The Behavioral Genetics of Psychopathology

A Clinical Guide

Kerry L. Jang

• Publisher: Routledge
• Year: 2005
• eBook ISBN: 9781135630713
• Topic: Biological Sciences
• Subtopic: Mental Health in Psychology
• Index: Biological Sciences

New discoveries about the genetic underpinnings of many kinds of human experience are now continually being made. This book explores the impact of these discoveries on the ways in which the common mental disorders are best conceptualized and treated.

Most people think of research in genetics as the search for genes. This is only one focus of effort, and even with the reliable identification of susceptibility genes, the clinical applications of their discovery, such as gene therapies and new drug development, are a long way off. For the present, the impact of genetic research on our understanding of mental illness is tied to our ability to estimate the effect of all genes by means of family, twin, and adoption studies. The results of these studies challenge some deeply cherished ideas and theories, and support others.

Of course, the effect of genes is only half the equation. The role of experience, environment, and living conditions accounts for as much, often considerably more, of the variability in psychopathology. In this book, Kerry Jang attempts not to answer questions about what is "genetic" and what is not, but about what a knowledge of the relative influence of genes versus environment means at a psychological level of analysis--to show how it changes common assumptions about classification, etiology, diagnosis, and intervention.

He first offers an overview of contemporary behavioral genetics, dispels common misconceptions, responds to the criticisms that have been leveled at this new field, and describes in basic terms how genetic and environmental effects are estimated and how susceptibility genes are pinpointed. He then points to new directions in which standard nosological systems are likely to evolve as new information about vulnerabilities and covariances emerges. Finally, he synthesizes and evaluates the consistency of the last decade's findings for the most common categories of psychopathology that have been studied by behavior geneticists: mood, personality, and anxiety disorders, substance abuse; and schizophrenia and the psychotic disorders.

Clinicians and researchers alike need to understand the genetic influences on the feelings and behaviors they are seeking to change or study if they are to be effective in their work. The Behavioral Genetics of Psychopathology: A Clinical Guide empowers them with this understanding.

II
The Behavioral Genetics of the Common Mental Disorders

Chapter 4
The Mood Disorders

The prevalence of depression has earned the disorder the title of the common cold of psychopathology.

—Dunn, Sham, and Hand (1993)

The mood disorders are among the most heavily researched in behavioral genetics. Heritability studies have shown that the magnitude of genetic and environmental effects varies considerably—from 0% to 70%, depending on the definition. The molecular genetic research is equally broad and varied. It has investigated a wide range of genes with mixed results: from genes known to control neurotransmitter and hormone systems to genes that have no known function.

Typically, inconsistency in results is something researchers dread. However, the variability provides the backdrop for an important theme in behavioral genetics— linking this disparity to differences in how and on whom depression was measured to address key questions about the disorder, such as: “What forms of depression are heritable?” Are specific symptoms differentially heritable? Is a general liability to depression inherited? Is depression in females the same as in males? What is the relationship between different forms of depression, such as bipolar and unipolar? The outcome of the molecular and behavioral genetic research on these questions has resulted in a shift in the understanding of depression from a broad and monolithic disorder to a collection of individual symptoms that vary in severity and aetiology. The notion that depression may not be inherited as a unitary disorder is highlighted by the frequently contradictory results from molecular genetic studies.

IDENTIFYING SUSCEPTIBILITY GENES FOR DEPRESSION

Neurotransmitter Studies

The serotoninergic system has received a great deal of attention because clinical studies have shown that transport of this neurotransmitter is significantly lower in patients diagnosed with major depressive disorder (MDD or “unipolar depression”; see also Cowen, 1993, for a review). Serotonin has also been implicated in related forms of depression, such as seasonal affective disorder (SAD). For example, positron emission tomography (PET) studies have shown reduced levels of the serotonin transporter and low production of tryptophan hydroxylase, an amino acid precursor of the synthesis of serotonin in SAD patients (e.g., Willeit et al., 2000).

The gene that controls the transport of serotonin is the serotonin transporter repeat-length polymorphism, or 5-HTTLPR, gene. The gene comes in either a short (s) or a long (l) form. Each parent contributes one of these genes that would yield one of three possible genotypes: s/s, l/l, or s/l. Early association studies suggested that the short form of the allele was the putative disease gene. For example, Collier, Arranz, Sham, and Battersby (1996) showed that frequency of genotypes containing the short form of the gene was elevated in a large sample of 454 patients diagnosed with bipolar or unipolar depression, compared to 570 healthy controls. However, this elevation in scores was not quite statistically significant. Recently, Geijer et al. (2000) replicated this lack of association. They did not find any differences in the frequency of 5-HTTPLR polymorphisms in suicide attempters diagnosed with unipolar depression compared to healthy normal controls. They also found no differences in the gene frequencies for the serotonin receptor 2A or the tryptophan hydroxylase gene.

