The Neurobiology of Schizophrenia

The Neurobiology of Schizophrenia begins with an overview of the various facets and levels of schizophrenia pathophysiology, ranging systematically from its genetic basis over changes in neurochemistry and electrophysiology to a systemic neural circuits level. When possible, the editors point out connections between the various systems. The editors also depict methods and research strategies used in the respective field. The individual backgrounds of the two editors promote a synthesis between basic neuroscience and clinical relevance.

• Provides a comprehensive overview of neurobiological aspects of schizophrenia
• Discusses schizophrenia at behavioral, cognitive, clinical, electrophysiological, molecular, and genetic levels
• Edited by a translational researcher and a psychiatrist to promote synthesis between basic neuroscience and clinical relevance
• Elucidates connections between the various systems depicted, when possible

• Language: English
• Publisher: Academic Press
• Release date: Jul 8, 2016
• ISBN: 9780128018774

Book preview

Tucker.

Part I

Introduction

Outline

Chapter 1 Historical and Clinical Overview

Chapter 1

Historical and Clinical Overview

Implications for Schizophrenia Research

T. Nickl-Jockschat¹,² and T. Abel³,⁴,    ¹Department of Psychiatry, Psychotherapy and Psychosomatics, RWTH Aachen University, Aachen, Germany,    ²Juelich-Aachen Research Alliance–Translational Brain Medicine, Juelich/Aachen, Germany,    ³Department of Biology, University of Pennsylvania, Philadelphia, PA, United States,    ⁴Institute for Translational Medicine and Therapeutics, University of Pennsylvania, Philadelphia, PA, United States

Abstract

Schizophrenia is not an etiologically defined disorder, but rather a syndrome. Diagnostic criteria of the two most widely used diagnostic systems, ICD-10 and DSM-V, are based on clinical symptoms and differ between them. Moreover, because of these two distinct classification systems, two patients can be diagnosed with schizophrenia even though they do not share a single symptom at the time of examination. This clinical heterogeneity poses serious challenges for the study of the neurobiological underpinnings of schizophrenia. In this introductory chapter, we briefly review historical definitions of schizophrenia and their influence on current diagnostic criteria. Therapeutic approaches, as well as basic research, have fueled new insight into the pathophysiology of the disorder. We also provide a short overview of the major developments in the history of neurobiological research of schizophrenia. Addressing the clinical heterogeneity found in schizophrenia will be a major challenge for future studies on the disorder.

Keywords

Diagnostic criteria; ICD-10; DSM-V; history of medicine; basic research

Chapter Outline

Introduction 3

A Brief History of the Definition of Schizophrenia 5

Lessons From Therapeutic Approaches 9

The Neurobiology of Schizophrenia 10

References 12

Introduction

Schizophrenia is a severe neuropsychiatric disorder that not only causes a high burden of disease but also challenges our understanding of how the mind and brain work. Certainties for healthy subjects (eg, that our thoughts or our actions are controlled by ourselves) are shattered for the schizophrenia patient. Consequently, a better understanding of the neurobiology of this disorder not only might help to identify better strategies for early diagnosis, therapy, and personalized medicine but also may answer some open questions about the biological correlates of our mind and consciousness.

Schizophrenia is defined as a syndrome. The diagnosis is based on a constellation of clinical symptoms and not on a common pathomechanism, as is the case for ischemic stroke or cardiac infarction. The two most widely used diagnostic systems, ICD-10 (International Statistical Classification of Diseases and Related Health Problems, WHO, 2010) and DSM-V (Diagnostic and Statistical Manual of Mental Disorders, American Psychiatric Association, 2013), both provide a catalog of symptoms and demand that a certain number of this pool must to be present over a given period of time for a diagnosis to be made (Table 1.1). It should be noted that because of these two classification systems, two patients can be diagnosed with schizophrenia even though they do not share a single symptom at the time of examination. According to DSM-V, a patient presenting with delusions and hallucinations can be diagnosed with schizophrenia, as can a patient presenting with disorganized speech and negative symptoms. Although both hypothetical patients arguably differ in their clinical presentations, they are both given the same diagnosis. In other words, schizophrenia can manifest with totally different clinical phenotypes.

Table 1.1

Diagnostic Criteria for Schizophrenia According to DSM-V and ICD-10

It should be noted that both diagnostic systems require additional tests, mainly to exclude secondary symptom genesis due to a somatic disorder.

It should be clear that this heterogeneity poses significant challenges for the identification of the neurobiological underpinnings of this disorder. If schizophrenia can present with strikingly distinct symptoms in different patients, then distinct pathophysiological mechanisms with specific etiological factors might be subsumed under the broad concept of schizophrenia (cf. Tandon et al., 2013). Consequently, the identification of causal factors might be severely hampered due to this heterogeneity.

