Ming T. Tsuang - US grants

Considerable evidence has accumulated during the past thirty years which suggests that "perinatal complications", broadly defined, may be considered as one of the major known risk factors for a variety of psychiatric and neurologic disorders. Further investigation is required to support this claim, to further clarify the definition of risk and to quantify the magnitude of risk for specific disorders. Research in this area could result in the early identification of individuals at risk for subsequent psychiatric disorder and the implementation of preventive measures. We are proposing to complete a twenty-five year prospective study of the association between perinatal complications and psychiatric disorders. We will select a sample of 500 births with perinatal complications (prolonged fetal hypoxia, non-hypoxic complications and prematurity) and a matched control sample of 500 normal births from the Providence cohort of the National Collaborative Perinatal Project (NCPP) who were assessed at age seven. The mothers of these subjects were monitored systematically during pregnancy, and the children were evaluated at birth and throughout the first seven years of life on a wide range of neurological, cognitive, developmental and behavioral measures. Using follow-up methods tested and refined through two years of pilot work, we will contact these subjects (now ages 20-26) and administer the Diagnostic Interview Schedule, the Wechsler Adult Intelligence Scale and other instruments designed to gather psychiatric, psychosocial and neuropsychological outcome data. Quantitative estimates of the risk for various forms of psychiatric disorder will be obtained for a number of previously suggested risk conditions, including "perinatal complications", "prolonged fetal hypoxia", "non-hypoxic complications", "prematurity" and a range of additional variables contained in the baseline NCPP data set. The primary psychiatric disorders we plan to study are alcohol abuse and dependence, drug abuse and dependence, antisocial personality disorder, affective disorders, and minimal brain dysfunction. In addition to the quantification of the magnitude of risk associated with these risk conditions, our analyses will investigate potential etiologic mechanisms which have been proposed to account for the association between perinatal events and psychiatric outcomes.

There is some descriptive evidence demonstrating that there are gender differences in schizophrenia regarding age of onset, premorbid history, family history, psychopathology, and course. However, much of the literature on schizophrenia assumes that the illness is similar in men and women. The validity of the effect of gender on schizophrenia has important consequences for understaanding the nature of the illness. That is, if there are significant differences in the family history, expression, and course of the illness, perhaps men and women express different subtypes of schizophrenia. This has been proposed by some who argue that schizophrenic men may be at higher risk for an amotivational (negative) syndrome than schizophrenic women. The proposed study will test this hypothesis by examining four of the validity criteria of Robins and Guze (1970): the clinical description, family history, and course of the disorder, and the specificity of the gender effects on schizophrenia compared to major affective disorder. Hypotheses will be tested using the data from the well-known retrospective cohort studies, the Iowa 500 and Non-500 studies. Subjects have been rediagnosed by expert diagnosticians using DSM-III criteria, which will allow for stringent tests of the gender hypotheses. This data set provides a sample size with the power to test for the validity of gender effects, has a long observational period, and has important variables necessary to test for gender effects such as age of onset, premorbid history, family history, symptomatology, and social functioning, and includes psychiatric and normal controls. Finally, this data set allows for the opportunity to test models that begin to explain the effect of gender on schizophrenia, which have important implications for understanding the nature of the illness. The hypotheses will be tested using powerful data analytic techniques that have been recently developed for use with the type of data available for this proposal. They include survival analysis to estimate age of onset distributions and family morbidity risk, maximum likelihood procedures to test competing models of familial transmission, latent class analysis to test for gender differences in psychopathology, and log linear analyses to test models to explain the effect of gender on schizophrenia. This study will provide a valuable contribution to understanding the heterogeneity of schizophrenia, has methodological consequences for estimating the morbidity risk, and costs considerably low given the benefits.

Genetic and neuropsychological studies have each made important contributions to the understanding of schizophrenia. Family, twin and adoption studies provide strong support for a genetic component in the etiology and schizophrenia, and neurodiagnostic approaches strongly indicate the presence of brain dysfunction in many schizophrenics. Moreover, variability in the occurrence of mental illness in the relatives of schizophrenics and variability in the types and severity of structural, physiological and functional abnormalities suggests significant genetic and neuropsychological heterogeneity within schizophrenia. Linking these two research domains thus represents a crucial step toward elucidating the heterogeneity of schizophrenia. This proposal is for a unified investigation of genetic and neuropsychological heterogeneity in schizophrenia with the goal of determining whether specific types of neuropsychological deficits reflect familial or nonfamilial etiological factors. Using the family study method, the proposed research plan is to diagnose psychiatric hospital patients and normal controls, and their first degree relatives using structured psychiatric interviews, and to assess all of these individuals with a comprehensive neuropsychological battery. Genetic heterogeneity will be assessed by dividing schizophrenics into familial and nonfamilial subgroups defined on the basis of the family study. Specificity of neuropsychological deficits to familial and nonfamilial schizophrenia will be assessed by comparing the schizophrenic subgroups to bipolar and normal control groups. The study will compare the ability of seven different models of schizophrenia to account for the patterns of neuropsychological variability among the schizophrenic groups and their relatives. The results will also be used to determine the heritability of neuropsychological deficits, to determine which measures may be vulnerability markers for schizophrenia, and to study neuropsychological impairment as an alternative schizophrenic phenotype for use in segregation analyses. By elucidating the genetic and neuropsychological heterogeneity of schizophrenia, the proposed study will clarify underlying mechanisms and aid in the development of homogeneous subgroups that will be informative for future etiological and pathophysiological research. The delineation of neuropsychological profiles and familial and nonfamilial forms of schizophrenia will contribute to psychiatric genetic counseling and to the development of treatment-relevant subgroups.

