but not in males. There was less glucocorticoid binding in Vis compared with Sc PAs in females, measured by maximal binding capacity (P=0.035) or single saturating dose glucocorticoid binding (Bssd) (P=0.019). There was no regional difference in specific glucocorticoid binding in males. There was a gender difference with fewer GRs in Vis PAs in females compared with males measured by Bssd (P=0.006). In summary, GRs are present in human PAs. There is a lower GR density in Vis compared with Sc PAs in females, and females have fewer GRs in Vis PAs compared with males. These differences are likely to affect regional aromatase activity and to contribute to the smaller visceral fat mass in females compared with males.

Genetic association and cellular function of MC1R variant alleles in human pigmentation

We have examined MC1R variant allele frequencies in the general population of South East Queensland and in a collection of adolescent dizygotic and monozygotic twins and family members to define statistical associations with hair and skin color, freckling, and mole count. Results of these studies are consistent with a linear recessive allelic model with multiplicative penetrance in the inheritance of red hair. Four alleles, D84E, R151C, R160W, and D294H, are strongly associated with red hair and fair skin with multinomial regression analysis showing odds ratios of 63, 118, 50, and 94, respectively. An additional three low-penetrance alleles V60L, V92M, and R163Q have odds ratios 6, 5, and 2 relative to the wild-type allele. To address the cellular effects of MC1R variant alleles in signal transduction, we expressed these receptors in permanently transfected HEK293 cells. Measurement of receptor activity via induction of a cAMP-responsive luciferase reporter gene found that the R151C and R160W receptors were active in the presence of NDP-MSH ligand, but at much reduced levels compared with that seen with the wild-type receptor. The ability to stimulate phosphorylation of the cAMP response element binding protein (CREB) transcription factor was also apparent in all stimulated MC1R variant allele-expressing HEK293 cell extracts as assessed by immunoblotting. In contrast, human melanoma cell lines showed wide variation in the their ability to undergo cAMP-mediated CREB phosphorylation. Culture of human melanocytes of known MC1R genotype may provide the best experimental approach to examine the functional consequences for each MC1R variant allele. With this objective, we have established more than 300 melanocyte cell strains of defined MC1R genotype.

Chemokines in human periodontal disease tissues

An immunoperoxidase technique was used to examine IP-10 (interferon-gamma inducible protein 10), RANTES (regulated on activation normal T cell expressed and secreted), MCP-1 (monocyte chemoattractant protein-1), and MIP-1alpha (macrophage inflammatory protein-1alpha) in gingival biopsies from 21 healthy/gingivitis and 26 periodontitis subjects. The samples were placed into 3 groups according to the size of infiltrate. MIP-1alpha+ cells were more abundant than the other chemokines with few MCP-1+ cells. The mean percent MIP-1alpha+ cells was higher than the percent MCP-1+ cells (P = 0.02) in group 2 (intermediate size infiltrates) lesions from periodontitis subjects, other differences not being significant due to the large variations between tissue samples. Analysis of positive cells in relation to CD4/CD8 ratios showed that with an increased proportion of CD8+ cells, the mean percent MIP-1alpha+ cells was significantly higher in comparison with the mean percent RANTES+ and MCP-1+ cells (P < 0.015). Endothelial cells were MCP-1+ although positive capillaries were found on the periphery of infiltrates only. Keratinocyte expression of chemokines was weak and while the numbers of healthy/gingivitis and periodontitis tissue sections positive for IP-10, RANTES and MCP-1 reduced with increasing inflammation, those positive for MIP-1alpha remained constant for all groups. In conclusion, fewer leucocytes expressed MCP-1 in gingival tissue sections, however, the percent MIP-1alpha+ cells was increased particularly in tissues with increased proportions of CD8 cells and B cells with increasing inflammation and also in tissues with higher numbers of macrophages with little inflammation. Further studies are required to determine the significance of MIP-1alpha in periodontal disease.

