INTEGRATIVE BIOMARKER IDENTIFICATION AND CLASSIFICATION USING HIGH THROUGHPUT ASSAYS

It is well accepted that tumorigenesis is a multi-step procedure involving aberrant functioning of genes regulating cell proliferation, differentiation, apoptosis, genome stability, angiogenesis and motility. To obtain a full understanding of tumorigenesis, it is necessary to collect information on all aspects of cell activity. Recent advances in high throughput technologies allow biologists to generate massive amounts of data, more than might have been imagined decades ago. These advances have made it possible to launch comprehensive projects such as (TCGA) and (ICGC) which systematically characterize the molecular fingerprints of cancer cells using gene expression, methylation, copy number, microRNA and SNP microarrays as well as next generation sequencing assays interrogating somatic mutation, insertion, deletion, translocation and structural rearrangements. Given the massive amount of data, a major challenge is to integrate information from multiple sources and formulate testable hypotheses. This thesis focuses on developing methodologies for integrative analyses of genomic assays profiled on the same set of samples. We have developed several novel methods for integrative biomarker identification and cancer classification. We introduce a regression-based approach to identify biomarkers predictive to therapy response or survival by integrating multiple assays including gene expression, methylation and copy number data through penalized regression. To identify key cancer-specific genes accounting for multiple mechanisms of regulation, we have developed the integIRTy software that provides robust and reliable inferences about gene alteration by automatically adjusting for sample heterogeneity as well as technical artifacts using Item Response Theory. To cope with the increasing need for accurate cancer diagnosis and individualized therapy, we have developed a robust and powerful algorithm called SIBER to systematically identify bimodally expressed genes using next generation RNAseq data. We have shown that prediction models built from these bimodal genes have the same accuracy as models built from all genes. Further, prediction models with dichotomized gene expression measurements based on their bimodal shapes still perform well. The effectiveness of outcome prediction using discretized signals paves the road for more accurate and interpretable cancer classification by integrating signals from multiple sources.

Reduced neutrophil count in people of African descent is due to a regulatory variant in the Duffy antigen receptor for chemokines gene.

Persistently low white blood cell count (WBC) and neutrophil count is a well-described phenomenon in persons of African ancestry, whose etiology remains unknown. We recently used admixture mapping to identify an approximately 1-megabase region on chromosome 1, where ancestry status (African or European) almost entirely accounted for the difference in WBC between African Americans and European Americans. To identify the specific genetic change responsible for this association, we analyzed genotype and phenotype data from 6,005 African Americans from the Jackson Heart Study (JHS), the Health, Aging and Body Composition (Health ABC) Study, and the Atherosclerosis Risk in Communities (ARIC) Study. We demonstrate that the causal variant must be at least 91% different in frequency between West Africans and European Americans. An excellent candidate is the Duffy Null polymorphism (SNP rs2814778 at chromosome 1q23.2), which is the only polymorphism in the region known to be so differentiated in frequency and is already known to protect against Plasmodium vivax malaria. We confirm that rs2814778 is predictive of WBC and neutrophil count in African Americans above beyond the previously described admixture association (P = 3.8 x 10(-5)), establishing a novel phenotype for this genetic variant.

Identification of antigen-encoding genes of Enterococcus faecalis during human infections and characterization of a gene cluster for the biosynthesis of an antigenic polysaccharide

Genomic libraries of two Enterococcus faecalis strains, OG1RF and TX52 (an isolate from an endocarditis patient), were constructed in Escherichia coli and were screened with serum from a rabbit immunized with surface proteins of an E. faecalis endocarditis isolate and sera from four patients with enterococcal endocarditis. Thirty-eight immunopositive cosmid clones reacted with at least two of the patient sera and contained distinct inserts based on their DNA restriction patterns. These were chosen for further subcloning in a pBluescript SK ($-$) vector. Each sublibrary was screened with one of the five sera. Analysis of sequences from the immunopositive subclones revealed similarities to a range of proteins, including bacterial virulence factors, transporters, two-component regulators, metabolic enzymes, and membrane or cell surface proteins. Fourteen subclones did not show significant similarity to any sequence in the databases and may contain novel genes. Thirteen of the immunopositive cosmid clones did not yield immunopositive subclones and one such cosmid clone, TX5159, produced an antigenic polysaccharide in Escherichia coli. The insert of TX5159 was found to contain a multicistronic gene cluster containing genes similar to those involved in the biosynthesis and export of polysaccharides from both Gram-positive and Gram-negative organisms. Insertions in several genes within the cluster abolished the immunoreactivity of TX5159. RT-PCR of genes within the cluster with total RNA from OG1RF showed that these genes are transcribed. The polysaccharide was detected in two recently reported E. faecalis mucoid strains using specific antibody, but not in the other strains tested. This is the first report on a gene cluster of E. faecalis involved in the biosynthesis of an antigenic polysaccharide. ^