To understand the current definition of schizophrenia, it seems inevitable that we must look back at its historical roots. The diagnosis criteria in both ICD-10 and DSM-V are the result of clinical and scientific developments spanning several decades and are founded on descriptions that were published in the first half of the twentieth century. Thus, when speaking of schizophrenia today, we do not refer to an etiologically defined disease but rather to a syndrome with a history of its own.

Here, we briefly summarize the developments that influenced the definitions of this disorder. We point out that we do not claim completeness with regard to medical history; rather, we aim to exemplify the importance—and the inherent challenges—that these definitions pose for current research into the neurobiological basis of schizophrenia.

A Brief History of the Definition of Schizophrenia

Since the beginning of psychiatry as a medical discipline, the syndrome called schizophrenia has been a major challenge for clinicians and scientists. Although schizophrenia-like symptoms have been reported by physicians of ancient Greece, we start our brief history of schizophrenia with the work of the German psychiatrist Wilhelm Griesinger (1817–68) (Fig. 1.1). Griesinger’s ideas were influential on a scientific level and a clinical level. In general, he demanded that clinical psychiatry should be based on empirical research rather than speculation, and he highlighted the need for neurobiological (specifically, neuroanatomical) studies. Given his predecessors who stemmed from psychiatry in the era of German Romanticism and, accordingly, emphasized the importance of individual passions and irrational impulses for psychopathology, this concept appeared revolutionary (Hoff and Theodoridou, 2008). The famous phrase attributed to Griesinger that mental illnesses are illnesses of the brain (Geisteskrankheiten sind Gehirnkrankheiten) is certainly an abbreviation and simplification of his ideas, but this phrase underlines his passionate attempt to further neurobiological research in psychiatry. With regard to schizophrenia, Griesinger formulated the concept of the unitary psychosis (Einheitspsychose) that is based on the assumption of a continuum between all psychiatric disorders. Griesinger proposed a model that starts with an early affective phase, followed by a delusional stage that finally leads to irreversible cognitive deficits (Griesinger, 1861). Given recent genetic studies that suggest a continuum between variants mediating vulnerability for schizophrenia and bipolar disorder, this idea of a continuum between psychiatric disorders appears surprisingly modern.

Figure 1.1 Wilhelm Griesinger (1817–68).

Distinct from Griesinger’s unitary psychosis, Emil Kraepelin (1856–1926) (Fig. 1.2) proclaimed a dichotomy of psychoses. He distinguished two major classes of psychotic disorders: dementia praecox (premature dementia)—corresponding to our current use of the term schizophrenia—and manisch-depressives Irreseyn (manic depression)—corresponding to the current range of affective disorders (Kraepelin, 1896). Remarkably, the long-term outcome and the course of the illness played a decisive role in Kraepelin’s classification. Broadly speaking, dementia praecox was thought to be associated with a poor prognosis, whereas manic depression was considered a comparatively benign disease. Kraepelin also explicitly acknowledged that individual symptoms did not distinguish between the two disease entities, but rather the pattern of symptoms leads to a distinction between these disorders. This basic Kraepelian dichotomy laid the foundation for the idea of distinct psychiatric disease entities and can be regarded as the first step toward our modern classification systems.

Figure 1.2 Emil Kraepelin (1856–1926).

Although the Kraepelian dichotomy allowed a valid distinction between two types of disorders based on their prognosis and their course, there was still a lively discussion about which symptoms were able to distinguish between schizophrenia and disorders with a better prognosis. The most important input at this time that still influences our concepts of schizophrenia today came from the Swiss psychiatrist Eugen Bleuler (1857–1939) and his German contemporary Kurt Schneider (1887–1967).

Bleuler (Fig. 1.3) was the one to coin the term schizophrenia or, more exactly, the group of schizophrenias (Bleuler, 1911). Before this, the Kraepelian term of dementia praecox was widely used. One of the reasons for Bleuler to shift the terminology away from the term dementia to the phrase group of schizophrenias was that he disagreed with Kraepelin in two important ways. Kraepelin treated schizophrenia as a single condition that was invariably incurable. Bleueler pointed out that some cases had a comparatively good prognosis and did not proceed into a dementia-like phenotype. Moreover, Kraepelin regards dementia praecox as a disorder that manifests in young people; therefore, he coined the term premature dementia. In contrast to this, Bleuler pointed out that the disease could also manifest at later ages (Fusar-Poli and Politi, 2008). This challenged the Kraepelian notion that distinguished dementia praecox, as its name insinuates, from other disorders based on its malignant progress of debilitating symptoms.

Figure 1.3 Eugen Bleuler (1857–1939).

