Richard L. Amdur

Convergent functional genomics of schizophrenia: from comprehensive understanding to genetic risk prediction

We have used a translational convergent functional genomics (CFG) approach to identify and prioritize genes involved in schizophrenia, by gene-level integration of genome-wide association study data with other genetic and gene expression studies in humans and animal models. Using this polyevidence scoring and pathway analyses, we identify top genes (DISC1, TCF4, MBP, MOBP, NCAM1, NRCAM, NDUFV2, RAB18, as well as ADCYAP1, BDNF, CNR1, COMT, DRD2, DTNBP1, GAD1, GRIA1, GRIN2B, HTR2A, NRG1, RELN, SNAP-25, TNIK), brain development, myelination, cell adhesion, glutamate receptor signaling, G-protein–coupled receptor signaling and cAMP-mediated signaling as key to pathophysiology and as targets for therapeutic intervention. Overall, the data are consistent with a model of disrupted connectivity in schizophrenia, resulting from the effects of neurodevelopmental environmental stress on a background of genetic vulnerability. In addition, we show how the top candidate genes identified by CFG can be used to generate a genetic risk prediction score (GRPS) to aid schizophrenia diagnostics, with predictive ability in independent cohorts. The GRPS also differentiates classic age of onset schizophrenia from early onset and late-onset disease. We also show, in three independent cohorts, two European American and one African American, increasing overlap, reproducibility and consistency of findings from single-nucleotide polymorphisms to genes, then genes prioritized by CFG, and ultimately at the level of biological pathways and mechanisms. Finally, we compared our top candidate genes for schizophrenia from this analysis with top candidate genes for bipolar disorder and anxiety disorders from previous CFG analyses conducted by us, as well as findings from the fields of autism and Alzheimer. Overall, our work maps the genomic and biological landscape for schizophrenia, providing leads towards a better understanding of illness, diagnostics and therapeutics. It also reveals the significant genetic overlap with other major psychiatric disorder domains, suggesting the need for improved nosology.

Outcomes following diagnosis of acute renal failure in U.S. veterans: focus on acute tubular necrosis

When patients develop acute kidney injury, a small fraction of them will develop end-stage renal disease later. The severity of renal impairment in the remaining patients is uncertain because studies have not carefully examined renal function over time or the precise timing of entry into a late stage of chronic kidney disease. To determine these factors, we used a United States Department of Veterans Affairs database to ascertain long-term renal function in 113,272 patients. Of these, 44,377 had established chronic kidney disease and were analyzed separately. A cohort of 63,491 patients was hospitalized for acute myocardial infarction or pneumonia and designated as controls. The remaining 5,404 patients had diagnostic codes indicating acute renal failure or acute tubular necrosis. Serum creatinine, estimated glomerular filtration rates, and dates of death over a 75-month period were followed. Renal function deteriorated over time in all groups, but with significantly greater severity in those who had acute renal failure and acute tubular necrosis compared to controls. Patients with acute kidney injury, especially those with acute tubular necrosis, were more likely than controls to enter stage 4 chronic kidney disease, but this entry time was similar to that of patients who initially had chronic kidney disease. The risk of death was elevated in those with acute kidney injury and chronic kidney disease compared to controls after accounting for covariates. We found that patients who had an episode of acute tubular necrosis were at high risk for the development of stage 4 disease and had a reduced survival time when compared to control patients.

Polygenic dissection of diagnosis and clinical dimensions of bipolar disorder and schizophrenia

Bipolar disorder and schizophrenia are two often severe disorders with high heritabilities. Recent studies have demonstrated a large overlap of genetic risk loci between these disorders but diagnostic and molecular distinctions still remain. Here, we perform a combined genome-wide association study (GWAS) of 19 779 bipolar disorder (BP) and schizophrenia (SCZ) cases versus 19 423 controls, in addition to a direct comparison GWAS of 7129 SCZ cases versus 9252 BP cases. In our case–control analysis, we identify five previously identified regions reaching genome-wide significance (CACNA1C, IFI44L, MHC, TRANK1 and MAD1L1) and a novel locus near PIK3C2A. We create a polygenic risk score that is significantly different between BP and SCZ and show a significant correlation between a BP polygenic risk score and the clinical dimension of mania in SCZ patients. Our results indicate that first, combining diseases with similar genetic risk profiles improves power to detect shared risk loci and second, that future direct comparisons of BP and SCZ are likely to identify loci with significant differential effects. Identifying these loci should aid in the fundamental understanding of how these diseases differ biologically. These findings also indicate that combining clinical symptom dimensions and polygenic signatures could provide additional information that may someday be used clinically.

