- Email: [email protected]
TRANSCRIPTIONAL SIGNATURES FROM BLOOD AND BRAIN
S54
Abstracts of the 3rd Biennial Schizophrenia International Research Conference / Schizophrenia Research 136, Supplement 1 (2012) S1–S375
biomarkers may have advantages or disadvantages compared to imaging or electrophysiological biomarkers (ease of use versus proximity to brain). This symposium will provide an update on which blood biomarkers are showing the most promise and consider how combining transcriptomics and proteomics may be a fruitful next step in biomarker development.
BRAIN DERIVED NEUROTROPHIC FACTOR: ILLUSTRATIVE OF A PUTATIVE BIOMARKER FOR SCHIZOPHRENIA Peter F. Buckley Georgia Health Sciences University, Augusta, Georgia, USA The identification and characterization of putative biomarkers as objective indicators of illness cause, course, and/or treatment response represents a major challenge for schizophrenia research, enduring from early “Pink Spot” methylation studies to current investigation of potential biomarker candidates – inter alia – Brain Derived Neurotrophic Factor (BDNF). Convergent data from multiple, diverse neuroscience and clinical studies suggest that BDNF might serve as a novel and heuristic potential biomarker of both molecular neurobiology and disease trajectory of schizophrenia. A metaanalysis (Green at al, 2011) suggests that – inspite of heterogeneity – reductions in BDNF are observed in both first and multiepisode schizophrenia with minimal drug effect while our initial analyses from a longitudinal study suggests that low BDNF is associated with later relapse. This presentation, drawing from extant literature as well as data from our collaborative studies, will illuminate basic principles of biomarker discrimination - including requisite specificity and sensitivity expectations through a critical evaluation of BDNF findings to date as well as pointing to gaps in information that would support or negate BDNF’s biomarker portfolio.
DISEASE BIOMARKERS FOR SCHIZOPHRENIA - FROM LABORATORY TO PATIENT BEDSIDE Sabine Bahn University of Cambridge, Cambridge, United Kingdom Schizophrenia is a multifaceted neuropsychiatric disorder. Its onset is the result of complex interactions between genetic, developmental and environmental factors. It almost certainly presents a heterogeneous group of aetiologies which may not be reflected in the symptomatic/clinical presentation of patients. Therefore, a better molecular understanding of the disease onset and progression is urgently needed. Multi-omics profiling approaches can be employed to investigate large numbers of patient and control samples in a single experiment. These large scale experiments are required to identify disease intrinsic molecular signatures as well as patient subgroups with potentially distinct biochemical pathways underpinning their symptoms. I will present results from our biomarker discovery studies. We have identified a number of highly significant peptides and proteins that distinguish first-onset paranoid schizophrenia patients from healthy controls. Our findings suggest alterations in glucoregulatory processes in CSF of drug-naïve patients with first-onset schizophrenia. Short-term treatment with atypical antipsychotic medication resulted in a normalization of the CSF disease signature in half the patients well before a clinical improvement would be expected. Furthermore, our results suggest that the initiation of antipsychotic treatment during a first psychotic episode may influence treatment response and/or outcome. More recently, we have identified a candidate biomarker panel in patient serum, specifically up-or down-regulated in drug naive, first onset schizophrenia patients compared to healthy controls using high throughput proteomic profiling and multiplexed immunoassay profiling technology. A panel of 51 markers was found to yield an average sensitivity and specificity of >85% across five clinical centres. Abnormalities remained significant after adjustment for all recorded baseline characteristics. The panel has now been developed into a test which can be used to help confirm thediagnosis of schizophrenia.
TRANSCRIPTIONAL SIGNATURES FROM BLOOD AND BRAIN Marquis P. Vawter University of California, Irvine, California, USA The consistency of peripheral gene expression data and the overlap with brain expression have not been evaluated in biomarker discovery, nor has it been reported in tissues from the same subjects on a genome wide transcript level. The effects of processing whole blood, transformation, and passaged cell lines on gene expression profiling have not been studied in healthy subjects. Gene expression in cell lines derived from antipsychotic treatment responders versus non-responders with schizophrenia have never been evaluated. The gene expression of brain and PBMCs was compared in postmortem human and rat tissues. The effects of blood processing and Epstein-Barr virus (EBV) transformed lymphocytes were compared to passaged lymphoblastic cell lines (LCLs), whole blood from Tempus™ (ABI) tubes, and Ficoll extracted peripheral blood mononuclear cells (PBMCs). LCLs were obtained from 30 first episode patients in the initial phase of treatment onset and expression profiles were compared for responders versus non-responders. Profiling studies were conducted using an exon array mixed model analysis, discriminant analysis was used to identify predictors of response. Postmortem subjects’ brain and PBMC profiles showed similar expression levels for 22% - 90% of Refseq transcripts depending on the criteria stringency. There were 6,813 transcripts differentially expressed between blood preparations using Bonferroni correction. For treatment responders compared to non-responders, there were 22 transcripts with statistically significant treatment responsex probeset effects, 245 transcripts with statistically significant medication × probeset effects and 210 transcripts with statistically significant treatment response × medication × probeset. Discriminant analysis identified expression of probesets in two genes, NETO1 and CPS1, as predictors of antipsychotic response. Pathway analysis revealed an over-representation of treatment response genes related to axonal guidance signaling. EBV infection and blood preparation methods are critical variables across studies of peripheral biomarker expression. These initial estimates of coexpressed genes in blood and brain indicate potential usefulness of peripheral transcript expression in the biomarker discovery process. The over-representation of treatment response genes related to axonal guidance signaling involves several genes previously implicated in schizophrenia. These results provide insight into possible individual differences that predict antipsychotic response in first episode patients with schizophrenia. The use of cell lines derived from peripheral blood of subjects with schizophrenia to identify antipsychotic response related gene expression alterations may be useful in predicting response in a larger study.