PHENOTYPIC ASSOCIATION OF GENE AMPLIFICATION WITH MULTIPLE DRUG RESISTANCE IN VINCRISTINE RESISTANT CHINESE HAMSTER OVARY CELLS

Resistance of tumors to pharmacologic agents poses a significant problem in the treatment of human malignancies. This study overviews the scope of clinical resistance and focuses upon current research attempts toward investigation of the phenomenon of multidrug resistance (MDR).^ The objective of this investigation was to determine whether gene amplification had a role in the development of the MDR phenotype in Chinese hamster ovary cells (CHO) primarily selected for resistance to vincristine (VCR). A DNA fragment, previously shown to be amplified in two independently derived Chinese hamster cell lines exhibiting the MDR phenotype, was also amplified in VCR hamster lines. Sequences flanking this fragment were shown to contain coding information for a 4.3 kb transcript overproduced in VCR cells. These sequences were not enriched in double minute DNA preparations isolated from VCR cells. There was an approximately forty-fold increase in both the level of gene amplification and transcript overproduction in the VCR cell lines, independent of the level of primary resistance. This DNA amplification and overproduction of the 4.3 kb transcript was also demonstrated in CHO cells independently selected for resistance to Adriamycin and vinblastine.^ All the DNA sequences of two hamster cDNA clones containing 785 and 932 base pair inserts showed direct homology to the published mouse mdr sequences (about 90%). This sequence conservation held for only portions of the gene when the human mdr1 sequences were compared with those from either the mouse or hamster.^ Somatic cell hybrids, constructed between VCR CHO cells and sensitive murine cells, were used to determine whether there was a functional relationship between the chromosome bearing the amplified sequences and the MDR phenotype. Concordant segregation between vincristine resistance, the MDR phenotype, the presence of MDR-associated amplified sequences, overexpression of the mRNA encoded by these sequences, overexpression of the mRNA encoded by these sequences, and CHO chromosome Z1 was consistent with the hypothesis that there is an amplified gene on chromosome Z1 of the VCR CHO cells which is responsible for MDR in these cells. ^

Comparison of multiple comparison methods for identifying differential gene expression in simulated and real papillary thyroid cancer microarray data

The difficulty of detecting differential gene expression in microarray data has existed for many years. Several correction procedures try to avoid the family-wise error rate in multiple comparison process, including the Bonferroni and Sidak single-step p-value adjustments, Holm's step-down correction method, and Benjamini and Hochberg's false discovery rate (FDR) correction procedure. Each multiple comparison technique has its advantages and weaknesses. We studied each multiple comparison method through numerical studies (simulations) and applied the methods to the real exploratory DNA microarray data, which detect of molecular signatures in papillary thyroid cancer (PTC) patients. According to our results of simulation studies, Benjamini and Hochberg step-up FDR controlling procedure is the best process among these multiple comparison methods and we discovered 1277 potential biomarkers among 54675 probe sets after applying the Benjamini and Hochberg's method to PTC microarray data.^

Regulation of {\it neu\/} gene expression by the simian virus 40 large T antigen and tumor suppressors Rb and p53