Moreover, Beuler introduced diagnostic criteria for schizophrenia focused on fundamental symptoms that are regarded as core symptoms of the disorder and accessory symptoms that are often seen but are not necessary for diagnosis. His fundamental symptoms consisted of the loosening of associations (today referred to as formal thought disorder or disorganized speech), disturbances of affectivity, ambivalence, and autism (the latter to describe the social withdrawal of the patients and the first mentioning of the term autism in the medical literature). These have become famous as the four As (for an overview of Bleuler’s fundamental and accessory symptoms, please see Table 1.2). It is important to point out that Bleuler’s fundamental symptoms, which are central to his definition of the disorder, largely reflect what are today called negative symptoms, whereas positive symptoms, such as delusions or hallucinations, are missing from Bleuler’s description.

Table 1.2

Diagnostic Criteria for Schizophrenia (the Four As) According to Eugen Bleuler

Similar to Bleuler’s fundamental and accessory symptoms, Kurt Schneider (Fig. 1.4) distinguished between first-rank (core) and second-rank (less important) symptoms; however, in contrast to Bleuler’s approach, Schneider’s first-rank symptoms are closely related to positive symptoms. In this classification system, auditory hallucinations per se—and especially imperative voices—were not first-rank symptoms. Only auditory hallucinations of a dialoguing or commenting quality were viewed as first-rank symptoms. Schneider characterized two or more voices that are talking about the patient like a dialogue as dialoguing, whereas voices that are commenting on the actions of the patient were labeled as commenting. Many of his first-rank symptoms are centered around his idea of Ich-Störung (in the English language usually only referred to as a subgroup of delusions): bizarre phenomena such as thought withdrawal (the idea that thoughts are taken away from the patient’s mind), thought insertion (the intrusion of thoughts into the patient’s mind, usually believed to be caused by another person), thought dispersion or broadcast (the idea that the thoughts of the patient can be read or heard by other people), and passivity experiences or delusions of control (over the mind or body of the patient). Moreover, Schneider highlighted the importance of delusional perceptions, which are normal sensory perceptions linked to a bizarre, delusional conclusion by the patient (eg, seeing a STOP traffic sign leads the patient to the believe that this signals his or her own near death) (Schneider, 1950) (for an overview of Schneider’s first- and second-rank symptoms, please see Table 1.3).

Figure 1.4 Kurt Schneider (1887–1967).

Table 1.3

First- and Second-Rank Symptoms of Schizophrenia According to Kurt Schneider

Kurt Schneider’s diagnostic concept has exerted its influence over the years. Three out of the first four diagnostic criteria of the ICD-10 directly refer to first-rank symptoms, and only the fourth criterion—bizarre delusional beliefs—is a second-rank symptom. Although the authors certainly realize that current diagnostic systems are the result of decade-long discussions rather than the simple continuation of old concepts, they think that Schneider’s impact on the current concepts in ICD-10 might serve well as an example of the importance of historical definitions of schizophrenia in other contemporary classification systems. Despite decades of discussions between basic researchers and clinicians, one of the two most important diagnostic systems still is largely based on Schneider’s first-rank criteria.

It is important in this context that the two major classification systems—ICD-10 and DSM-V—that are used in research and in clinics do not use the same criteria for the diagnosis of schizophrenia. One of the differences most easily recognizable is the time during which symptoms need to be present: 1 month according to ICD-10, but 6 months in most cases according to DSM-V (WHO, 2010; American Psychiatric Association, 2013). Consequently, these differences define at least the fringes of the patient pools included in the respective diagnosis and thus might lead to different conclusions with regard to the neurobiological underpinnings of this complex disorder.

Lessons From Therapeutic Approaches

On January 23, 1934, the Hungarian psychiatrist Ladislas Meduna (1896–1964) used a camphor solution to induce a generalized seizure in a schizophrenia patient. As reported in Meduna’s autobiography, the patient, 33-year-old Zsoltan L., achieved full remission after 4 years of illness and was discharged from the hospital (Meduna, 1985). Although other sources questioned the therapeutic success, as described by Meduna (Baran et al., 2008), first chemoconvulsive therapy, and later its modified version of electroconvulsive therapy (Bini, 1937) became the first effective biological treatment strategy that is still used today, although admittedly rarely for schizophrenia (cf. Loh et al., 2013).

With the discovery of chlorpromazine as the first antipsychotic drug in 1952, treatment strategies were revolutionized. Chlorpromazine was initially used as a sedative agent to enhance anesthesia. However, due to its sedating potential, it was believed to have a potential role in the treatment of psychiatric patients. On January 19, 1952, it was first administered to a patient with acute mania. Remarkably, this patient was discharged after 3 weeks of treatment (López-Muñoz et al., 2005). Following this remarkable success, the first clinical trial followed the same year (Delay and Deniker, 1955). For the first time in medical history, drugs existed that were easy to administer and had the potential to cure a disease that was previously regarded as incurable (López-Muñoz et al., 2005). Although the advent of atypical antipsychotics helped to minimize some of the extrapyramidal motor side effects, the basic molecular mechanism was still antidopaminergic (Gründer et al., 2009). With regard to our topic of neurobiological schizophrenia research, the use of antipsychotics led to important conclusions and helped to point out the dopaminergic system as a key player in schizophrenia pathophysiology (Carlsson, 2006). However, if dopamingergic antagonism alone would explain the antipsychotic properties, then a given patient should be fully remitted minutes after intravenous application of haloperidole, when the receptor occupancy needed for the antipsychotic effect is reached. The fact that a therapeutic response takes much longer points to the idea that other mechanisms are potentially involved.