Rosenthal et al. (1998) reported that the frequency of the short allele was higher (44.8%) in a sample of 97 patients with SAD compared to a sample of 71 healthy controls (32.4%). This is important because SAD is defined as a variant of recurrent MDD whose essential feature is the onset and remission of major depressive episodes at characteristic times of the year (DSM-IV, 1994). Unfortunately, several studies have failed to replicate the association. Lenzinger et al. (1999) carefully matched 18 drug-naïve (people who have not received medication for their condition) SAD patients with healthy controls and found no association. This finding was also replicated using samples drawn from Sweden, Finland, and Germany (Johansson et al., 2001).

The frustration that characterizes research on serotonin also applies to the research on the other major neurotransmitters. Neither the genes controlling the transport nor those controlling the reception of dopamine have been consistently associated with bipolar depression (e.g., Byerley, Hoff, Holik, & Coon, 1994; Holmes, Brynjolfsson, Brett, & Curtis, 1991; Serretti et al., 1999). Similarly, few significant associations between the peripheral benzodiazapine receptor gene and bipolar or other depressive disorders (e.g., Kurumaji, Nomoto, Yamada, Yoshikawa, & Toru, 2001) or differences in monoamine oxidase gene variants (e.g., Syagailo et al., 2001) have been found.

Hormones, Proteins, and Depression

Another strategy has been to examine other biochemical systems for possible candidates. Recent research on age-related changes in hormone systems has yielded some positive associations worth investigating. Seidman, Araujo, Roose, and McKinlay (2001) focused on the androgen receptor gene because of clinical findings that depressive symptomology increases as levels of age-related testosterone levels decrease. In a sample of 1,000 men, 110 were classified as depressed using a cutoff score on the Center for Epidemiologic Studies Depression Scale (CES-D: Radloff, 1977), a popular self-report scale of depressive symptomology. All subjects had testosterone levels measured and were genotyped for the repeat length of the CAG gene, a marker associated with androgen receptor function. They found that depressive symptomology was significantly and inversely associated with total testosterone levels in men with shorter CAG repeat lengths, but not in men with moderate and longer repeat lengths.

Another age-related target of research is the Apolipoprotein-E e4 allele (APOE). Although this gene is best known as a risk marker for Alzheimer’s disease, it is thought to be important in depression because depression is one of the diagnostic criteria for dementia. Several positive associations have been reported between the APOE allele and late-onset depression (e.g., Steffens et al., 1997). More recently, Stewart, Russ, and Richards (2001) found that the APOE allele was present in 69% of subjects who display subjective memory impairments with depression, as opposed to only 28% of subjects with either depression or impairments in subjective memory. Despite these positive associations, Mauricio et al. (2000) found no association between the APOE allele and changes in depression scores in a sample of 113 seniors who were followed longitudinally for 5 years.

Other genes under active study are those that mediate the immune system response (e.g., activity of natural killer cells, antibody production, T-cell activation) and that are suspected to differ between patients diagnosed with major depression and other populations (see Maes, Meltzer, Scharpe, & Bosmans, 1993, for a review). On a more positive note, a research trend producing fascinating and clinically relevant results does not search for depression genes per se, but rather genes that control the effects of antidepressants. A good example of this research is the work on the genes that control liver enzymes and how they metabolize medication (e.g., Murphy, Kremer, Rodrigues, & Schatzberg, 2003). Another interesting twist is the report of significant associations between depression and genes that have no known function. For example, Zubenko, Hughes, and Stiffler (2002) found that the frequency of the gene D2S2944—found on the long arm of chromosome 2—was about three times higher in the female patients diagnosed with DSM-III-R (1987) recurrent early-onset major depressive disorder (MDD) compared to healthy female controls. In contrast, no increase in the frequency of the D2S2944 gene was found in males with this disorder, suggesting that a major sex difference exists in the aetiology of early onset MDD.

At this time, some of the most consistent molecular genetic results have come from linkage studies. Recent reviews of bipolar depression have noted several significant linkages on chromosome 18p (see Gershon et al., 1998). Chromosome 18 is of interest for another reason: several linkages to schizophrenia have been reported on this chromosome (see Gershon, 2000) and both disorders (bipolar depression and schizophrenia) have also been linked to chromosome 13q (Blouin et al., 1998; Detera-Wadleigh et al., 1999). These linkage studies suggest that what is inherited is not a specific disorder, but rather a general liability to psychopathology. Gershon (2000) refered to these as “nonspecific psychopathology genes”—genes that are shared by families but do not coaggregate in families. It is unclear at this time what this general vulnerability to psychopathology might be, but if these genes exist, their identification would help us to understand the biology of susceptibility, develop new diagnostic tests for this vulnerability, and focus attention on the genetic and environmental factors that differentiate various manifestations of disorder.