The major goal of this research is to study the relationship between the extent of family history (FH) of alcoholism and outcome of treatment for alcoholism using state-of-the-art methods for defining and measuring both FH status and treatment outcome. It is hypothesized that family history positive (FH+) alcoholics have a double set of deficits to overcome in their bid for sobriety and will have a worse treatment outcome over time than will family history negative (FH-) alcoholics. FH+, as compared to FH-, alcoholics may be impaired not only by a more severe form of alcoholism, but also may be developmentally impaired from having grown up in an alcoholic family. Specifically, 180 male alcoholics in an inpatient alcoholism treatment program will be divided into three FH groups of 60 each according to the alcoholism of their parents: both parents alcoholic, father only alcoholic, and no alcoholic parents. FH statue will be determined from a convergence of the reports of the subject, the subjects's sibling and the subject's mother or other knowledgeable parent generation relative. Conservative criteria for both the presence and the absence of parental alcoholism will be used to minimize false positives and, most importantly, false negatives. These three groups of alcoholics will be followed for 18 months to determine outcomes after inpatient treatment. Treatment outcome evaluation will include careful planning to minimize subject attrition and measurement of multiple life health outcomes. Alcoholics will be breath tested to determine whether they are alcohol-free when interviewed, and corroborating evidence from collateral reports and biochemical markers will be used with the alcoholics' self-reports in a convergent validity approach to the measurement of drinking and related behavior. A second goal of this project is to contribute to the growing literature on differences between FH+ and FH- alcoholics in course and severity of alcoholism, treatment entry psychosocial adjustment, and life history experiences and adaptation. To this end, a multidimensional assessment will be conducted while the alcoholic is in the hospital and cover the following domains on which the three FH groups will be compared: family background and childhood history, current and past cognitive functioning, drinking and related behavior, social and vocational adjustment, and mental and physical health. Finally, a pilot study will compare FH+ alcoholics with their nonalcoholic adult male biological siblings on this multidimensional assessment battery to better understand this selective vulnerability.

The goal of the project is to determine the effects of familial environment, non-familial environment, and genetic factors on the etiology and course of drug abuse and dependence. There have been a number of family studies, one adoption study, and no rigorous twin studies of genetic factors in drug abuse/dependence. The identification of environmental factors involved in the etiology and maintenance of drug abuse/dependence has implications for treatment and prevention. The identification of genetic factors has potential for identifying high risk groups. The project will use the Vietman Era Veterans Twin Registry, comprising 6000 pairs, compiled by the National Academy of Sciences (NAS). Data will be collected by the Institute for Survey Research under subcontract to NAS. Data will be collected in 3 stages. The first stage will be a questionnaire mailed to all twins eliciting information on personal history and screening information on drug use and mood and antisocial personality disorders. In stages II, 4916 subjects, including all of those screened positive for drug use, will be included. Data collection entails structured telephone interviews for diagnosing drug abuse and dependence and mood and antisocial personality disorders. Stage III will include 2000 of the subjects using a mailed questionnaire to identify putative risk/protective factors. These data will be relevant for assessing if: (1) different drugs are influenced differently by familial, nonfamilial, and genetic factors; (2) different types of problematic use of the same drug are influenced differently by familial, nonfamilial, and genetic factors; (3) groups of drugs can be identified that are under the control of the same familial, nonfamilial, and genetic factors; (4) psychiatric morbidity mediates the genetic influence on drug abuse/dependence; (5) environmental risk/protective factors can be identified.

schizophrenia; mental disorder diagnosis; mental health epidemiology; disease /disorder classification; longitudinal human study; tranquilizer; mental disorder chemotherapy; haloperidol; methylphenidate; neurobiology; neuroendocrine system; behavioral genetics; social adjustment; visual tracking; evoked potentials; electroencephalography; brain mapping; human subject; nervous system disorder diagnosis; neuropsychological tests; magnetic resonance imaging;

The evidence from family, twin, and adoption studies has established that the risk for developing schizophrenia involves a substantial genetic component (McGuffin, Farmer, and Gottesman, 1987). The major challenge now confronting researchers is to determine the mode or modes of genetic transmission. Advances in the last decade in recombinant DNA technology, especially improvements in methods for generating and isolating restriction fragment length polymorphisms (RFLPs), have made it possible to use DNA sequence polymorphisms to trace alleles at specific genomic loci. The power of these methods have recently been demonstrated by two landmark studies (Barrett et al., 1988; Sherrington et al., 1988) providing the first evidence for genetic linkage in schizophrenia. These findings foreshadow a series of breakthroughs in clarifying the etiology and pathophysiology of schizophrenia. The major objective of this proposed Diagnostic Center is to work as one of three centers in developing a sample of 200 multiplex schizophrenic families for subsequent linkage studies. To this end, approximately 67 families will be recruited from a Harvard Medical School consortium of hospitals from the 6-month to the 4 1/2 year period of a 5-year contract. Families will be sought with two or more affected siblings and an average of seven first-degree relatives, including both parents. More two- generation families. Selected probands and relatives will receive extensive assessment using state of the art tools in psychiatric epidemiology and neuropsychology. Blood samples will also be obtained from all subjects for transformation into lymphoblast cell lines for subsequent linkage analyses. The principal hypothesis of this study is that a major gene or genes contributes to the etiology of this disorder, which can be identified through genetic linkage analysis of a well-defined sample of multiplex families. The key question is whether there are specific DNA markers which co-segregate with the disease trait. Linkage analyses of our center sample will focus first on the chromosome 5 region where linkage has been shown as well on DNA markers linked to loci implicated in catecholamine metabolism.