The use of antipsychotics remains the gold standard in schizophrenia treatment (American Psychiatric Association, 2013). However, despite their overall efficacy, some patients benefit earlier and more completely from antipsychotic treatment than others, and a subgroup does not respond sufficiently to pharmacological treatment (Hasan et al., 2012). Consequently, new therapeutic strategies are needed. However, only a better understanding of schizophrenia pathophysiology can help to discover new treatment approaches. Therefore, a better understanding of the underlying neurobiological pathomechanisms not only will help to shed more light on basic functions of our minds and brains but also will help to develop new treatment strategies for this severe mental disorder.

The Neurobiology of Schizophrenia

The idea that schizophrenia was a biological disorder is nearly as old as the first depiction of its symptoms. Hippocrates tried to find physical causes for psychiatric disorders, including those with schizophrenia-like symptoms, but he suspected that imbalances of body fluids, not pathologies of the brain, caused the symptoms.

More than two millennia later, the improved clinical characterization of psychiatric disorders that were previously labeled as insanity without any further distinction also sparked new interest and new approaches in neurobiological research. As depicted, Wilhelm Kraepelin was the first to distinguish schizophrenia (or, in his words, dementia praecox) from other psychiatric disorders. Remarkably, his clinical ideas were compatible with increased scientific efforts to unravel the pathophysiological mechanisms underlying these diseases. Kraepelin took the initiative to found the Deutsche Forschungsanstalt für Psychiatrie (German Research Institute of Psychiatry), a predecessor of the Max Planck Institute for Psychiatry. Under this framework, scientists including Alois Alzheimer (1864–1915), Franz Nissl (1860–1919), and Robert Gaupp (1870–1953) researched the genetics and neuropathology of schizophrenia.

The next important revolution in neurobiological research came with the introduction of antipsychotics as a treatment strategy. As depicted, antipsychotic medication has directed the attention toward distinct neurotransmitter systems early and has helped to unravel some of the molecular changes in schizophrenia patients.

Although neuroanatomical research on the brains of schizophrenia patients has already been systematically conducted in Kraepelin’s research institute, several important obstacles prevented the researchers from a robust characterization of structural changes in schizophrenia. As we know today, brain structure changes are subtle and inter-individual variance among patients is high. Consequently, there is a need for large groups in such studies, but these are difficult to acquire for a postmortem study. The introduction of modern imaging techniques, especially magnetic resonance imaging (MRI), allowed the inclusion of cohorts with sufficient sample sizes to counter that problem (Shenton et al., 2001). They also yielded the great advantage of longitudinal study designs that helped to characterize the temporal dynamics of these brain structure changes. In addition to structural imaging, functional imaging techniques have led to important breakthroughs. They have provided the unique chance to study functional changes in the brains of schizophrenia patients. Currently, functional MRI (fMRI) is a widely used method to characterize neural correlates of schizophrenia symptoms. The acquisition of large cohorts, often by collaborative approaches, has significantly increased the number of individuals available for a given analysis and, therefore, helped to solidify the reliability of their results.

It has been known since the beginning of the twentieth century that schizophrenia is a disorder with high heritability. Consequently, the identification of susceptibility variants has been a major effort in psychiatric genetics. However, most common variants in schizophrenia have rather low odds ratios for the actual manifestation of the disease. It took the pooling of large cohorts to achieve the robust identification of genetic variants with genome-wide significance. Modern genetics has also shifted the focus to large deletions or duplications of the genome, so-called copy number variations (CNVs). In contrast to more common variants, some CNVs show rather high odds ratios for schizophrenia and other neuropsychiatric disorders. Given this, the biological characterization of CNVs seems to be a promising approach to understanding the molecular underpinnings of this disorder.