Association of periodontitis with rheumatoid arthritis: a pilot study.

BACKGROUND Similarities exist in the epidemiology and immunopathogenesis of periodontitis and rheumatoid arthritis (RA), but the associations between their respective disease activities and severities are less well documented. We evaluated the prevalence and severity of periodontitis in United States (U.S.) veterans with RA and their relationship to RA disease activity and severity. METHODS Patients with RA from an outpatient rheumatology clinic were eligible, and patients with osteoarthritis (OA) served as controls. Dentists, masked to the rheumatologic diagnoses, performed periodontal probing and examined dental panoramic radiographs to assess the presence and severity of periodontitis. Associations of periodontitis with RA were examined using multivariate regression, whereas the association of periodontitis with disease-severity measures in RA was examined using the chi(2) test. RESULTS Sixty-nine patients with RA (57 males and 12 females) and 35 patients with OA (30 males and five females) were studied. Moderate to severe periodontitis was more prevalent in patients with RA (51%) than controls (26%) (P = 0.03), an association independent of age, race, smoking, diabetes mellitus, and gender. Patients with RA who were seropositive for rheumatoid factor (RF) were more likely to have moderate to severe periodontitis (59%) than patients who were RF negative (15%) (P = 0.02). Likewise, patients with RA who were positive for the anti-cyclic citrullinated peptide (CCP) antibodies were more likely to have moderate to severe periodontitis (56%) than patients who were anti-CCP negative (22%) (P = 0.01). There were no associations of periodontitis status with other measures of RA disease activity or severity. CONCLUSIONS In a cohort of U.S. veterans, periodontitis was more common and severe in patients with RA compared to patients with OA. Although unrelated to disease activity, the presence of periodontitis in patients with RA was associated with seropositivity for RF and the anti-CCP antibody, which was highly relevant given the associations of these autoantibodies with poor outcomes and disease pathogenesis in RA.

Association between AKI and Long-Term Renal and Cardiovascular Outcomes in United States Veterans

BACKGROUND AND OBJECTIVES AKI is associated with major adverse kidney events (MAKE): death, new dialysis, and worsened renal function. CKD (arising from worsened renal function) is associated with a higher risk of major adverse cardiac events (MACE): myocardial infarction (MI), stroke, and heart failure. Therefore, the study hypothesis was that veterans who develop AKI during hospitalization for an MI would be at higher risk of subsequent MACE and MAKE. DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS Patients in the Veterans Affairs (VA) database who had a discharge diagnosis with International Classification of Diseases, Ninth Revision, code of 584.xx (AKI) or 410.xx (MI) and were admitted to a VA facility from October 1999 through December 2005 were selected for analysis. Three groups of patients were created on the basis of the index admission diagnosis and serum creatinine values: AKI, MI, or MI with AKI. Patients with mean baseline estimated GFR<45 ml/min per 1.73 m(2) were excluded. The primary outcomes assessed were mortality, MAKE, and MACE during the study period (maximum of 6 years). The combination of MAKE and MACE-major adverse renocardiovascular events (MARCE)-was also assessed. RESULTS A total of 36,980 patients were available for analysis. Mean age±SD was 66.8±11.4 years. The most deaths occurred in the MI+AKI group (57.5%), and the fewest (32.3%) occurred in patients with an uncomplicated MI admission. In both the unadjusted and adjusted time-to-event analyses, patients with AKI and AKI+MI had worse MARCE outcomes than those who had MI alone (adjusted hazard ratios, 1.37 [95% confidence interval, 1.32 to 1.42] and 1.92 [1.86 to 1.99], respectively). CONCLUSIONS Veterans who develop AKI in the setting of MI have worse long-term outcomes than those with AKI or MI alone. Veterans with AKI alone have worse outcomes than those diagnosed with an MI in the absence of AKI.