We propose to continue a line of research, enabled by an NIMH Merit Award, which has been testing hypotheses about neurobiologic manifestations of schizotaxia (the predisposition to schizophrenia) among schizophrenic patients and the nonpsychotic adult relatives of schizophrenic patients. Our work-showing that schizotaxia is associated with negative symptoms, neuropsychological dysfunction and structural brain abnormalities-converges with data from other centers to show that schizotaxia is a subtle brain disorder affecting about 20 to 50 percent of the nonpsychotic relatives of schizophrenic patients. The data collected during prior funding periods have allowed us to demonstrate a) neuropsychological deficits in schizophrenic patients and their relatives, b) gender differences in the expression of these deficits, c) stability of these deficits over time, d) structural brain abnormalities in patients and relatives, e) functional MRI abnormalities in patients and relatives, and f) how the psychometric features of neuropsychological tests make them useful for assessing phenotypes in genetic linkage studies of schizophrenia. We have decided to pursue three major aims in this continuation proposal that will help us better understand the neural substrates of schizotaxia and how they lead to schizophrenia. First, we will identify predictors of social dysfunction and psychopathology in adolescent children of schizophrenia patients. Second, we will better describe the neural substrate of schizophrenia prior to the onset of psychosis and lay the foundation for work that will examine if neurodegeneration occurs after illness onset. Third, we will establish the infrastructure required to monitor adolescents at risk for psychosis so that future proposals can select adolescents at risk for schizophrenia for prevention protocols. To accomplish these aims, we will assess 300 adolescent children of schizophrenic patients (ACSZ) and 50 normal controls with neuropsychological, psychosis proneness, psychosocial functioning, and family adversity measures. All controls and 150 randomly selected ACSZ will also be evaluated with magnetic resonance imaging. All 350 subjects will be monitored for adverse outcomes at six month intervals. Given the wide age range for the onset of schizophrenia, we plan to follow this sample for many years. Thus, we will also lay the foundation for future proposals that will monitor the incidence of psychosis in this sample through young adulthood. This will eventually allow us to assess 1)the predictive validity of schizotaxia measures and environmental adversity for subsequent psychosis and 2)longitudinal changes in neuro-psychological functioning and brain structure in subjects who do not become psychotic. Achievement of the second goal will help clarify which brain abnormalities in schizophrenic patients can be attributed to neurodevelop-mental events prior to onset and which are due to neurodegeneration after onset.

DESCRIPTION (Adapted from the Applicant's Abstract): The investigators propose to supplement existing data from the Providence and Boston cohorts of the National Collaborative Perinatal Project (NCPP) in this first of a two-phase project. In the first of four major aims listed, the investigators will test a hypothesis that prenatal and perinatal complications require a familial predisposition to schizophrenia (or to psychosis) for the development of characteristic neurobehavioral deficits at age 7. The second aim will further explore the importance of diagnostic specificity (schizophrenia versus affective psychosis) in this effect.The third and fourth of these aims will require the full data set generated by phase 2. Aim 3 will determine the relative risks for specific disorders among offspring as functions of prenatal and perinatal complications, familial predisposition, and neurobehavioral deficits manifested in childhood. Aim 4 will further explore the interactions between these predisposing factors and adult psychopathology. The investigators will begin with the results of screening questions obtained at the time of the 7-year followup to locate and diagnostically evaluate 125 parents with a lifetime history of psychosis leading to hospitalization. Based on pilot data, they expect that 50 of these individuals will have DSM-III-R schizophrenia, 50 will have affective psychoses, and 25 will have other DSM-III-R diagnoses. Control NCPP parents will be matched to these psychotic NCPP parents on the basis of age, sex, ethnicity, socioeconomic status, site, parity, sex of offspring, and the presence or absence of prenatal and perinatal complications (PPCs) in offspring. Normal controls will lack Axis I (except adjustment disorders) or Axis II disorders. Phase 2 is to add an additional 75 parents to each group for a total of 200. For each group of parents studied in phase 1, 100 offspring will be evaluated; phase 2 will add another 150 parents in each group for a total of 250. The Diagnostic Interview for Genetic Studies (DIGS) will be the core diagnostic instrument. Various measures of attention, short-term memory, concept formation, and neuromotor function will be used as probes of genetic risk factors; various measures of long-term memory and olfaction are intended to quantify the sequelae of specific perinatal complications. Finally, measures of cerebral laterality will be used to contrast subjects with schizophrenia and subjects with affective disorder.

Considerable research has been directed towards assessing the long-term consequences of illicit drug abuse, but little definitive information is available because of difficulties in distinguishing causes of drug abuse from effects. Observed differences might reflect either the consequences of drug abuse or a vulnerability to drug abuse that predated drug usage. This case-control study will evaluate the long-term impact of drug abuse on brain function by capitalizing on our epidemiological sample of over 8,000 twins from the Vietnam Era Twin Registry who have been interviewed by us within the last three years about their drug use history. We will compare twins who previously abused drugs to their co-twins who have never abused drugs. We propose to study the two most frequently abused categories of drugs in our sample, marijuana and stimulants (amphetamine & cocaine). We define abuse as the regular use of the drug (i.e., at least once per week) for at least one year. This definition identifies subjects with substantial histories of use (for marijuana: median length of regular use = 8 years, median days/week of use = 7; for stimulants: median length of regular use = 3 years, median days/week of use = 4). Subjects who have abused drugs other than marijuana or stimulants or experienced tolerance or withdrawal to alcohol will be excluded. Non-abusers never used any drug more than five times in their lives. Because marijuana is the most frequently abused illicit drug, our sample includes enough monozygotic pairs to comprise a sample of discordant pairs (n= 118 pairs) adequate for testing differences between abusers and non-abusers. Stimulant abuse is less prevalent, so there are fewer discordant monozygotic pairs available for study (n=69 pairs). Therefore, for stimulants we will supplement the discordant monozygotic pairs with discordant dizygotic pairs (n=55 pairs). A projected individual response rate of 76% will yield a total sample of approximately 141 pairs in which both twins participate. We will compare former drug abusers to their drug non-abusing co-twins in a classic epidemiological matched pair design. The goal of the study is to determine if marijuana and/or stimulants have long-term consequences on brain function as reflected in neuropsychological performance (memory, attention, problem solving, executive functions, etc.), electrophysiology (in a subset of monozygotic twins discordant for stimulant abuse), psychiatric status (Axis I and II psychopathology), and personality. We will also investigate dose-response relationships between the drugs and brain function, the possible role of age of initiation, and recovery of functioning. The consequences of marijuana will be compared to the consequences of stimulants. Twins from the discordant pairs will travel to one of the collaborating sites (Harvard Univ., Washington Univ., or Univ. of Illinois) or he visited in their home city for the administration of structured interviews, neuropsychological testing, and self-report measures. Those pairs participating in electrophysiological assessment will travel to the Harvard site.

