Search

Abstract Background As studies of molecular biology system attempt to achieve a comprehensive understanding of a particular system, Type 1 errors may be a significant problem. However, few investigators are inclined to accept the increase in Type 2 errors (false positives) that may result when less stringent statistical cut-off values are used. To address this dilemma, we developed an analysis strategy that used a stringent statistical analysis to create a list of differentially expressed genes that served as "bait" to "fish out" other genes with similar patterns of expression. Results Comparing two strains of mice (NOD and C57Bl/6), we identified 93 genes with statistically significant differences in their patterns of expression. Hierarchical clustering identified an additional 39 genes with similar patterns of expression differences between the two strains. Pathway analysis was then employed: 1) identify the central genes and define biological processes that may be regulated by the genes identified, and 2) identify genes on the lists that could not be connected to each other in pathways (potential false positives). For networks created by both gene lists, the most connected (central) genes were interferon gamma (IFN-γ) and tumor necrosis factor alpha (TNF-α). These two cytokines are relevant to the biological differences between the two strains of mice. Furthermore, the network created by the list of 39 genes also suggested other biological differences between the strains. Conclusion Taken together, these data demonstrate how stringent statistical analysis, combined with hierarchical clustering and pathway analysis may offer deeper insight into the biological processes reflected from a set of expression array data. This approach allows us to 'recapture" false negative genes that otherwise would have been missed by the statistical analysis.

Background: We investigated sera from elderly subjects with and without age-related macular degeneration (AMD) for presence of autoantibodies (AAbs) against human macular antigens and characterized their identity. Methods: Sera were collected from participants in the Age-Related Maculopathy Ancillary (ARMA) Study, a cross-sectional investigation ancillary to the Health ABC Study, enriched with participants from the general population. The resulting sample (mean age: 79.2±3.9 years old) included subjects with early to advanced AMD (n = 131) and controls (n = 231). Sera were tested by Western blots for immunoreactive bands against human donor macular tissue homogenates. Immunoreactive bands were identified and graded, and odds ratios (OR) calculated. Based on these findings, sera were immunoprecipitated, and subjected to 2D gel electrophoresis (GE). Liquid chromatography-tandem mass spectrometry (LC-MS/MS) was used to identify the targets recognized by circulating AAbs seen on 2D-GE, followed by ELISAs with recombinant proteins to confirm LC-MS/MS results, and quantify autoreactivities. Results: In AMD, 11 immunoreactive bands were significantly more frequent and 13 were significantly stronger than in controls. Nine of the more frequent bands also showed stronger reactivity. OR estimates ranged between 4.06 and 1.93, and all clearly excluded the null value. Following immunoprecipitation, 2D-GE and LC-MS/MS, five of the possible autoreactivity targets were conclusively identified: two members of the heat shock protein 70 (HSP70) family, HSPA8 and HSPA9; another member of the HSP family, HSPB4, also known as alpha-crystallin A chain (CRYAA); Annexin A5 (ANXA5); and Protein S100-A9, also known as calgranulin B that, when complexed with S100A8, forms calprotectin. ELISA testing with recombinant proteins confirmed, on average, significantly higher reactivities against all targets in AMD samples compared to controls. Conclusions: Consistent with other evidence supporting the role of inflammation and the immune system in AMD pathogenesis, AAbs were identified in AMD sera, including early-stage disease. Identified targets may be mechanistically linked to AMD pathogenesis because the identified proteins are implicated in autophagy, immunomodulation, and protection from oxidative stress and apoptosis. In particular, a role in autophagy activation is shared by all five autoantigens, raising the possibility that the detected AAbs may play a role in AMD via autophagy compromise and downstream activation of the inflammasome. Thus, we propose that the detected AAbs provide further insight into AMD pathogenesis and have the potential to contribute to disease biogenesis and progression. [ABSTRACT FROM AUTHOR]

Background: As studies of molecular biology system attempt to achieve a comprehensive understanding of a particular system, Type 1 errors may be a significant problem. However, few investigators are inclined to accept the increase in Type 2 errors (false positives) that may result when less stringent statistical cut-off values are used. To address this dilemma, we developed an analysis strategy that used a stringent statistical analysis to create a list of differentially expressed genes that served as "bait" to "fish out" other genes with similar patterns of expression. Results: Comparing two strains of mice (NOD and C57Bl/6), we identified 93 genes with statistically significant differences in their patterns of expression. Hierarchical clustering identified an additional 39 genes with similar patterns of expression differences between the two strains. Pathway analysis was then employed: 1) identify the central genes and define biological processes that may be regulated by the genes identified, and 2) identify genes on the lists that could not be connected to each other in pathways (potential false positives). For networks created by both gene lists, the most connected (central) genes were interferon gamma (IFN-?) and tumor necrosis factor alpha (TNF-a). These two cytokines are relevant to the biological differences between the two strains of mice. Furthermore, the network created by the list of 39 genes also suggested other biological differences between the strains. Conclusion: Taken together, these data demonstrate how stringent statistical analysis, combined with hierarchical clustering and pathway analysis may offer deeper insight into the biological processes reflected from a set of expression array data. This approach allows us to 'recapture" false negative genes that otherwise would have been missed by the statistical analysis. [ABSTRACT FROM AUTHOR]