MICRORNA BIOMARKERS OF SCHIZOPHRENIA IN BLOOD Murray Cairns Schizophrenia Research Institute, Sydney, New South Wales, Australia MicroRNA are small non-coding RNA that play a significant role as the specificity factors or guide strands in post-transcriptional gene silencing. These molecules are enriched in the brain and are emerging as key regulators of brain development and neural function. MicroRNA genes have also been shown to be genetically associated with schizophrenia and their cortical expression has been shown to be altered in postmortem brain tissue. While these changes were seen in the central nervous system, it is plausible that schizophrenia-associated miRNA signatures also exist in non-neural tissue, providing the basis of an accessible biomarker in living subjects. To this end we investigated miRNA expression in peripheral blood mononuclear cells (PBMCs) in patients with schizophrenia and a non-psychiatric comparison group using a commercial microarray platform. This analysis identified 33 downregulated miRNA after correction for multiple testing (FDR=0%), and a further 50 downregulated molecules with an FDR<5%. A large sub-group consisting of 17 miRNAs were transcribed from a single imprinted locus at the maternally expressed DLK1-DIO3 region on chromosome 14q32. The expression of this large cluster is known to be brain enriched and the encoded microRNA are predicted to be involved in neural and immune system function. We believe that these molecules and their disease-associated expression signatures could have application as biomarkers for schizophrenia and related sub-phenotypes.
Abstracts of the 3rd Biennial Schizophrenia International Research Conference / Schizophrenia Research 136, Supplement 1 (2012) S1–S375
biomarkers may have advantages or disadvantages compared to imaging or electrophysiological biomarkers (ease of use versus proximity to brain). This symposium will provide an update on which blood biomarkers are showing the most promise and consider how combining transcriptomics and proteomics may be a fruitful next step in biomarker development.
BRAIN DERIVED NEUROTROPHIC FACTOR: ILLUSTRATIVE OF A PUTATIVE BIOMARKER FOR SCHIZOPHRENIA Peter F. Buckley Georgia Health Sciences University, Augusta, Georgia, USA The identification and characterization of putative biomarkers as objective indicators of illness cause, course, and/or treatment response represents a major challenge for schizophrenia research, enduring from early “Pink Spot” methylation studies to current investigation of potential biomarker candidates – inter alia – Brain Derived Neurotrophic Factor (BDNF). Convergent data from multiple, diverse neuroscience and clinical studies suggest that BDNF might serve as a novel and heuristic potential biomarker of both molecular neurobiology and disease trajectory of schizophrenia. A metaanalysis (Green at al, 2011) suggests that – inspite of heterogeneity – reductions in BDNF are observed in both first and multiepisode schizophrenia with minimal drug effect while our initial analyses from a longitudinal study suggests that low BDNF is associated with later relapse. This presentation, drawing from extant literature as well as data from our collaborative studies, will illuminate basic principles of biomarker discrimination - including requisite specificity and sensitivity expectations through a critical evaluation of BDNF findings to date as well as pointing to gaps in information that would support or negate BDNF’s biomarker portfolio.