It is proposed to study 90 sib-pairs with the disorder. Specific aims are: (1) to clinically assess a pedigree sample having adequate power to detect genes for schizophrenia; (2) to conduct a genome scan to find such loci; and (3) to transmit all data to the NIMH designated cell repository and data management centers. The goals will be attained by achieving the following: (1) from Taiwan and China, 900 Han Chinese sib- pairs having DSM-IV schizophrenia will be collected. (2) The investigators will examine all family members using the Diagnostic Interview for Genetic Studies and the Family Interview for Genetic Studies. The PI participated in the development and field testing of these interviews and has an already established training program for their use. They have been translated into Mandarin by the Taiwanese investigators, who have used them in prior studies. (3) Blood samples will be sent to the NIMH designated cell repository for creation of lymphoblastoid cell lines. (4) Clinical data will be entered using the database software created for the NIMH Human Genetics Initiative. Data will be vetted and sent to the NIMH designated data management center. (5) The investigators will complete a genome scan using 450 markers spaced at an average of 10 cM intervals using markers that have been optimized for use in the Han Chinese population. The scan will be completed with no cost to the NIMH through an agreement with Millenium Pharmaceuticals, a biotechnology company in the Boston area that the PI has worked with on a prior genetic linkage study of schizophrenia. All genetic analyses will be approved by the consultant, Eric Lander, Ph.D. (6) all clinical data will be made available to the scientific community by the end of the funding period. All genotypes will be available on year after they are created but no later than a year after the funding period. This project is feasible because: (1) The PI has already coordinated one multi-site genetic linkage study of schizophrenia and has participated in a second. (2) The investigators have a longstanding relationship with the Taiwanese collaborators and an effective, albeit, more recent working relationship with the Chinese collaborator. (3) The investigators have conservatively estimated that each site has access to more than enough available families having two schizophrenic siblings. (4) The PI's Harvard team has had prior experience collaborating on genotyping and linkage analysis projects with Millennium Pharmaceuticals. This, and Millennium's prior genotyping experience shows that the genotyping phase of the work is feasible.

This R25 application responds to the NIMH RFA, "Training Future Mental Health Clinical Researchers," and seeks to establish a new educational clinical research training program in psychiatric genetics. The overall aim of the proposed training program is to fill a gap in the post-doctoral training opportunities available in clinical research. Currently, it is difficult for a young scientist to acquire a focused training in psychiatric genetics. There are many excellent training programs that focus on components of the psychiatric genetics curriculum, but few that provide complete training in all areas of this discipline. The Executive Director of this training program has worked in the field of psychiatric genetics for over three decades, and has created an extensive research infrastructure that has brought together a variety of independent investigators who have long collaborated together in productive efforts to understand the genetic basis of psychiatric disorders. The training and development of new investigators has always been a cornerstone of our collective efforts. In this proposal, we seek to formalize our commitment to prepare future generations of independent researchers by creating a comprehensive, multidisciplinary post-doctoral training program to provide promising young mental health clinicians with the tools needed to pursue clinical research in psychiatric genetics. The goals of the program are: 1) To create a coherent didactic and experiential post-doctoral training program in psychiatric 2) To implement this training program using faculty from multiple disciplines: psychiatry, psychology, statistical genetics, molecular genetics, behavioral genetics, neuropsychology and neuroscience; and 3) To evaluate this program by assessing the ability of trainees to attain the following goals during their training: a) maintenance of a 3.5 grade point average in didactic courses; b) publication of two papers per year and c) submission and funding of a competitive grant application. Finally, we will seek to create a training infrastructure that will allow students to benefit from the rapid advances of science and technology on the one hand, and their ethical, legal and social consequences, on the other.

Our proposal is a response to NIDA's Request for Applications entitled `Molecular Genetics of Drug Addiction Vulnerability. The main goal of the proposed study is to detect one or more genes responsible for the genetic transmission of heroin dependence. Our Specific Aims respond to those specified in the RFA: 1) To collect and clinically characterize a large sib-pair sample with adequate statistical power for identifying genomic regions that may harbor loci conferring susceptibility to heroin dependence; 2) To conduct a whole-genome scan to establish the chromosomal localization of such loci; 3) To follow-up regions of interest from the whole-genome scan and evaluate candidate genes; and 4) To make the clinical and genotypic data quickly available to other investigators in the scientific community. To accomplish our aims, we have established a collaboration with two psychiatrists in Yunnan Province, China. This province, which borders the "Golden Triangle" -- the source of much of the world's heroin -- has a comprehensive drug abuse registration system to which our colleagues have access. About 30,000 heroin addicts are in the registry and can be easily located by our Chinese collaborator, the Director of the Yunnan Institute for Drug Abuse. We will collect blood and diagnostic information (using a structured diagnostic interview) from 1000 sib-pairs having DSM-IV defined heroin dependence as well as from their parents and other affected and unaffected siblings. Blood samples will be sent to a cell repository at Coriell Laboratories for creation of lymphoblastoid cell lines. In collaboration with a colleague at Washington University, we will complete a genome scan using 350 markers spaced at an average of 10 cM intervals. Genotype and clinical data will be entered using database software. We will conduct a multipoint linkage analysis using the guidelines of Lander and Kruglyak to assert statistical significance. We will follow-up regions of interest with a denser set of markers and evaluate candidate genes. All clinical data will be made available to the scientific community by the end of the funding period. All genotypes will be available one year after they are generated, but no later than a year after the end of the funding period. We are submitting this proposal using the RO1 mechanism.