The neu gene (also c-erbB-2 or HER2) encodes a 185 kilodalton protein that is frequently overexpressed in breast, ovarian and non-small cell lung cancers. Study of the regulation of neu indicates that neu gene expression can be modulated by c-myc or by the adenovirus 5 E1a gene product. This study demonstrates that the transforming protein, large T antigen, of the simian virus 40 represses neu promoter activity. Repression of neu by large T antigen is mediated through the region $-$172 to $-$79 (relative to first ATG) of the neu promoter--unlike through $-$312 to $-$172 for c-myc or E1a. This suggests a different pathway for repression of neu by large T antigen. The 10 amino acid region of large T required for binding the tumor suppressor, retinoblastoma gene product, Rb, is not necessary for repression of neu. Moreover, the tumor suppressors, Rb and p53 can independently inhibit neu promoter activity. Rb inhibits neu through a 10 base pair G-rich enhancer (GTG element) ($-$243 to $-$234) and also through regions close to transcription initiation sites ($-$172 to $-$79). Mutant Rb unable to complex large T is able to repress the region close to transcription initiation but not the GTG enhancer. Thus, Rb inhibits the two regulatory domains of the neu gene by different mechanisms. Both Rb and p53 can repress the transforming activity of activated neu in focus forming assays. These data provide evidence that tumor suppressors regulate expression of growth stimulatory genes such as neu. Therefore, one reason for the overexpression of neu that is frequently seen in breast cancer cells may be due to functional inactivation of Rb and p53 which is also a common occurrence in breast cancer cells. ^

MULTIPLE POSTTRANSCRIPTIONAL REGULATORY FEATURES CONTROL EXPRESSION OF ETHANOLAMINE UTILIZATION GENES IN ENTEROCOCCUS FAECALIS

Enterococcus faecalis is a Gram-positive bacterium that lives as a commensal organism in the mammalian gastrointestinal tract, but can behave as an opportunistic pathogen. Our lab discovered that mutation of the eutK gene attenuates virulence of E. faecalis in the C. elegans model host. eutK is part of the ethanolamine metabolic pathway which was previously unknown in E. faecalis. I discovered the presence of two unique posttranscriptional regulatory features that control expression of eut locus genes. The first feature I found is an AdoCBL riboswitch, a cis-acting RNA regulatory element that acts as a positive regulator of gene expression. The second feature I discovered is a unique two-component system, EutVW. The EutV response regulator contains an ANTAR family domain, which binds RNA to trigger transcriptional antitermination. I determined that induction of expression of several genes in the eut locus is dependent on ethanolamine, AdoCBL and the two-component system. AdoCBL and ethanolamine are both required for induction of eut locus gene expression. Additionally, I discovered eutG is regulated by a unique mechanism of antitermination. Both the AdoCBL riboswitch and EutV response regulator control the expression of the downstream gene eutG. EutV potentially acts through a novel antitermination mechanism in which a dimer of EutV binds to a pair of mRNA stem loops forming an antitermination complex. My data show a unique mechanism by which two environmental signals are integrated by two different posttranscriptional regulators to regulate a single locus.

Identification and phenotypic characterization of a second collagen adhesin, Scm, and genome-based identification and analysis of 13 other predicted MSCRAMMs, including four distinct pilus loci, in Enterococcus faecium.