DISEASE BIOMARKERS FOR SCHIZOPHRENIA - FROM LABORATORY TO PATIENT BEDSIDE Sabine Bahn University of Cambridge, Cambridge, United Kingdom Schizophrenia is a multifaceted neuropsychiatric disorder. Its onset is the result of complex interactions between genetic, developmental and environmental factors. It almost certainly presents a heterogeneous group of aetiologies which may not be reflected in the symptomatic/clinical presentation of patients. Therefore, a better molecular understanding of the disease onset and progression is urgently needed. Multi-omics profiling approaches can be employed to investigate large numbers of patient and control samples in a single experiment. These large scale experiments are required to identify disease intrinsic molecular signatures as well as patient subgroups with potentially distinct biochemical pathways underpinning their symptoms. I will present results from our biomarker discovery studies. We have identified a number of highly significant peptides and proteins that distinguish first-onset paranoid schizophrenia patients from healthy controls. Our findings suggest alterations in glucoregulatory processes in CSF of drug-naïve patients with first-onset schizophrenia. Short-term treatment with atypical antipsychotic medication resulted in a normalization of the CSF disease signature in half the patients well before a clinical improvement would be expected. Furthermore, our results suggest that the initiation of antipsychotic treatment during a first psychotic episode may influence treatment response and/or outcome. More recently, we have identified a candidate biomarker panel in patient serum, specifically up-or down-regulated in drug naive, first onset schizophrenia patients compared to healthy controls using high throughput proteomic profiling and multiplexed immunoassay profiling technology. A panel of 51 markers was found to yield an average sensitivity and specificity of >85% across five clinical centres. Abnormalities remained significant after adjustment for all recorded baseline characteristics. The panel has now been developed into a test which can be used to help confirm thediagnosis of schizophrenia.
TRANSCRIPTIONAL SIGNATURES FROM BLOOD AND BRAIN Marquis P. Vawter University of California, Irvine, California, USA The consistency of peripheral gene expression data and the overlap with brain expression have not been evaluated in biomarker discovery, nor has it been reported in tissues from the same subjects on a genome wide transcript level. The effects of processing whole blood, transformation, and passaged cell lines on gene expression profiling have not been studied in healthy subjects. Gene expression in cell lines derived from antipsychotic treatment responders versus non-responders with schizophrenia have never been evaluated. The gene expression of brain and PBMCs was compared in postmortem human and rat tissues. The effects of blood processing and Epstein-Barr virus (EBV) transformed lymphocytes were compared to passaged lymphoblastic cell lines (LCLs), whole blood from Tempus™ (ABI) tubes, and Ficoll extracted peripheral blood mononuclear cells (PBMCs). LCLs were obtained from 30 first episode patients in the initial phase of treatment onset and expression profiles were compared for responders versus non-responders. Profiling studies were conducted using an exon array mixed model analysis, discriminant analysis was used to identify predictors of response. Postmortem subjects’ brain and PBMC profiles showed similar expression levels for 22% - 90% of Refseq transcripts depending on the criteria stringency. There were 6,813 transcripts differentially expressed between blood preparations using Bonferroni correction. For treatment responders compared to non-responders, there were 22 transcripts with statistically significant treatment responsex probeset effects, 245 transcripts with statistically significant medication × probeset effects and 210 transcripts with statistically significant treatment response × medication × probeset. Discriminant analysis identified expression of probesets in two genes, NETO1 and CPS1, as predictors of antipsychotic response. Pathway analysis revealed an over-representation of treatment response genes related to axonal guidance signaling. EBV infection and blood preparation methods are critical variables across studies of peripheral biomarker expression. These initial estimates of coexpressed genes in blood and brain indicate potential usefulness of peripheral transcript expression in the biomarker discovery process. The over-representation of treatment response genes related to axonal guidance signaling involves several genes previously implicated in schizophrenia. These results provide insight into possible individual differences that predict antipsychotic response in first episode patients with schizophrenia. The use of cell lines derived from peripheral blood of subjects with schizophrenia to identify antipsychotic response related gene expression alterations may be useful in predicting response in a larger study.
MICRORNA BIOMARKERS OF SCHIZOPHRENIA IN BLOOD Murray Cairns Schizophrenia Research Institute, Sydney, New South Wales, Australia MicroRNA are small non-coding RNA that play a significant role as the specificity factors or guide strands in post-transcriptional gene silencing. These molecules are enriched in the brain and are emerging as key regulators of brain development and neural function. MicroRNA genes have also been shown to be genetically associated with schizophrenia and their cortical expression has been shown to be altered in postmortem brain tissue. While these changes were seen in the central nervous system, it is plausible that schizophrenia-associated miRNA signatures also exist in non-neural tissue, providing the basis of an accessible biomarker in living subjects. To this end we investigated miRNA expression in peripheral blood mononuclear cells (PBMCs) in patients with schizophrenia and a non-psychiatric comparison group using a commercial microarray platform. This analysis identified 33 downregulated miRNA after correction for multiple testing (FDR=0%), and a further 50 downregulated molecules with an FDR<5%. A large sub-group consisting of 17 miRNAs were transcribed from a single imprinted locus at the maternally expressed DLK1-DIO3 region on chromosome 14q32. The expression of this large cluster is known to be brain enriched and the encoded microRNA are predicted to be involved in neural and immune system function. We believe that these molecules and their disease-associated expression signatures could have application as biomarkers for schizophrenia and related sub-phenotypes.