This is a proposed Phase III of our Neurodevelopmental Study of Schizophrenia, in which we intend to assess the consequences of genetic and/or pre- and perinatal complications (PPCs) using high resolution structured magnetic resonance imaging (MRI). Studies of schizophrenia have implicated gray and white matter abnormalities in limbic-diencephalic, paralimbic, and prefrontal brain regions. These brain abnormalities result from genetic and/or environmental (i.e., obstetric) factors. Further, some nonpsychotic relatives of patients with schizophrenia suffer from similar, milder, ("subsyndromal") dysfunctions. We are proposing a 5-year study to continue a 40-year prospective high risk study to directly test the consequences of genetic vulnerability (assessed by psychosis in the parent) and specific PPCs (i.e., chronic fetal hypoxia and infections in the second trimester), on cortical and subcortical brain volumes in adult offspring of parents with schizophrenia or affective psychoses. We have a unique opportunity to re-evaluate subjects that we have carefully studied by clinical and neuropsychological evaluations in Phase II of the study. The sample was originally ascertained from a community cohort of pregnancies drawn from the Providence and Boston cohorts of the National Collaborative Perinatal Project. At these sites, 17,741 pregnancies were followed prospectively and systematically recorded, and the offspring's mental and physical development were assessed at 4 and 8 months, and 1,4, and 7 years of age. We have systematically located, recruited, and diagnosed 200 parents with psychotic disorders, and 200 normal comparison parents, individually-matched on specific parent and offspring characteristics. 403 of their adult offspring, ages 31-37, have been identified in Phase II. We estimate that we will ascertain 85 percent of these 403 offspring for MRI. This study is unique in that we can specify developmental predictors of structural brain abnormalities and their functional consequences due to risk and obstetric status, for offspring who become psychotic, or exhibit subsyndromal expressions of the genetic diathesis. Further, we will identify the contribution of PPCs and/or genetic vulnerability to neural circuit abnormalities and demonstrate the specificity for schizophrenia versus affective psychosis. We will use a high resolution MRI and highly detailed, reliable image analysis techniques programs to link etiological predisposition to adult brain volumes (i.e., in limbic-diencephalic, paralimbic and cortical regions, and white matter tracts). This study has important implications for understand2ing the etiology and development of schizophrenia.

DESCRIPTION (provided by applicant): Neurobiological deficits that serve as informative endophenotype markers have been demonstrated in schizophrenia by a number of different paradigms. Neurophysiological deficits are prominent in P50 event related suppression, prepulse inhibition (PPI) of the startle response, and the antisaccade (AS) task for eye movement dysfunction. Neurocognitive deficits in schizophrenia are revealed by poor performance on the CPT, verbal memory, and tests of working memory. Each of these deficits has also been demonstrated in clinically unaffected relatives of schizophrenia patients, which is evidence that they may reflect part of the heritable risk for the illness. This conclusion is reinforced by findings of deficits in non-psychotic, unmedicated schizophrenia patients, and schizotypal patients. The null hypothesis is that all 6 deficits reflect a single, common underlying heritable dysfunction in all schizophrenia patients. A test of that hypothesis requires measurement of all of these deficits in the same group of schizophrenia patient probands and their relatives. If they are all manifestations of the same genetic dysfunction (although perhaps expressed in different brain areas), then a multivariate analysis would show that they all contribute to a single dimension in both relatives and schizophrenia patients. An alternative hypothesis is that only one or a small subset of deficits is present in each family, which is consistent with the heterogeneity found in current genetic linkage studies. In that case, the multivariate analysis would show the different measures or subsets of them loading onto different dimensions. Schizophrenia itself is likely to be the result of multiple deficits in any individual. Therefore, the analysis is performed in the same cohort of schizophrenia patient probands and their relatives to take advantage of Mendel's second law, which holds that genetically independent deficits segregate independently. Hence, although schizophrenia patient probands themselves have multiple deficits, if the deficits are caused by different genetic factors, then they will segregate to different groups of relatives. This 7 site collaborative RO1 project will gather a combined total of 420 pedigrees (1680 subjects) and 525 normal subjects over 5 years (each site will contribute 1/7th of these totals). Findings of heritable deficits in specific measures will be used to guide the next generation of studies of the genetics of schizophrenia.