Attention has recently been drawn to Enterococcus faecium because of an increasing number of nosocomial infections caused by this species and its resistance to multiple antibacterial agents. However, relatively little is known about the pathogenic determinants of this organism. We have previously identified a cell-wall-anchored collagen adhesin, Acm, produced by some isolates of E. faecium, and a secreted antigen, SagA, exhibiting broad-spectrum binding to extracellular matrix proteins. Here, we analysed the draft genome of strain TX0016 for potential microbial surface components recognizing adhesive matrix molecules (MSCRAMMs). Genome-based bioinformatics identified 22 predicted cell-wall-anchored E. faecium surface proteins (Fms), of which 15 (including Acm) had characteristics typical of MSCRAMMs, including predicted folding into a modular architecture with multiple immunoglobulin-like domains. Functional characterization of one [Fms10; redesignated second collagen adhesin of E. faecium (Scm)] revealed that recombinant Scm(65) (A- and B-domains) and Scm(36) (A-domain) bound to collagen type V efficiently in a concentration-dependent manner, bound considerably less to collagen type I and fibrinogen, and differed from Acm in their binding specificities to collagen types IV and V. Results from far-UV circular dichroism measurements of recombinant Scm(36) and of Acm(37) indicated that these proteins were rich in beta-sheets, supporting our folding predictions. Whole-cell ELISA and FACS analyses unambiguously demonstrated surface expression of Scm in most E. faecium isolates. Strikingly, 11 of the 15 predicted MSCRAMMs clustered in four loci, each with a class C sortase gene; nine of these showed similarity to Enterococcus faecalis Ebp pilus subunits and also contained motifs essential for pilus assembly. Antibodies against one of the predicted major pilus proteins, Fms9 (redesignated EbpC(fm)), detected a 'ladder' pattern of high-molecular-mass protein bands in a Western blot analysis of cell surface extracts from E. faecium, suggesting that EbpC(fm) is polymerized into a pilus structure. Further analysis of the transcripts of the corresponding gene cluster indicated that fms1 (ebpA(fm)), fms5 (ebpB(fm)) and ebpC(fm) are co-transcribed, a result consistent with those for pilus-encoding gene clusters of other Gram-positive bacteria. All 15 genes occurred frequently in 30 clinically derived diverse E. faecium isolates tested. The common occurrence of MSCRAMM- and pilus-encoding genes and the presence of a second collagen-binding protein may have important implications for our understanding of this emerging pathogen.

Construction of improved temperature-sensitive and mobilizable vectors and their use for constructing mutations in the adhesin-encoding acm gene of poorly transformable clinical Enterococcus faecium strains.

Inactivation by allelic exchange in clinical isolates of the emerging nosocomial pathogen Enterococcus faecium has been hindered by lack of efficient tools, and, in this study, transformation of clinical isolates was found to be particularly problematic. For this reason, a vector for allelic replacement (pTEX5500ts) was constructed that includes (i) the pWV01-based gram-positive repAts replication region, which is known to confer a high degree of temperature intolerance, (ii) Escherichia coli oriR from pUC18, (iii) two extended multiple-cloning sites located upstream and downstream of one of the marker genes for efficient cloning of flanking regions for double-crossover mutagenesis, (iv) transcriptional terminator sites to terminate undesired readthrough, and (v) a synthetic extended promoter region containing the cat gene for allelic exchange and a high-level gentamicin resistance gene, aph(2'')-Id, to distinguish double-crossover recombination, both of which are functional in gram-positive and gram-negative backgrounds. To demonstrate the functionality of this vector, the vector was used to construct an acm (encoding an adhesin to collagen from E. faecium) deletion mutant of a poorly transformable multidrug-resistant E. faecium endocarditis isolate, TX0082. The acm-deleted strain, TX6051 (TX0082Deltaacm), was shown to lack Acm on its surface, which resulted in the abolishment of the collagen adherence phenotype observed in TX0082. A mobilizable derivative (pTEX5501ts) that contains oriT of Tn916 to facilitate conjugative transfer from the transformable E. faecalis strain JH2Sm::Tn916 to E. faecium was also constructed. Using this vector, the acm gene of a nonelectroporable E. faecium wound isolate was successfully interrupted. Thus, pTEX5500ts and its mobilizable derivative demonstrated their roles as important tools by helping to create the first reported allelic replacement in E. faecium; the constructed this acm deletion mutant will be useful for assessing the role of acm in E. faecium pathogenesis using animal models.

Further characterization of the epa gene cluster and Epa polysaccharides of Enterococcus faecalis.