With the recent major advances in psychiatric genetics and the robust identification of schizophrenia susceptibility genes, genetic animal models have gained increasing importance. These models yield several great advantages. Besides a homogeneous genetic background, animals can also be raised under standardized conditions. Consequently, environmental factors can be expected to play only a minor role in shaping the behavioral phenotype. They are also perfect models to study molecular and electrophysiological changes due to the modeled genetic variants. As depicted, there is considerable clinical heterogeneity among schizophrenia cases. Due to DSM-V and ICD-10, these heterogeneous cases are summarized under one diagnostic label. Consequently, considerable heterogeneity of the underlying pathophysiological processes might obscure the identification of the neurobiological underpinnings. One potential option to address this dilemma is the use of newly developed classification schemes that are based on dimensions of observable behavior and neurobiological measures rather than an artificial grouping of heterogeneous syndromes with different pathophysiological mechanisms into one disorder (Wong et al., 2010; Ford et al., 2014). Because of these considerations, the National Institute of Mental Health (NIMH) has developed so-called research domain criteria (RDoCs) that treat psychopathological symptoms as situated within a continuum that reaches from normal functioning to pathological extremes. This is intended to minimize biological heterogeneity within the studied cases and might be a decisive step toward the identification of neurobiological foundations of schizophrenia.

Our aim with this textbook is to provide an overview of the current state of this lively field of research. Therefore, we focus on behavioral, cognitive, clinical, electrophysiological, molecular, and genetic levels.

As depicted, schizophrenia is a group of complex and potentially pathophysiologically heterogeneous disorders with high heritability involved with pronounced changes in brain structure and function. The symptoms are complex and sometimes bizarre, involving disturbances of a multitude of cognitive and affective domains. Despite significant progress in various areas of research, the etiopathogenesis of these complex disorders is still not fully understood. The intricate interplay between various susceptibility genes and structural alterations in neuronal connectivity along with changes at epigenetic and neurochemical levels make the neurobiology of schizophrenia a complex topic. Therefore, we have organized and focused discussions on critical areas of research.

Given the complex nature of schizophrenia neurobiology, various professions are involved in research, ranging from molecular biologists to physicians and psychologists. However, students of the respective disciplines are often confronted with—and confused by—the puzzling riddle of this disorder. We explicitly aimed for our textbook to be easily comprehensible for this heterogeneous audience. Therefore, we include chapters briefly describing basic methods and techniques reported in the section. We hope that this approach makes it easy for MD and PhD students to introduce themselves to this field of research and help to spark new scientific approaches for advanced researchers.

References

1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders fifth ed. Washington, DC: Author; 2013.

2. Baran B, Bitter I, Ungvari GS, Nagy Z, Gazdag G. The beginnings of modern psychiatric treatment in Europe Lessons from an early account of convulsive therapy. Eur Arch Psychiatry Clin Neurosci. 2008;25:434–440.

3. Bini, L., 1937. Richerche sperimentali nell’accesso epilettico da corrente elettrica. In: Die Therapie der Schizophrenie: Insulinschock, Cardiazol, Dauerschlaf. Bericht über die wissenschaftlichen Verhandlungen auf der 89. Versammlung der Schweizerischen Gesellschaft für Psychiatrie in Münsingen b. Bern am 29.-31. Mai 1937. Schweiz Arch Neurol Psychiatr, 39 (Suppl.), pp. 121–122.

4. Bleuler E. Dementia praecox oder Gruppe der Schizophrenien. In: Aschaffenburg G, ed. Handbuch der Psychiatrie Spezieller Teil, 4 Abteilung, 1 Hälfte. Leipzig: Deuticke; 1911.

5. Carlsson A. The neurochemical circuitry of schizophrenia. Pharmacopsychiatry. 2006;39(Suppl. 1):S10–S14.

6. Delay J, Deniker P. Neuroleptic effects of chlorpromazine in therapeutics of neuropsychiatry. J Clin Exp Psychopathol. 1955;16(2):104–112.

7. Ford JM, Morris SE, Hoffman RE, et al. Studying hallucinations within the NIMH RDoC framework. Schizophr Bull. 2014;40(Suppl. 4):S295–S304.

8. Fusar-Poli P, Politi P. Paul Eugen Bleuler and the birth of schizophrenia (1908). Am J Psychiatry. 2008;165(11):1407.

9. Griesinger W. Die Pathologie und Therapie der psychischen Krankheiten Stuttgart: Krabbe; 1861.

10. Gründer G, Hippius H, Carlsson A. The atypicality of antipsychotics: a concept re-examined and re-defined. Nat Rev Drug Discov. 2009;8(3):197–202.

11. Hasan A, Falkai P, Wobrock T, et al. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for biological treatment of schizophrenia, part 1: update 2012 on the acute treatment of schizophrenia and the management of treatment resistance. World J Biol Psychiatry. 2012;13(5):318–378.

12. Hoff P, Theodoridou A. Schizophrene Psychosen im Spannungsfeld von Kognition, Affekt und Volition – Die psychiatriehistorische Perspektive. In: Kircher T, Gauggel S, eds. Neuropsychologie der Schizophrenie. Heidelberg: Springer; 2008.

13. Kraepelin E. Psychiatrie Ein Lehrbuch für Studierende und Aerzte Leipzig: Barth; 1896.

14. Loh N, Nickl-Jockschat T, Sheldrick AJ, Grözinger M. Accessibility, standards and challenges of electroconvulsive therapy in Western industrialized countries: a German example. World J Biol Psychiatry. 2013;14(6):432–440.