DESCRIPTION (provided by applicant): Bipolar disorder is a common psychiatric illness with devastating consequences for affected individuals, their families, and society. The exact causes of this illness are presently not known, despite considerable research effort dedicated to this objective, in part due to the complexity of the illness, both in its phenotypes and its etiology. Due to this complexity, and although twin and adoption studies clearly demonstrate that bipolar disorder is highly heritable, risk-gene- identification efforts for this illness have been hampered by low power. Research into the biological basis of bipolar disorder is currently advancing in humans and--to a lesser extent--in animals, more or less independently. Each independent line of investigation (i.e., animal and human studies) is contributing to the incremental gains in knowledge of bipolar disorder etiology witnessed in the last decade. Yet, it is now clear that the lack of integration between these two lines of investigation is hindering the pace of risk-gene identification or, perhaps more accurately, constitutes a missed opportunity for optimizing the approach to gene discovery. Our group has laid the foundation for overcoming this barrier to resolving the genetic etiology of bipolar disorder by implementing a convergent functional genomics approach that capitalizes on multiple sources of information from various disciplines to narrow the search for bipolar disorder susceptibility loci and genes, and increase the stringency and accuracy of designating a candidate gene as a causal factor in the illness. In the present application, we propose to apply this methodology to the study of distinct features of the bipolar disorder phenotype (e.g., cycling and switching), and extend it by testing novel candidate genes for association with the illness in a phenotypically enriched and presumably more homogeneous--sample of patients with pediatric onset of bipolar disorder. The goal of this work is to identify the genes that contribute to the emergence of bipolar disorder and regulate its most prominent features: mood cyclicity and switching. The identification of these genes may have numerous consequences for the future of bipolar disorder, including the improvement of diagnostic approaches to the illness, the construction of individually tailored pharmacological and psychosocial treatments for and interventions in the progression of the illness, and, ultimately, the prevention of its occurrence.

DESCRIPTION (provided by applicant): The propensity to abuse illicit drugs is at least partially determined by genes. In fact, the genetic contribution to the total risk for drug abuse varies from about one-third to two-thirds, depending on the class of drug. Yet, despite the fact that substance use disorders (SUDs), including those involving alcohol and tobacco, have a sizable genetic component, the specific genes that confer this risk have been quite elusive. Undoubtedly, some of the difficulty in identifying the susceptibility genes for SUDs stems from the underlying etiologic complexity of these phenotypes. Drug abuse is presumed to have a multifactorial polygenic etiology in which numerous genes and environmental factors all make small contributions to the overall risk for the illness. Compounding the difficulty in identifying genes that impart risk for SUDs may be the heterogeneous nature of the phenotypes that have traditionally been chosen for genetic studies of the liability toward SUDs. Alternative definitions of SUDs may facilitate the identification of genetic and environmental influences. These alternative phenotypes may be subtypes of SUDs and/or quantitative traits. Using alternative SUD classes and/or quantitative measures of SUDs could resolve genetic heterogeneity. In other words, alternative phenotypes may provide a stronger 'signal' in the search for underlying risk genes using linkage and association paradigms. Untangling the complicated pathway from genotype to phenotype is one of the biggest challenges facing investigators searching for individual genes for substance abuse. The same genes may give rise to varying phenotypes; conversely, behaviors that appear to be related phenotypically may in fact be etiologically unrelated, at least genetically. The definition of phenotypic boundaries is essential to the success of linkage and association studies to find genes predisposing to substance abuse. The identification of individual genes will pave the way for improved treatment by leading to, for example, the early identification of high-risk individuals, advances in clinical decision-making regarding different forms of treatment, and individually tailored drugs. Research aimed at identifying more genetically homogeneous phenotypes for illicit drug abuse lags well behind that for alcohol and tobacco use and abuse. The enormous health and financial costs of illicit drug abuse, both to the individual and to society, underscores the pressing need for such research.

DESCRIPTION (provided by applicant): The diagnosis of schizophrenia presently relies on essentially the same methods of clinical interviewing, symptom identification, and application of diagnostic criteria that have been in use for nearly 25 years. Given the considerable genetic contribution toward the etiology of schizophrenia, a biologically based set of diagnostic markers may be more efficient for validating the diagnosis and its subtypes, and for improving diagnostic sensitivity and specificity. Moreover, if diagnostic markers can be identified in peripheral blood rather than in postmortem brain tissue, the promise of early identification and improved treatment might be realized. As a first step toward establishing such biologically based markers, the main objective of this project is to identify and validate highly reliable and specific patterns of gene expression in peripheral blood cells that differentiate patients with schizophrenia from control subjects. We have generated pilot data identifying differences in gene expression profiles between patients with schizophrenia, patients with bipolar disorder, and control subjects; in the proposed project, novel applications of oligonucleotide microarrays and advanced statistical techniques will be used to extend and validate these preliminary findings. To accomplish this, we propose the following specific aims: 1) Quantify gene expression levels in peripheral lymphocytes from patients with schizophrenia and control subjects; 2) Quantify levels of gene expression induced by the application of typical and atypical antipsychotic medications to cultured lymphocytes obtained from control subjects, in order to rule out candidate biomarker genes that respond to medication but are not specific to the disease process; 3) Prioritize candidate genes for subsequent verification and validation as disease biomarkers using novel data-analytic and statistical methods that control the rate of inferential errors; 4) Verify the differential expression of top candidate genes in peripheral lymphocytes by more precise mRNA quantification techniques; and 5) Validate the differential expression of top candidate genes in postmortem brain tissue obtained from patients with schizophrenia and control subjects. Long-term goals (beyond this proposal) will be to replicate and extend our findings, test for other potential confounders of peripheral bloodbased gene expression profiling, and test for specificity vis-a-vis psychiatric comparison groups. Ultimately, biologically based diagnostic markers may substantially enhance the possibilities for early identification, intervention, and prevention, and also facilitate the identification of etiologic factors in the illness.