Type 1 diabetes (T1D) is an autoimmune disease hypothesized, to be enhanced or triggered by viral infections. In this study we used organ donor pancreas to isolate islets that expressed known signs of viral infections, and subsequently investigated pathways that may be secondarily affected through analysis of gene expression profiles. Using samples provided by the Network for Pancreatic Organ Donors with Diabetes, (nPOD), islets were collected, by laser-capture, from nondiabetic donors (Control), autoantibody positive nondiabetic donors (AB+) and donors with T1D (T1D). Islets were categorized based on the presence of markers associated with viral infection (VIMs), HLA, Mx1, dsRNA, and PKR, identified via immunohistochemical staining. From each donor islets were pooled based on the number of VIMs (0 VIMs, 1 VIMs or ≥2VIMs). RNA was extracted, and microarray used to assess transcriptomes. GeneSpring software was used to generate a list of genes with differential expression between donors/VIMs. Using WebGestalt we identified pathways enriched by this list. A total of 383 genes with a fold change of ≥1.1 and p-value=0.001 had differential expression between islets with 0 VIMs and ≥1 VIMs. Pathway analysis showed strongest enrichment for genes in the insulin secretion pathway (p< 5.7E-5). As an example; CACNA1C and ABCC8 (important for the secretion of insulin), had lower expression in the islets with ≥1VIMs compared to 0 VIMs (p<0.03). CACNA1C also had lower expression in islets with 0 VIM vs. ≥1VIMs when only comparing samples from control donors (p=0.02). Islets selected for high expression for genes associated with viral infection had decreased expression of genes important for insulin secretion. This may suggest a pattern of dedifferentiation and/or functional impairment of beta cells in the setting of viral infection. Disclosure: G. Nelson: None. N.I. Lenchik: None. T. Rodriguez-Calvo: None. D. Balcacean: None. I.C. Gerling: None. Funding: JDRF; National Institute of Diabetes and Digestive and Kidney Diseases; Lebonheur Children's Hospital [ABSTRACT FROM AUTHOR]

NOD mice develop clear signs of leukocyte infiltration around their islets of Langerhans as early as age 5 weeks. Our studies focus on the molecular networks and processes preceding that pathological stage. Mononuclear spleen leukocytes were collected from NOD and control mouse strains (NON and C57B1/6) at ages 2 and 4 weeks (n=5 for each strain, at each age). Both proteins and RNA were extracted from each leukocyte sample. Differential expression was evaluated, in proteomes by 2D gel electrophoresis, and trancriptomes by Affymetrix expression arrays. Gene expression data at the trancript anal protein levels were subjected to analysis separately. The two expression patterns of interest were: A) Genes with unique high or low expression in NOD mice regardless of age, and, B) Genes with a unique change in expression, between age 2 and 4 weeks, only in NOD mice. Statistical analysis (ANOVA and t-test) was used to create lists of genes that follow those patterns, and those lists were further refined using hierarchical clustering. Gene lists were mined at a commercially available service (www.ingenuity.com) that uses a database of gene connections to create molecular networks from each uploaded gene list. Each network consists of genes from the list (called focus genes) and genes needed to connect these focus genes into tight molecular networks (non-focus genes). Although the list of genes following expression patterns A (strain differences) at the protein level and at the transcript level, did not share a single gene, the networks created from these two lists showed remarkable similarities sharing many of the same central non-focus genes. A similar observation (lack of overlap at the individual gene level but strong overlap at the pathway (systems) level), were seen when we compared proteome and transcriptome gene list following patterns B (developmental differences). An overall evaluation of the data pointed to a central role of the cytokines TNF-alfa, IFN-gamma and IL1-beta in explaining the unique gene expression in spleen leukocytes from young NOD mice compared to the two control strains. Based on this data we propose a new model for the pathogenesis of autoimmune diabetes in NOD mice. This model suggests that NOD mice are caught in a genetic catch 22. Their lack of responsiveness to TNF-alfa, IFN-gamma and IL1-beta prevents them from eliminating auto-reactive leukocytes. However, they are precluded from increasing their base level of these cytokines due to the toxicity of these cytokines to insulin producing beta cells. [ABSTRACT FROM AUTHOR]

Abstract: The B-lymphocyte marker CD72 has multiple alleles and serves crucial and nonredundant roles in B-cell development and activation. B-lymphocytes play an important role in development of autoimmune diabetes in NOD mice, therefore we examined the patterns of expression of CD72 and its alloantigens on splenic lymphocytes from 4-week-old NOD/LtJ mice. Comparisons of CD72 expression were made between NOD mice and three non-diabetic strains, NON/LtJ mice, C57BL/6J and AKR/J. Use of allele-specific monoclonal antibodies revealed that the previously uncharacterized NON strain expresses either the a or d allele, whereas NOD and AKR mice were confirmed to express the rare c allele. Flow cytometric analysis revealed differential expression between the strains. Whereas NON, C57BL/6 and AKR mice expressed CD72 on 98, 94 and 92% of their B-lymphocytes (CD19+ cells), the NOD mice only expressed this regulatory marker on 78% of their B-lymphocytes. Furthermore, CD72 expression levels on CD72 positive cells were lower in NOD mice than in other three non-diabetic strains. The presence of the CD72c allele, as well as its low level of expression in NOD mice at 4 weeks of age, may be associated with B-lymphocyte hyper-responsiveness and resistance to activation-induced cell death. [Copyright &y& Elsevier]

The article offers information on a study that explores the islet expression of type I interferon response sensors is associated with immune infiltration and viral infection in type 1 diabetes. It highlights the importance of therapeutic interventions aimed at eliminating potentially persistent infections and diminishing inflammation in individuals with type 1 diabetes.