15. López-Muñoz F, Alamo C, Cuenca E, Shen WW, Clervoy P, Rubio G. History of the discovery and clinical introduction of chlorpromazine. Ann Clin Psychiatry. 2005;17(3):113–135.

16. Meduna L. Autobiography. Convuls Ther. 1985;1(43–57):121–135.

17. Schneider K. Klinische Psychopathologie Stuttgart: Thieme; 1950.

18. Shenton ME, Dickey CC, Frumin M, McCarley RW, et al. A review of MRI findings in schizophrenia. Schizophr Res. 2001;49:1–52.

19. Tandon R, Gaebel W, Barch DM, et al. Definition and description of schizophrenia in the DSM-5. Schizophr Res. 2013;150(1):3–10.

20. Wong EH, Yocca F, Smith MA, Lee CM. Challenges and opportunities for drug discovery in psychiatric disorders: the drug hunters’ perspective. Int J Neuropsychopharmacol. 2010;13(9):1269–1284.

21. World Health Organization, 2010. International Statistical Classification of Diseases and Related Health Problems. <http://www.who.int/classifications/icd/en/bluebook.pdf?ua=1> (accessed 07.03.16.).

Part II

The Genetic and Epigenetic Basis of Schizophrenia

Outline

Chapter 2 Progress and Prospects for Endophenotypes for Schizophrenia in the Time of Genomics, Epigenetics, Oscillatory Brain Dynamics, and the Research Domain Criteria

Chapter 3 Insights From Genome-Wide Association Studies (GWAS)

Chapter 4 Sequencing Approaches to Map Genes Linked to Schizophrenia

Chapter 5 Epigenetic Approaches to Define the Molecular and Genetic Risk Architectures of Schizophrenia

Chapter 6 Exploring Neurogenomics of Schizophrenia With Allen Institute for Brain Science Resources

Chapter 2

Progress and Prospects for Endophenotypes for Schizophrenia in the Time of Genomics, Epigenetics, Oscillatory Brain Dynamics, and the Research Domain Criteria

G.A. Miller¹ and B.S. Rockstroh²,    ¹Department of Psychology and Department of Psychiatry and Biobehavioral Sciences, University of California, Los Angeles, CA, United States,    ²Department of Psychology, University of Konstanz, Konstanz, Germany

Abstract

Endophenotypes are biological or psychological phenomena of a disorder believed to be in the causal chain between genetic contributions to a disorder and diagnosable symptoms of psychopathology. Decades of psychophysiological research have identified a variety of established and candidate endophenotypes for schizophrenia without offering much insight into genetic contributions to the disorder and without substantially improving the understanding of the complex psychopathology. This chapter surveys issues surrounding the pursuit of endophenotypes for schizophrenia and some of the findings being established. Diverse lines of research not only complement each other in indicating fundamental characteristics of psychopathology (such as dimensional aspects) but also suggest that progress in understanding the nature and origins of schizophrenia would benefit from a new perspective on psychopathology research. The NIMH Research Domain Criteria initiative offers a promising framework for identifying factors contributing to schizophrenia.

Keywords

Endophenotypes; psychopathology; genes; epigenetics; Research Domain Criteria; brain connectivity; chronnectomics; schizophrenia

Chapter Outline

Endophenotypes in Schizophrenia: Current Status 22

An Emerging Frontier for Endophenotypes: Epigenetics 26

A New Frontier for Endophenotypes: Brain Connectivity Dynamics 28

Limitations and Future Directions 30

References 31

Endophenotypes are defined as biological or psychological phenomena of a disorder believed to be in the causal chain between genetic contributions to a disorder and diagnosable symptoms of psychopathology (Glahn et al., 2014; Gould and Gottesman, 2006; Gottesman and Shields, 1972, 1973; Gottesman and Gould, 2003; Iacono and Lykken, 1979; Lenzenweger, 2010, 2013; Miller and Rockstroh, 2013; Pearlson, 2015). Endophenotypes can be biological or psychological, as long as there is a foreseeable causal path from the gene through endophenotype(s) to phenotype (clinical expression). The endophenotype literature frames the causal paths to illness as beginning with genes (rather than with environments—social, economic, physical—or in terms of a larger temporal continuum along which individuals are manifestations of slower processes), which is an arguable starting point (eg, Schumann, 2014). But the identification of endophenotypes should facilitate the identification of genetic contributions to the psychiatric disorder and the causal mechanisms between them.