RNA trafficking signals in human immunodeficiency virus type 1

Intracellular trafficking of retroviral RNAs is a potential mechanism to target viral gene expression to specific regions of infected cells. Here we show that the human immunodeficiency virus type 1 (HIV-1) genome contains two sequences similar to the hnRNP A2 response element (A2RE), a cis-acting RNA trafficking sequence that binds to the trans-acting trafficking factor, hnRNP A2, and mediates a specific RNA trafficking pathway characterized extensively in oligodendrocytes. The two HIV-1 sequences, designated A2RE-1, within the major homology region of the gag gene, and A2RE-2, in a region of overlap between the vpr and tat genes, both bind to hnRNP A2 in vitro and are necessary and sufficient for RNA transport in oligodendrocytes in vivo. A single base change (A8G) in either sequence reduces hnRNP A2 binding and, in the case of A2RE-2, inhibits RNA transport. A2RE-mediated RNA transport is microtubule and hnRNP A2 dependent. Differentially labelled gag and vpr RNAs, containing A2RE-1 and A2RE-2, respectively, coassemble into the same RNA trafficking granules and are cotransported to the periphery of the cell. tat RNA, although it contains A2RE-2, is not transported as efficiently as vpr RNA. An A2RE/hnRNP A2-mediated trafficking pathway for HIV RNA is proposed, and the role of RNA trafficking in targeting HIV gene expression is discussed.

The pyramidal cell in cognition: A comparative study in human and monkey

Here we present evidence that the pyramidal cell phenotype varies markedly in the cortex of different anthropoid species. Regional and species differences in the size of, number of bifurcations in, and spine density of the basal dendritic arbors cannot be explained by brain size. Instead, pyramidal cell morphology appears to accord with the specialized cortical function these cells perform. Cells in the prefrontal cortex of humans are more branched and more spinous than those in the temporal and occipital lobes. Moreover, cells in the prefrontal cortex of humans are more branched and more spinous than those in the prefrontal cortex of macaque and marmoset monkeys. These results suggest that highly spinous, compartmentalized, pyramidal cells (and the circuits they form) are required to perform complex cortical functions such as comprehension, perception, and planning.

Gag-Pol supplied in trans is efficiently packaged and supports viral function in human immunodeficiency virus type 1

The intracellular trafficking and subsequent incorporation of Gag-Pol into human immunodeficiency virus type 1 (HIV-1) remains poorly defined. Gag-Pol is encoded by the same mRNA as Gag and is generated by ribosomal frameshifting. The multimerization of Gag and Gag-Pol is an essential step in the formation of infectious viral particles. In this study, we examined whether the interaction between Gag and Gag-Pol is initiated during protein translation in order to facilitate the trafficking and subsequent packaging of Gag-Pol into the virion. A conditional cotransfection system was developed in which virion formation required the coexpression of two HIV-1-based plasmids, one that produces both Gag and Gag-Pol and one that only produces Gag-Pol. The Gag-Pol proteins were either immunotagged with a His epitope or functionally tagged with a mutation (K65R) in reverse transcriptase that is associated with drug resistance. Gag-Pol packaging was assessed to determine whether the Gag-Pol incorporated into the virion was preferentially packaged from the plasmid that expressed both Gag and Gag-Pol or whether it could be packaged from either plasmid. Our data show that translation of Gag and Gag-Pol from the same mRNA is not critical for virion packaging of the Gag-Pol polyprotein or for viral function.

Phenotypic characterization of five dendritic cell subsets in human tonsils

Heterogeneous expression of several antigens on the three currently defined tonsil dendritic cell (DC) subsets encouraged us to re-examine tonsil DCs using a new method that minimized DC differentiation and activation during their preparation. Three-color flow cytometry and dual-color immunohistology was used in conjunction with an extensive panel of antibodies to relevant DC-related antigens to analyze lin(-) HLA-DR+ tonsil DCs. Here we identify, quantify, and locate five tonsil DC subsets based on their relative expression of the HLA-DR, CD11c, CD13, and CD123 antigens. In situ localization identified four of these DC subsets as distinct interdigitating DC populations. These included three new interdigitating DC subsets defined as HLA-DRhi CD11c(+) DCs, HLA-DRmod CD11c+ CD13(+) DCs, and HLA-DRmod CD11c(-) CD123(-) DCs, as well as the plasmacytoid DCs (HLA-DRmod CD11c- CD123(+)). These subsets differed in their expression of DC-associated differentiation/activation antigens and co-stimulator molecules including CD83, CMRF-44, CMRF-56, 2-7, CD86, and 4-1BB ligand. The fifth HLA-DRmod CD11c(+) DC subset was identified as germinal center DCs, but contrary to previous reports they are redefined as lacking the CD13 antigen. The definition and extensive phenotypic analysis of these five DC subsets In human tonsil extends our understanding of the complexity of DC biology.