Schizophrenia Gene Networks and Pathways and Their Applications for Novel Candidate Gene Selection

Background Schizophrenia (SZ) is a heritable, complex mental disorder. We have seen limited success in finding causal genes for schizophrenia from numerous conventional studies. Protein interaction network and pathway-based analysis may provide us an alternative and effective approach to investigating the molecular mechanisms of schizophrenia. Methodology/Principal Findings We selected a list of schizophrenia candidate genes (SZGenes) using a multi-dimensional evidence-based approach. The global network properties of proteins encoded by these SZGenes were explored in the context of the human protein interactome while local network properties were investigated by comparing SZ-specific and cancer-specific networks that were extracted from the human interactome. Relative to cancer genes, we observed that SZGenes tend to have an intermediate degree and an intermediate efficiency on a perturbation spreading throughout the human interactome. This suggested that schizophrenia might have different pathological mechanisms from cancer even though both are complex diseases. We conducted pathway analysis using Ingenuity System and constructed the first schizophrenia molecular network (SMN) based on protein interaction networks, pathways and literature survey. We identified 24 pathways overrepresented in SZGenes and examined their interactions and crosstalk. We observed that these pathways were related to neurodevelopment, immune system, and retinoic X receptor (RXR). Our examination of SMN revealed that schizophrenia is a dynamic process caused by dysregulation of the multiple pathways. Finally, we applied the network/pathway approach to identify novel candidate genes, some of which could be verified by experiments. Conclusions/Significance This study provides the first comprehensive review of the network and pathway characteristics of schizophrenia candidate genes. Our preliminary results suggest that this systems biology approach might prove promising for selection of candidate genes for complex diseases. Our findings have important implications for the molecular mechanisms for schizophrenia and, potentially, other psychiatric disorders.

Gains in Upper Extremity Function After Stroke via Recovery or Compensation: Potential Differential Effects on Amount of Real-World Limb Use

Abstract In terms of integration of the paretic upper extremity in activities of daily living (ADLs), outcome is poor after stroke. Furthermore, amount of real-world arm use appears only weakly correlated with laboratory motor function scales. Therefore, amount of arm use may depend critically on the location, extent, and type of functional gains, which can be quantified with comprehensive kinematic and EMG analysis of ADL performance. Gains in upper extremity function can occur via compensation or recovery of premorbid movement and EMG patterns, and traditional treatment approaches encourage adoption of compensatory strategies early in the postacute period that can inhibit potential recovery. A new treatment approach called Accelerated Skill Acquisition Program (ASAP) focuses on impairment reduction coupled with repetitive, task-specific training of the paretic arm during ADLs. We present pilot data that show recovery in subjects who received the ASAP, while a usual care control subject showed increased use of compensation over the same period. Finally, we discuss the advantages of data reduction methods such as principal components analysis, confirmatory factor analysis, and structural equation modeling, which can potentially distill large kinematic and EMG data sets into the key latent variables that predict amount of real-world use.

Hyperglycemia is associated with increased risk of morbidity and mortality after colectomy for cancer

BACKGROUND The relationship of hyperglycemia to general surgery outcomes is not well-understood. We studied the association of operative day and postoperative day 1 (POD1) blood glucose (BG) with outcomes after open colectomy for cancer. STUDY DESIGN We retrospectively analyzed the 2000-2005 Veterans Affairs Surgical Quality Improvement Program database, linked with Veterans Affairs Decision Support System BG values. Median BG was categorized as hypoglycemic (<80 mg/dL); normoglycemic (BG 80-120 mg/dL); or mildly (BG 121-160 mg/dL), moderately (BG 161-200 mg/dL), or severely (BG >200 mg/dL) hyperglycemic. The relationship of BG to postoperative outcomes was assessed with multivariable logistic regression. RESULTS We identified 9,638 colectomies. We excluded 511 procedures for emergency status or preoperative coma, mechanical ventilation, or sepsis. After excluding patients without recorded BG, we analyzed operative day and POD1 BG in 7,576 and 5,773 procedures, respectively. On multivariable analysis, operative day moderate hyperglycemia was associated with surgical site infection (odds ratio = 1.44; 95% CI, 1.10-1.87). POD1 severe hyperglycemia was associated with cardiac arrest (odds ratio = 2.31; 95% CI, 1.08-4.98) and death (odds ratio = 1.97; 95% CI, 1.23-3.15). POD1 mild (odds ratio = 2.20; 95% CI, 1.05-4.60), moderate (odds ratio = 3.44; 95% CI, 1.51-7.84), and severe (odds ratio = 3.94; 95% CI, 1.64-9.58) hyperglycemia and hypoglycemia (odds ratio = 6.74; 95% CI, 1.75-25.97) were associated with myocardial infarction. Associations were similar in diabetic and nondiabetic patients. CONCLUSIONS Even mild hyperglycemia was associated with adverse outcomes after colectomy, suggesting that a perioperative BG target of 80 to 120 mg/dL, although avoiding hypoglycemia, might be appropriate. Randomized clinical trials are needed to confirm these findings.