This chapter identifies and examines several conceptual issues concerning endophenotypes in schizophrenia research and in psychopathology research more generally. A selective review of recent endophenotype research in schizophrenia is enriched with some insights from genomics, epigenetics, and oscillatory brain dynamics to enhance the goal of understanding mechanisms of schizophrenia pathophysiology and psychopathology. The potential of adopting the Research Domain Criteria (RDoC) in pursuit of this goal is briefly discussed. Profound logical and conceptual issues remain to be resolved concerning the causal and other relationships between psychological and biological phenomena (Berenbaum, 2013; Lilienfeld, 2007, 2014; Miller, 1996, 2010; Miller and Keller, 2000; Miller and Rockstroh, 2013), but the empirical literature is nonetheless advancing.

If one assumes that there is a causal path between genes and manifest psychopathology and that that the path has multiple stages (whether or not there are multiple paths), then one believes that there are endophenotypes, which are the stages along that path. Fig. 2.1 illustrates the simplest possible example of such a model, which comprises essentially what is known in the statistical literature and is known as a mediation model. The concept is that at least some of the causal force of genetic input works via an intermediate step, the endophenotype. The figure conveys how the endophenotype may be closer, causally, to both genetic input and clinical manifestation.

Figure 2.1 An endophenotype as an interim step in the causal chain between genes and observable clinical manifestation.

The extent to which one should target endophenotypes is a strategic question for the psychopathology research literature. Much of the rationale in the endophenotype literature is that we can make progress by dissecting and understanding small spans along the causal path. One can focus on the mechanisms by which genes lead to endophenotypes or the mechanisms by which endophenotypes lead to a clinically significant disorder. One might treat some endophenotypes as milder versions of the disorder (such as was proposed for schizotypy regarding schizophrenia) or as fundamentally nonclinical features of one’s personality (eg, low levels of behavioral inhibition), one’s biochemistry (eg, aberrant DNA methylation), or other individual differences that in certain circumstances can foster the development or recurrence of disorder.

Nothing about the concept of endophenotype diminishes the role of environment, nor does it entail an assumption that genetic contributors are more important than environmental contributions. The assumptions are merely that there are genetic contributors and that it will prove very fruitful to identify their sequelae through the causal chain(s). As discussed, it is now widely understood that genes alone are likely to play only a limited role in most forms of psychopathology, that the genetic contributions are likely to be numerous and complex, and that the relationships of genetic contributions to environmental contributions are also likely to be complex. There is no doubt that environment will be a critical part of the schizophrenia story.

The endophenotype concept was introduced to the psychopathology literature more than 40 years ago (Gottesman and Shields, 1972), which had little impact until an invitation to Gottesman 30 years later to revisit the concept (Gottesman and Gould, 2003). In the schizophrenia literature, a substantial body of evidence has evaluated potential endophenotypes following the criteria of Gottesman (discussed at length in Lenzenweger, 2010 and in Miller and Rockstroh, 2013). Table 2.1 lists the criteria proposed by Gottesman and colleagues. Table 2.2 cites some endophenotypes proposed for schizophrenia as meeting some, and in some cases all, of the Gottesman criteria.

Table 2.1

Criteria for Endophenotypes

Source: Proposed by Gottesman and Shields (1972). See also Hasler (2006), Lenzenweger (2010), and Ritsner and Gottesman (2011).

Table 2.2

Commonly Proposed Endophenotypes for Schizophrenia

Examples (not comprehensive) updated from review by Miller and Rockstroh (2013).

In the past decade, the potential of endophenotypes has been an active subject of debate (eg, Baker, 2014; Braff, 2014; Ford, 2014; Goldman, 2014; Iacono, 2014; Iacono et al., 2014; Miller et al., 2014; Miller et al., 2016; Munafò and Flint, 2014; Patrick, 2014; Schumann, 2014; Wilhelmsen, 2014; Yee et al., 2015). A common misunderstanding is that endophenotypes must be biological phenomena. In fact, interest in psychological endophenotypes reflects the growing recognition that cognitive deficits are among the most important symptoms of schizophrenia in terms of impact on society and on individual quality of life. Accordingly, efforts are growing to develop and evaluate psychological interventions for specific phenomena associated with schizophrenia, especially the cognitive remediation of performance deficits. An important reason for this shift is that pharmacological treatment, emphasizing biological features of psychopathology, has not delivered on its promise to alleviate or substantially improve cognitive decline and negative symptoms, and, if anything, prospects have declined in recent years (Keefe and Harvey, 2012; Yee et al., 2015).

Intellectually, this literature is something of an antidote to the nearly exclusive focus on genetics and pharmacology that has dominated the schizophrenia treatment literature in recent decades. That tradition reflected a widely shared but naively reductionist view of mental illness that overemphasizes the clearly sizable contribution of genetic and other biological factors at the expense of psychological factors and gene ×environment (G ×E) interactions, correlations, and other joint relationships. Fig. 2.2 illustrates this much richer but much more challenging picture now taking shape. In addition to showing that an endophenotype may have far more complex inputs, it shows that both endophenotype and clinical manifestation may feed back into those inputs, altering their contribution over time. Such feedback loops over time are surely crucial to the prevention, development, remission, resurgence, and treatment of schizophrenia.