DESCRIPTION (provided by applicant): This proposal responds to Request for Applications RFA-MH-08-131, which seeks Collaborative R01 applications that propose to enrich pre-existing resources for schizophrenia in the NIMH Human Genetics Initiative and to apply genomic methods to further our understanding of the molecular etiology of the disorder. The overarching aims of this proposal are to quickly and cost-effectively ascertain a large sample of trio families affected by schizophrenia, and to discover causal variants for the disorder in the first family-based genome-wide association study (GWAS) of the illness. In our recently completed NIMH-funded Genetic Linkage Study of Schizophrenia (R01MH059624;PI: Ming T. Tsuang), we established a large and efficient ascertainment network and infrastructure in Taiwan, which will again be utilized and expanded in the proposed study. Through additional ascertainment within this framework, we will collect an aggregate sample of 5,000 trios with adequate power for detecting in a GWAS those variants that make even small contributions to the risk for the disorder. We will meet the overarching goals of this project by accomplishing several Specific Aims, as follows: 1) Supplement our previously collected sample of 1,200 Han Chinese schizophrenia-affected nuclear families from Taiwan by rapidly screening and collecting an additional 3,800 trios from ten ascertainment sites in Taiwan;2) Assess the association of schizophrenia with a genome-wide panel of single-nucleotide polymorphisms and their constituent haplotypes;3) Perform a genome-wide survey for copy-number variations related to schizophrenia;4) Test for gene-gene interactions (epistasis);5) Test for gene-environment interactions, such as the well-established effect of season of birth;6) Analyze quantitative schizophrenia phenotypes, such as symptom scores and age at onset;and 7) Enhance the NIMH Genetics Initiative collections by sending all clinical data, biomaterials, and genotypes to the appropriate repositories. The project would achieve the goals of the RFA by enriching the existing resources of the NIMH Human Genetics Initiative and by applying the latest genomic research methods to further our understanding of the molecular etiology of the disorder. Also, by capitalizing on an existing clinical infrastructure and an efficient screening and assessment protocol, we will obtain a well-powered sample in a very rapid and cost-effective manner.

DESCRIPTION (provided by applicant): Second-generation antipsychotic medications have proven quite effective in reducing symptoms of schizophrenia, but little progress has been made in improving behavioral outcomes of relevance to affected individuals, their families, and policymakers. A major goal of research on functional outcomes in schizophrenia should be to understand for whom and under what circumstances interventions aimed at improving functional abilities are effective. Currently, research on the assessment or improvement of functional outcomes is usually conducted in a primarily psychosocial context, while genetic studies are typically viewed as conceptually quite distant from such studies. However, it is highly likely that there are significant direct and/or indirect genetic influences on functional outcomes such as interpersonal skills, participation in community activities, and work skills. In this proposal, we seek to bridge these two seemingly distant ends of the spectrum of schizophrenia research. Our previous non-genetic research showed that cognitive functions and negative symptoms influence functional capacity, which in turn influences functional outcomes in schizophrenia. On the other hand, depressive symptoms independently predicted functional outcomes. Separate work by our group and others has demonstrated the heritability of many cognitive abilities and symptoms, and some progress has been made in identifying specific genes that influence some of these traits. We now propose to test the hypothesis that several specific genes will contribute to functional outcomes either through direct pathways or through their influences on cognition and symptoms. To accomplish this goal, we are proposing the following specific aims: 1) To determine which specific neuropsychological functions and symptom dimensions predict functional capacity and functional outcomes in schizophrenia;2) To detect direct and mediated effects of genetic polymorphisms on real-world functioning, specifically focusing on: a) empirically supported candidate polymorphisms;b) all haplotype-tagging single nucleotide polymorphisms in and around empirically supported candidate genes;and c) ontologically related candidate genes;and 3) To establish multivariate models of functional outcomes, by: a) evaluating possible additive and epistatic interactions between candidate genes;and b) evaluating pleiotropic effects of each candidate gene. Linking basic research with the study of functional performance is only a first step toward future development of novel therapeutics targeted at aberrant proteins, but it is an important one. The development of such treatments is likely to be a long-term process, but one that could ultimately enhance psychosocial interventions to improve real-world functioning. PUBLIC HEALTH RELEVANCE: We will test the hypothesis that several specific genes contribute to functional outcomes, either through direct pathways or through their influences on cognition and symptoms, in schizophrenic patients. This project has important implications for personalized treatment and case management. By linking basic genetic research with the study of functional performance, we take the first step toward the development of new types of treatments targeting aberrant proteins, which could ultimately enhance psychosocial interventions and lead to an improvement in real-world functioning.

DESCRIPTION (provided by applicant): Why do some individuals exposed to traumatic events develop Post-traumatic Stress Disorder (PTSD), while others do not? Risk and resilience are two concepts that are important in understanding the etiology of PTSD. Past research on PTSD has identified different factors that may put individuals at greater risk of developing PTSD, such as family history, childhood experiences, personality variables, and preexisting mental disorders. The concept of resilience, often considered to represent resistance to the negative effects of a traumatic event, is also very important in any consideration of PTSD. Resilience can be defined as "the ability to successfully adapt to stressors, maintaining psychological well-being in the face of adversity." Equally as important, however, is the concept of resilience as the ability to improve after the development of PTSD. The biological factors associated with the risk of developing PTSD, as well as both aspects of resilience are poorly understood. This project will investigate biological risk and resilience factors associated with the development of PTSD among a sample of Marines before and after they are exposed to combat stress utilizing genomic methodologies and a battery of behavioral and biological assessments. Specifically, we propose to investigate gene- expression based biomarkers of PTSD risk and resilience. The development of PTSD following initial traumatic exposure is quite variable, with some individuals never exhibiting signs of PTSD, while others are plagued with incapacitating symptoms despite years of combinations of therapy. At present, the basic characteristics underlying resilience are unknown, and it is not currently possible to predict the development of the disorder, much less any eventual outcome in any given individual. If biomarkers related to risk can be discovered, they may help to identify which individuals are at risk, and lead to more effective primary prevention protocols. Although environmental factors are clearly essential for the development of PTSD, the disorder also has a demonstrable genetic component. We plan in this study to identify gene-based biomarkers of PTSD in an effort to better understand the biological factors related to both the risk of developing PTSD;and to resilience as represented by resistance to development of PTSD. To accomplish these objectives, we have proposed two specific aims as follows: Specific Aim 1: Identify gene-expression-based biomarker profiles of PTSD risk and resilience in peripheral blood mononuclear cells and Specific Aim 2: Identify changes in gene-expression-based biomarker profiles of risk and resilience in peripheral blood mononuclear cells. PUBLIC HEALTH RELEVANCE: The development of Post Traumatic Stress Disorder (PTSD) following initial traumatic exposure is extremely variable. While some individuals never exhibit signs of PTSD, others are plagued with incapacitating symptoms despite years of combinations of therapy. Understanding the gene expression patterns of combat-exposed individuals who go one to experience PTSD, as compared to the gene expression patterns of combat-exposed individuals who did not go on to experience PTSD, will shed light on the biological mechanisms involved in PTSD and provide a basis for more effective prevention and treatment of this devastating disorder.