Association Study of 167 Candidate Genes for Schizophrenia Selected by a Multi-Domain Evidence-Based Prioritization Algorithm and Neurodevelopmental Hypothesis

Integrating evidence from multiple domains is useful in prioritizing disease candidate genes for subsequent testing. We ranked all known human genes (n = 3819) under linkage peaks in the Irish Study of High-Density Schizophrenia Families using three different evidence domains: 1) a meta-analysis of microarray gene expression results using the Stanley Brain collection, 2) a schizophrenia protein-protein interaction network, and 3) a systematic literature search. Each gene was assigned a domain-specific p-value and ranked after evaluating the evidence within each domain. For comparison to this ranking process, a large-scale candidate gene hypothesis was also tested by including genes with Gene Ontology terms related to neurodevelopment. Subsequently, genotypes of 3725 SNPs in 167 genes from a custom Illumina iSelect array were used to evaluate the top ranked vs. hypothesis selected genes. Seventy-three genes were both highly ranked and involved in neurodevelopment (category 1) while 42 and 52 genes were exclusive to neurodevelopment (category 2) or highly ranked (category 3), respectively. The most significant associations were observed in genes PRKG1, PRKCE, and CNTN4 but no individual SNPs were significant after correction for multiple testing. Comparison of the approaches showed an excess of significant tests using the hypothesis-driven neurodevelopment category. Random selection of similar sized genes from two independent genome-wide association studies (GWAS) of schizophrenia showed the excess was unlikely by chance. In a further meta-analysis of three GWAS datasets, four candidate SNPs reached nominal significance. Although gene ranking using integrated sources of prior information did not enrich for significant results in the current experiment, gene selection using an a priori hypothesis (neurodevelopment) was superior to random selection. As such, further development of gene ranking strategies using more carefully selected sources of information is warranted.

Genome-wide association study of clinical dimensions of schizophrenia: polygenic effect on disorganized symptoms

OBJECTIVE Multiple sources of evidence suggest that genetic factors influence variation in clinical features of schizophrenia. The authors present the first genome-wide association study (GWAS) of dimensional symptom scores among individuals with schizophrenia. METHOD Based on the Lifetime Dimensions of Psychosis Scale ratings of 2,454 case subjects of European ancestry from the Molecular Genetics of Schizophrenia (MGS) sample, three symptom factors (positive, negative/disorganized, and mood) were identified with exploratory factor analysis. Quantitative scores for each factor from a confirmatory factor analysis were analyzed for association with 696,491 single-nucleotide polymorphisms (SNPs) using linear regression, with correction for age, sex, clinical site, and ancestry. Polygenic score analysis was carried out to determine whether case and comparison subjects in 16 Psychiatric GWAS Consortium (PGC) schizophrenia samples (excluding MGS samples) differed in scores computed by weighting their genotypes by MGS association test results for each symptom factor. RESULTS No genome-wide significant associations were observed between SNPs and factor scores. Most of the SNPs producing the strongest evidence for association were in or near genes involved in neurodevelopment, neuroprotection, or neurotransmission, including genes playing a role in Mendelian CNS diseases, but no statistically significant effect was observed for any defined gene pathway. Finally, polygenic scores based on MGS GWAS results for the negative/disorganized factor were significantly different between case and comparison subjects in the PGC data set; for MGS subjects, negative/disorganized factor scores were correlated with polygenic scores generated using case-control GWAS results from the other PGC samples. CONCLUSIONS The polygenic signal that has been observed in cross-sample analyses of schizophrenia GWAS data sets could be in part related to genetic effects on negative and disorganized symptoms (i.e., core features of chronic schizophrenia).