Figure 2.2 Numerous inputs to an endophenotype and feedback pathways over time. G, gene; E, environment.

The overemphasis on genetics and pharmacology in the research agenda for schizophrenia and other severe mental illness has been undermined by the lack of progress in pharmacological treatments of major mental illness in recent decades, with no major advances, few prospects, and no impact on some critical areas of clinical manifestation, specifically negative symptoms (such as flat affect and poverty of speech) and social and cognitive function. It has come to be appreciated that treating the positive symptoms (such as delusions and hallucinations), for which medications are often beneficial (although often with substantial side effects), is far from adequate if we are to help patients beyond reducing positive symptoms, including how well they perform at work and in social functions in daily life.

In response to the growing realization that heavy reliance on pharmacological methods has been unsuccessful, a variety of promising psychological interventions are now undergoing development, targeting sensory, cognitive, and social deficits in schizophrenia (and other mental disorders) (eg, Cuthbert and Instel, 2013; Crocker et al., 2013; Miller, 2010; Insel and Cuthbert, 2015). Furthermore, psychological symptoms have increasingly come to be appreciated as potentially valuable outcome variables for evaluating pharmacological interventions. We have come to realize that positive symptoms are not the primary barrier to effective function in the daily lives of people with schizophrenia. We need ways of improving their cognitive and social functions, which phenomenologically may appear as or contribute to negative symptoms, as critical factors in vocational performance and quality of life. Thus, there is some return of appreciation of the fundamentally psychological nature of schizophrenia. However, the field is not where it needs to be in terms of the appropriate technologies for measuring cognitive and social functions. Some of the sensory, cognitive, and social targets in the recent psychological treatment literature may qualify as endophenotypes. In fact, approaching endophenotypes as intervention targets, rather than limiting interventions to monolithic disorders, is appealing, especially when it may be beneficial to intervene before the full clinical picture develops.

Endophenotypes in Schizophrenia: Current Status

Miller and Rockstroh (2013) summarized research on endophenotypes in schizophrenia (see also Table 2.2), emphasizing a significant role for executive functions, several psychophysiological phenomena including functional brain (event-related and oscillatory) measures, and the overlap in phenomena meeting Gottesman’s criteria for an endophenotype between schizophrenia and psychotic bipolar disorder. Some endophenotypes span traditional diagnostic categories, as conceived in the RDoC initiative (Kozak and Cuthbert, 2016; Miller et al., 2016).

A literature search in December 2014 (key words endophenotype, schizophrenia) confirmed that despite the overwhelming research on psychiatric genetics and some skepticism about the limited utility of endophenotypes (eg, Patrick, 2014), interest in endophenotypes in schizophrenia is unbowed. Endophenotype research is alive and prospering: for 2014, PubMed identified almost 80 publications. Moreover, results of consortia of researchers providing data from very large samples (eg, B-SNIP, COGS-1, COGS-2) offer an increasingly solid basis for optimism.

Many of the promising endophenotypes for schizophrenia are biological (in most cases psychophysiological), and several are psychological. Interestingly, overt performance measures appear to tap variance separate from those of individual biological measures. Recent factor analyses by Seidman et al. (2015) of data from 83 schizophrenia patients, 151 unaffected siblings, and 209 community comparison participants yielded five distinct cognitive factors (episodic memory, working memory, perceptual vigilance, inhibitory processing, and visual abstraction) with low association with neurophysiological measures. This divergence suggests that both classes of phenomena will be important for advancing explanatory stories about schizophrenia. Importantly, many such studies have evaluated a given potential endophenotype for schizophrenia with samples confined to a schizophrenia diagnosis (perhaps including patients’ relatives), so conclusions about the diagnostic or dimensional specificity of most endophenotypes often cannot yet be drawn. The need for such specificity recedes if one pursues a transdiagnostic approach to psychopathology, as the United States National Institute of Mental Health now advocates in its RoDC initiative, but the need for evaluation across diagnostic categories becomes even clearer.

Ambitious recent studies have delivered on the analytic approach pursued by Glahn and colleagues (eg, Glahn et al., 2012, 2015), defining endophenotypes in very large samples randomly selected from extended pedigrees. Results are important for confirming cognitive and macrolevel cortical endophenotypes grounded in significant associations between these measures and genetic liability for schizophrenia: digit-symbol substitution, facial memory, emotion recognition, and temporal and prefrontal cortical surface abnormality.

In parallel with traditional endophenotype research, psychiatric genetics, represented by large-scale genome-wide association studies (GWAS), is beginning to identify replicated genetic patterns in schizophrenia. Regarding the vast and diverse literature, some major insights from GWAS