Pharmacogenomics and Precision Psychiatry

Molecular medicine has opened new possibilities of personalized approaches in drug therapy. The development of evidence-based pharmacogenetic guidelines to steer therapy has slowly entered the field of psychiatric therapeutics. Some of the reasons behind the limited progress in psychiatric pharmacogenomics include the broad definition of clinical syndromes, limited knowledge of psychiatric pathophysiology, and limited understanding of psychotropics’ mechanisms of action. Pharmacogenomic markers have been reported for both pharmacodynamic and pharmacokinetic genes. However, only genetic variation in pharmacokinetic genes has shown to be helpful in clinical practice. There is little consensus as to when and if pharmacogenetic tests should be used in psychiatry. There are limited evidence-based dosing guidelines available for actionable gene–drug pairs. Future work in psychiatry may deepen our understanding of the biological underpinnings of psychiatric syndromes and provide the potential for individual tailored therapies.

Animal Models

Animal models of mental disorders are important for the development of new treatment and biomarkers. Animal models should satisfy three validity criteria. The choice of species used for animal models depends on the purposes of the study. There are several established models of chromosomal abnormalities for autism and schizophrenia. Stress-induced animal models of depression are controversial. There have been no established models for bipolar disorder, but the authors recently proposed a genetic animal model showing recurrent spontaneous depression-like episodes. Optogenetic manipulation of neural circuit is also used for modeling of mental disorders. Progress in psychiatric genetics will lead to the generation of valid animal models of psychiatric disorders.

Epigenetics in Psychiatry

Epigenetic modifications such as DNA methylation (DNAm), histone acetylation and methylation, and those directed by small RNAs, are widely studied in psychiatry and may play a role in the etiology and pathophysiology of psychiatric disorders. This chapter provides a brief overview of the mechanisms regulating these epigenetic marks and the challenges in obtaining biologically meaningful epigenetic data, given the inaccessibility of the living human brain. Significant results to date from studies on the epigenetics of psychiatric disorders are presented, including the impact of stress on DNAm in psychiatric risk genes such as FKBP5, and the impact of drugs of abuse and of psychiatric medications on histone modifications. Future directions are discussed, including the study of newly discovered aspects of DNAm: 5-hydroxymethylcytosine and non-CpG methylation. Ongoing work aims to uncover neurobiological mechanisms of illness and to find epigenetic biomarkers in peripheral tissues that inform diagnosis, prognosis, and therapeutic response.

Endophenotypes

Endophenotypes are traits that, while genetically related to an illness, are not used for diagnoses (e.g., a symptom). It is unlikely that specific genes directly code for any of our current psychiatric diagnoses. Rather, genes influence neurobiological processes that either increase or decrease risk for mental illness. One use of an endophenotype is to help characterize a genetic locus or gene previously identified as conferring risk for a particular illness. In this context, endophenotypes help to bridge the gap between a behavioral syndrome and molecular genetic variation. Alternately, endophenotypes can be used for novel locus or gene discovery, particularly when used in multivariate analyses. In this chapter, we define endophenotypes and describe different ways they have been applied to aid our understanding of the genetic architecture of psychiatric disorders.

The Role of Copy Number Variation in Psychiatric Disorders

Copy number variants (CNVs) are deletions, duplications, inversions, or translocations of large DNA segments. They can play a significant role in human disease. Thirteen CNVs have received strong statistical support for involvement in schizophrenia. They are all rare in cases (<1%), much rarer among controls, and have high odds ratios (ORs) for causing disease. The same CNVs also increase risk for autism spectrum disorders, developmental delay, and medical/physical comorbidities. The penetrance of these CNVs for any disorder is relatively high, ranging from 10% for 15q11.2 deletions to nearly 100% for deletions at 22q11.2. Strong selection pressure operates against carriers of these CNVs. Most of these are formed by non-allelic homologous recombination (NAHR), which leads to high mutation rates, thus maintaining the rates of these CNVs in the general population, despite the strong selection forces.