We previously identified a gene cluster, epa (for enterocococcal polysaccharide antigen), involved in polysaccharide biosynthesis of Enterococcus faecalis and showed that disruption of epaB and epaE resulted in attenuation in translocation, biofilm formation, resistance to polymorphonuclear leukocyte (PMN) killing, and virulence in a mouse peritonitis model. Using five additional mutant disruptions in the 26-kb region between orfde2 and OG1RF_0163, we defined the epa locus as the area from epaA to epaR. Disruption of epaA, epaM, and epaN, like prior disruption of epaB and epaE, resulted in alteration in Epa polysaccharide content, more round cells versus oval cells with OG1RF, decreased biofilm formation, attenuation in a mouse peritonitis model, and resistance to lysis by the phage NPV-1 (known to lyse OG1RF), while mutants disrupted in orfde2 and OG1RF_163 (the epa locus flanking genes) behaved like OG1RF in those assays. Analysis of the purified Epa polysaccharide from OG1RF revealed the presence of rhamnose, glucose, galactose, GalNAc, and GlcNAc in this polysaccharide, while carbohydrate preparation from the epaB mutant did not contain rhamnose, suggesting that one or more of the glycosyl transferases encoded by the epaBCD operon are necessary to transfer rhamnose to the polysaccharide. In conclusion, the epa genes, uniformly present in E. faecalis strains and involved in biosynthesis of polysaccharide in OG1RF, are also important for OG1RF shape determination, biofilm formation, and NPV-1 replication/lysis, as well as for E. faecalis virulence in a mouse peritonitis model.

Application of in vivo induced antigen technology (IVIAT) to Bacillus anthracis.

In vivo induced antigen technology (IVIAT) is an immuno-screening technique that identifies bacterial antigens expressed during infection and not during standard in vitro culturing conditions. We applied IVIAT to Bacillus anthracis and identified PagA, seven members of a N-acetylmuramoyl-L-alanine amidase autolysin family, three P60 family lipoproteins, two transporters, spore cortex lytic protein SleB, a penicillin binding protein, a putative prophage holin, respiratory nitrate reductase NarG, and three proteins of unknown function. Using quantitative real-time PCR comparing RNA isolated from in vitro cultured B. anthracis to RNA isolated from BALB/c mice infected with virulent Ames strain B. anthracis, we confirmed induced expression in vivo for a subset of B. anthracis genes identified by IVIAT, including L-alanine amidases BA3767, BA4073, and amiA (pXO2-42); the bacteriophage holin gene BA4074; and pagA (pXO1-110). The exogenous addition of two purified putative autolysins identified by IVIAT, N-acetylmuramoyl-L-alanine amidases BA0485 and BA2446, to vegetative B. anthracis cell suspensions induced a species-specific change in bacterial morphology and reduction in viable bacterial cells. Many of the proteins identified in our screen are predicted to affect peptidoglycan re-modeling, and our results support significant cell wall structural remodeling activity during B. anthracis infection. Identification of L-alanine amidases with B. anthracis specificity may suggest new potential therapeutic targets.

Gene by BMI Interactions Influencing C-Reactive Protein Levels in European-Americans

C-Reactive Protein (CRP) is a biomarker indicating tissue damage, inflammation, and infection. High-sensitivity CRP (hsCRP) is an emerging biomarker often used to estimate an individual’s risk for future coronary heart disease (CHD). hsCRP levels falling below 1.00 mg/l indicate a low risk for developing CHD, levels ranging between 1.00 mg/l and 3.00 mg/l indicate an elevated risk, and levels exceeding 3.00 mg/l indicate high risk. Multiple Genome-Wide Association Studies (GWAS) have identified a number of genetic polymorphisms which influence CRP levels. SNPs implicated in such studies have been found in or near genes of interest including: CRP, APOE, APOC, IL-6, HNF1A, LEPR, and GCKR. A strong positive correlation has also been found to exist between CRP levels and BMI, a known risk factor for CHD and a state of chronic inflammation. We conducted a series of analyses designed to identify loci which interact with BMI to influence CRP levels in a subsample of European-Americans in the ARIC cohort. In a stratified GWA analysis, 15 genetic regions were identified as having significantly (p-value < 2.00*10-3) distinct effects on hsCRP levels between the two obesity strata: lean (18.50 kg/m2 < BMI < 24.99 kg/m2) and obese (BMI ≥ 30.00 kg/m2). A GWA analysis performed on all individuals combined (i.e. not a priori stratified for obesity status) with the inclusion of an additional parameter for BMI by gene interaction, identified 11 regions which interact with BMI to influence hsCRP levels. Two regions containing the genes GJA5 and GJA8 (on chromosome 1) and FBXO11 (on chromosome 2) were identified in both methods of analysis suggesting that these genes possibly interact with BMI to influence hsCRP levels. We speculate that atrial fibrillation (AF), age-related cataracts and the TGF-β pathway may be the biological processes influenced by the interaction of GJA5, GJA8 and FBXO11, respectively, with BMI to cause changes in hsCRP levels. Future studies should focus on the influence of gene x bmi interaction on AF, age-related cataracts and TGF-β.