In summary, the most recent molecular genetic results remain as inconsistent as earlier research. Thus, Kendler, Neale, Kessler, Heath, and Eaves’ (1992a) comment that “in the absence of replicated positive results of linkage analysis, twin and adoption studies provide our only method for disentangling the genetic and non- genetic sources of familial resemblance for depression” (p. 257) is as valid today as it was a decade ago. The next section examines this now very large body of research.

THE HERITABILITY OF UNIPOLAR DEPRESSION

Female Depression

Heritability studies worldwide have shown that genetic factors account for between 30% and 40% of the total variation in MDD among general population females. In the United States, Kendler et al. (1992a) estimated the heritability of DSM-III-R MDD in a sample of 1,033 female twin pairs from the United States at 42%, with the remaining 58% due to nonshared environmental (e²) factors. More recently, Kendler and Prescott (1999a) reported the heritability of MDD at 39% using a sample of 3,790 female twin pairs from Virginia. Shared environmental (c²) effects were again estimated at zero, with the remaining 61% attributable to nonshared environmental effects.

These estimates were replicated in a large community-based sample of twins from Australia. Bierut et al. (1999) estimated the heritability of DSM-III-R (1987) MDD at 44% (95% CI: 29–53%). When the newer DSM-IV (1994) criteria were applied, the heritability estimate remained much the same at 36% (95% CI: 15%–46%). Heath et al. (1999) reported similar results: Additive genetic effects accounted for 26% of the variability in DSM-III-R MDD and 44% in DSM-IV MDD. To underscore the stability of these results across countries, a recent meta-analysis of family and twin studies worldwide estimated that 37% (95% CI: 31%–42%) of the variability in female MDD was attributable to genetic influences (Sullivan, Neale, & Kendler, 2000).

Male Depression

In contrast, the heritability of depression in males is noticeably lower. Among Australian male pairs, Bierut et al. (1999) estimated the heritability of male DSM-III-R (1987) depression at 24% (95% CI: 0.00%–39%). This estimate dropped a bit lower to 18% (95% CI: 0.00%–26%) when current DSM-IV (1994) criteria were applied. A significant aspect of these findings is that the lower boundary of the 95% confidence interval includes zero, suggesting that male depression may not be heritable at all. This was demonstrated when the definition of depression was modified to reflect severe depression. Among males, heritability dropped to a mere 1% (95% CI: 0.00%– 60%), but remained high at 38% (95% CI: 0.00%–52%) among females.

The question of gender differences was examined in further detail by Kendler and Prescott (1999a). They first estimated the heritability of depression in males and females. They found h²_A= 39% for each gender, which is inconsistent with reports from other studies. However, it can be argued that this is not the case because the estimates fall within reported confidence intervals from other studies. Nevertheless, the important point here is that it is commonly assumed that the causes of depression in males are not much different from the causes of depression in females. The similarity of the magnitude of male and female heritability in this study is consistent with this assumption, but remains to be tested by estimating the genetic correlation (r_G) between male and female depression. The r_G was far from unity (1.0), being estimated at 0.57, indicating that a significant proportion of the genetic factors underlying male and female depression are not shared, and that the route to depression is gender specific.

Heritability Based on Revised Diagnostic Criteria

The research just summarized is based entirely on standard diagnostic criteria. As outlined in Chapter 3, this is problematic because this system weights all symptoms equally and yields patient groups composed of different combinations of symptoms. Each patient thus presents with quite different forms of depression that could be caused by diverse genetic and environmental causes. This problem is exacerbated by the fact that diagnostic systems force behavioral phenomena into categories that deem the disorder present or absent without indication of actual severity.

A popular solution has been to revise the definitions to better reflect differences in severity. Two such systems, the Washington University criteria (WUC: Feighner, Robins, & Guze, 1972), and the Research Diagnostic Criteria (RDC: Spitzer, Endicott, & Robins, 1978) are commonly used across North America in clinical and research settings. The WUC and RDC criteria are frequently employed in conjunction with DSM-III (1980), DSM-III-R (1987), and DSM-IV (1994) criteria. Superficially, these definitions of depression are quite similar. They all specify periods of dysphoric mood or pervasive loss of interest or pleasure. The RDC is the most similar to the DSM system in that a diagnosis of depression requires that a patient meet a minimum five out of nine criteria. However, the RDC assesses more severe forms of depression because it requires that the duration of dysphoric features last at least 1 week and that the person sought or was referred to help from someone during the dysphoric period, took medication, or had impairment with family, at school, at work, or socially. The WUC criteria are the most different. Only the WUC was designed at the outset to allow differential diagnoses of “definite” or “probable” depression. The original RDC definitions have been modified to include these subtypes. The difference between definite and probable depression is in the number of criteria. Definite depression req...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Acknowledgments
• Preface
• I: Behavioral Genetic Basics
• II: The Behavioral Genetics of the Common Mental Disorders