Human Linkage and Association Analysis

The basic idea in linkage analysis is that a disease gene will segregate in a family with a close (linked) marker, and typing this marker will lead to its detection. The successes using this approach have been largely confined to Mendelian monogenic disorders or complex disorders with Mendelian subforms. During the last decade, psychiatric genetics abandoned linkage analysis and moved to case-control studies of association, with remarkable success in identifying susceptibility genes for mental disorders. In this chapter, we review the statistical underpinnings of linkage and association and discuss important issues such as population stratification, imputation, data cleaning, the genomic inflation factor, and QQ and Manhattan plots. The challenge for the next decade will be to understand the biology of these GWAS (genome-wide association study) hits.

Statistical Genetics

Over more than the last decade, hypothesis-free genome-wide association studies (GWAS) have been widely used to detect genetic factors influencing phenotypes of interest. The basic principle of GWAS has been unchanged since the beginning: a series of univariate tests is conducted on all genetic variants available across the genome. We present study designs and commonly used methods for genome-wide studies, with a focus on the analysis of common variants. The basic concepts required for an application of GWAS in psychiatric genetics are introduced, from power calculation to meta-analysis. This chapter will help the reader in gaining the knowledge required for participation in and realization of GWAS of both qualitative and quantitative traits.

Biobanking of Human Induced Pluripotent Stem Cells for Psychiatric Research

The discovery that human primary cells such as nucleated blood cells or cultured skin fibroblasts can be reprogrammed into induced pluripotent stem cells (hiPSC) has ushered in a new era for research on the genetic etiology of neuropsychiatric disorders. Such hiPSC can be differentiated into several types of neurons, which may provide a primitive model for studying cellular variation in neuronal function due to underlying genetic variants causing the disorder. It is critical that source cells for possible reprogramming and their derived hiPSC be banked in an accredited facility capable of proper quality assurance that includes a genetic profile for future authentication of secondary biomaterials (e.g., differentiated cellular derivatives). Nucleated blood cells are more easily obtained compared to skin fibroblasts and can be cryopreserved for many years before they are reprogrammed to hiPSC. However, to enable all possible future uses of biosamples, some of which may not yet even be contemplated, researchers and biobanks must obtain clear informed consent from subjects for broad use of their biosamples in research.

Epigenomic Exploration of the Human Brain

The exploration of the epigenome has become a flourishing area in the neurosciences. Scientists increasingly appreciate that even the position of genetic material within the nucleus is purposeful, and its spatial orientation conveys information with critical influence on transcription, genome integrity, and stability. Together, epigenetic and three-dimensional genome data hold promise to reveal how DNA variants and mutations come into play in brain disease. Powerful new technologies can now map transcriptome, DNA-methylome, and other epigenetic regulators on the level of single brain cells. Many of these findings are limited to preclinical studies. Nevertheless, the advent of chromatin-modifying drugs in cancer therapy and the discovery that approved medications such as valproic acid and lithium have a chromatin-modifying effect have spurred hopes for improved biological therapies. Here we summarize current concepts and emerging insights into epigenetic regulation, with a focus on human brain and the neurobiology and pharmacology of psychiatric disorders.

Contribution of Epidemiology to Our Understanding of Psychiatric Genetics

The field of psychiatric epidemiology has advanced both methodological and substantive knowledge in our understanding of mental disorders through the following contributions: (1) development of standardized tools that operationalize diagnostic criteria in order to obtain reliable estimates; (2) estimation of the magnitude, correlates and service patterns of mental disorders; (3) documentation of patterns of comorbidity; (4) quantification of disability attributable to mental disorders; and (5) identification of risk and protective factors for mental disorders and their core domains. Community surveys using standardized tools for ascertaining psychiatric disorders have shown that mental disorders are highly prevalent in the general population. With the growing success in identifying genetic risk factors for chronic human disorders, the field of epidemiology will play an important role in defining study designs, appropriate samples, population generalizability, and statistical tools that will facilitate our ability to identify the joint influence of genetic and environmental factors on the susceptibility to mental disorders.