THE EXPRESSION PROFILE OF A GWAS DETECTED GENE (NINJ2) IN THE BRAIN OF A MOUSE MODEL OF ISCHEMIC STROKE

Stroke is the third leading cause of death and a major debilitating disease in the United States. Multiple factors, including genetic factors, contribute to the development of the disease. Genome-wide association studies (GWAS) have contributed to the identification of genetic loci influencing risk for complex diseases, such as stroke. In 2010, a GWAS of incident stroke was performed in four large prospective cohorts from the USA and Europe and identified an association of two Single Nucleotide Polymorphisms (SNPs) on chromosome 12p13 with a greater risk of ischemic stroke in individuals of European and African-American ancestry. These SNPs are located 11 Kb upstream of the nerve injury-induced gene 2, Ninjurin2 (NINJ2), suggesting that this gene may be involved in stroke pathogenesis. NINJ2 is a cell adhesion molecule induced in the distal glial cells from a sciatic-nerve injury at 7-days after injury. In an effort to ascribe a possible role of NINJ2 in stroke, we have assessed changes in the level of gene and protein expression of NINJ2 following a time-course from a transiently induced middle cerebral artery ischemic stroke in mice brains. We report an increase in the gene expression of NINJ2 in the ischemic and peri-infarct (ipsilateral) cortical tissues at 7 and 14-days after stroke. We also report an increase in the protein expression of NINJ2 in the cortex of both the ipsilateral and contralateral cortical tissues at the same time-points. We conclude that the expression of NINJ2 is regulated by an ischemic stroke in the cortex and is consistent with NINJ2 being involved in the recovery time-points of stroke.

Structure-function relation in retinoid regulated gene expression

Retinoids are known to inhibit proliferation of and induce terminal differentiation of many normal and transformed cells. It has been postulated that retinoids exert their effect by altering gene expression. HL-60 cells and macrophages both respond to retinoic acid action by the rapid induction of the enzyme tissue transglutaminase. The induction has been shown to be due to increased transcription of the transglutaminase gene. The first part of the dissertation studied the structure-function relationship of retinoid-regulated transglutaminase induction, differentiation and proliferation in HL-60 cells using retinoid analogs. The results indicated strict structural constraints and a strong structure-function correlation between transglutaminase induction and differentiation; those retinoids that induced transglutaminase also induced differentiation, those analogs that did not induce transglutaminase could not induce differentiation. The ability of the retinoids to induce transglutaminase in HL-60 cells was paralleled in macrophages. However, the antiproliferative effect of the retinoids displayed less stringent structural constraints than their differentiation- and transglutaminase-inducing properties. Specifically all the retinoids were able to inhibit proliferation to varying extents. It is concluded that the induction of transglutaminase and of differentiation by retinoids is mediated by receptors. While receptor mediation cannot be entirely ruled out, with the current data no definitive statement can be made about the antiproliferative activity of retinoids. Also, the concordance in the ability of the retinoids to induce transglutaminase and the ability to induce differentiation of HL-60 cells suggests that the former is an early response of the cells to retinoids and differentiation a later consequence on the same pathway. Using the induction of transglutaminase as an index of the direct, or primary, effect of retinoids on gene expression, the second part of the dissertation investigates, by 2D gel electrophoresis, the alteration in the rates of synthesis of other proteins in macrophages and HL-60 cells in response to short incubations with retinoic acid. Any changes in parallel with transglutaminase were taken to indicate proteins directly under the control of retinoic acid. It is concluded that retinoic acid regulates the expression of a circumscribed set of genes in a cell-specific manner. The results support the hypothesis that retinoids exert their multiple effects on myeloid cells, in part, by receptor-mediated alternations in gene expression. ^