Both the NIH and the NIA-Alzheimer's Association have emphasized the importance of early identification beginning in midlife to predict Alzheimer's disease (AD) and cognitive decline. Two keys to early identification are accurate detection of mild cognitive impairment (MCI), which can be a precursor of AD, and identification of biomarkers of MCI risk with potential for screening large populations. We have shown that blood-based transcriptomic signatures accompany and, in some cases, predict the development of some psychiatric illnesses, and others have recently found the same for MCI and AD. However, existing cross-sectional case- control biomarker studies of MCI were not designed to illuminate whether peripheral blood transcriptome biomarkers are precursors, concomitants, or consequences of MCI, and they are unable to shed light on the relative influence of inherited and environmental factors on each component of the putative biomarker signature. This is important because MCI and AD are both partially heritable disorders. Our proposed project would address these pressing questions within the context of an ongoing study that was explicitly designed to allow such inferences: our longitudinal Vietnam Era Twin Study of Aging (VETSA). The VETSA, just beginning wave 3 of longitudinal data collection from 1151 twins, began studying subjects at an average age of 56 (range: 51-60). The mean age of subjects in VETSA 3 will be 67; thus it will provide data both before and during the key transition from midlife to early old age. A large twin sample with a narrow age range that has been longitudinally characterized for many years on multiple domains (cognitive, physiological, psychological, biomedical, and genetic) makes VETSA uniquely well suited to characterizing individual differences in cognitive aging with a focus on MCI, beginning in midlife. The present proposal seeks to expand our ability to detect MCI early in VETSA subjects in a highly efficient and cost-effective manner by integrating transcriptome measurements from peripheral blood cells into the existing VETSA 3 protocol. We propose to collect an additional blood sample from all subjects in VETSA 3, extract and sequence RNA from those samples, and merge these transcriptome measures with the other data collected in VETSA to pursue three Specific Aims as part of a new VETSA Gene Expression (VETSA-GEX) project: 1) Construct an atlas of genetic and environmental influences in expression levels of all RNA transcripts (including both coding mRNAs and short and long non-coding RNAs) and gene co-expression networks detected in peripheral blood at midlife; 2) Discover, replicate, and functionally characterize peripheral blood transcriptomic signatures of neuropsychologically defined MCI and MCI severity, both by non-twin analysis of the entire sample and by co- twin-control analysis; and 3) Integrate peripheral blood transcriptome measures with other putative MCI biomarkers already being measured on these twins, including plasma ?-amyloid (A?) and phosphorylated-tau (p-tau) levels, genetic risk scores, and psychophysiological measures such as task-evoked pupil dilation.

Project Summary Over the last decade, scientists have accelerated their efforts to understand Alzheimer?s disease (AD). This has led to unprecedented knowledge of the genetic and biological bases of AD risk, and vast stores of valuable data for further mining. Understanding the genetic and biological risk states for AD is, in itself, extraordinarily valuable for guiding mechanistic studies, developing better diagnostics, and formulating therapeutics. But an understanding of risk states also has the benefit of allowing research on resilience to AD. Research on the genetic and biological bases of resilience necessarily lags behind the discovery of risk factors. Now, as the risk architecture of AD is coming into view, it is feasible to study resilience to AD in individuals who are cognitively normal despite being at elevated risk for the disease. The approach we have devised for identifying resilience factors is straightforward yet, to our knowledge, unprecedented. We identify unaffected individuals at the highest levels of multivariate risk, match them to affected individuals at equivalent levels of risk, and contrast these two subgroups to find residual variation associated with the absence of disease. In this project, we will capitalize on the wealth of existing high-throughput AD risk-factor results and data, and our involvement in many of the world?s largest AD consortia, to efficiently map resilience to AD at three levels (genetics, transcriptomics, and neuroimaging), and to integrate across these levels. In Aim 1, we will identify genetic variation associated with resilience to AD in the presence of elevated genetic risk conferred by APOE ?4 alleles, an elevated AD polygenic risk score, or an elevated AD polygenic hazard score. In Aim 2, we will mega-analyze all available transcriptomic data from studies of postmortem hippocampal tissue and of peripheral blood in AD to identify transcriptomic risk scores and machine-learning algorithms that maximally distinguish AD from cognitively normal control subjects, and scores and algorithms that then identify residual transcriptomic variation that offsets the transcriptomic risk in resilient controls. In Aim 3, we will identify an MRI-based structural brain signature that is associated with resilience to AD in the presence of an AD- associated cortical risk signature. Lastly, in our exploratory Aim 4, we will integrate genetic, transcriptomic, brain structural, and clinical data to identify biological relationships across Aims, and novel phenotypes of resilience. Collectively, these Aims will identify multivariate, genetic, transcriptomic, and brain-structural profiles of resilience to AD, as well as molecular, neurobiological, and clinical phenotypes stemming from AD- resilience genotypes.