Traditional genetics, of the type investigated by Mendel and his followers, was able to identify genes that had large effects and often displayed typical patterns, such as those involving dominant, recessive, or sex-linked traits. Genes that affect human behaviors and exhibit such patterns are well-known, including Huntington's disease (caused by an autosomal dominant mutation) and phenylketonuria, or PKU (a recessive mutation). Symptoms of Huntington's disease's include degeneration of the nervous system, usually beginning in middle age and resulting in death. In this devastating disease, there is usually a gradual loss of intellectual ability and emotional control. The genetic pattern is that of a condition caused by a rare, single, dominant gene. Since affected people have one copy of the dominant disease gene and one copy of a recessive gene (for a "normal" nervous system), half of their offspring develop the disease. Huntington's never skips a generation. Since the gene is dominant, the person who inherits it will manifest the disease (if he or she lives long enough). If one full sibling has the condition, there is a fifty-fifty chance that any other sibling will also get the disease.
In contrast to dominant conditions, recessive conditions show a very different pattern of occurrence. Recessive means that both copies of the gene must be of the same form (the same allele) in order to show the condition. Two parents, neither of whom shows a trait, can have a child affected by a recessive trait (this happens if both parents are carriers of one copy of the recessive allele—the child thus has two copies, one from each parent, and manifests the condition). Recessive traits can skip generations because parents and their offspring can carry one copy of the recessive gene and not display the associated trait. In the population there are many recessive genes that cause various abnormal conditions. Each particular recessive allele may be rare, but since there are many of them, their combined impact on a population can be substantial.
Among humans, a classic example of recessive inheritance is the condition of phenylketonuria (PKU). Individuals with PKU usually are severely mentally impaired. Most never learn to talk; many have seizures and display temper tantrums. PKU is a form of severe mental retardation that is both genetic and treatable. It is genetic in that it is caused by a recessive genetic allele. Without two copies of that particular allele, a person will not develop the set of symptoms, including mental impairment, that is characteristic of PKU. However, scientific knowledge has led to a treatment. It was discovered that the recessive PKU gene prevents the normal metabolism of a substance that is common in food, making many normal foods toxic to the individual with two PKU alleles. A special diet that is low in the offending substance can prevent or minimize the nervous system damage that leads to the profound intellectual disabilities of untreated PKU individuals.
The example of PKU demonstrates that inherited (genetic) conditions can be treated—that knowledge of specific causation can result in effective treatment. This is an extremely important point both ethically and philosophically, because it is often misunderstood and misinterpreted.
Well over one hundred different genes are known for which relatively rare recessive alleles cause conditions that include severe mental impairment among their symptoms. The rapidly developing knowledge of basic genetic chemistry, from molecular genetics to biotechnology and the Human Genome Project, which produced a mapping of some 30,000 human genes early in the twenty-first century on April 15, 2003, holds out the hope that many more of these devastating genetic conditions may soon be treatable. As part of the Human Genome Project, genes for Huntington's disease and PKU have been identified and sequenced, though as yet no new therapies have been developed for these disorders.
In spite of these clear scientific successes related to Mendelian genetic-pattern disorders, many human traits— including normal traits, as well as somatic, behavioral, and psychiatric disorders—have not exhibited clear Mendelian patterns of inheritance. For those traits, an extension of Mendel's work to quantitative traits that was first developed by Sir Ronald Fisher, has been used extensively. Beginning in the 1990s, an additional, more molecular, set of techniques was developed to examine possible influences of genetics on human behavior. These two broad approaches to studying the influences of nature and nurture in psychiatry are termed quantitative (or epidemiological) and molecular. A brief summary of the two approaches is presented here, including some examples of their results and their problems (an overview of them can be found in Neiderhiser and in
Schaffner , and a systematic analysis is presented in Behavioral Genetics by Plomin et al.).
QUANTITATIVE METHODS. Quantitative, or epidemiological, methods are utilized to distinguish genetic and environmental contributions to quantitative traits or features of an organism, as well as to assess correlations and interactions between genetic and environmental factors that account for differences between individuals. These methods do not examine individual genes, but report on proportions of differences in traits due to heredity or environment, or to their interactions, broadly conceived. The methods include family, twin, and adoption studies. Adoption studies examine genetically related individuals in different familial environments, and thus can prima facie disentangle contributions of nature and nurture. Twin studies compare identical and fraternal twins, both within the same familial environment and (in adoption studies) in different familial circumstances.
Twin studies have been used extensively in psychiatry to indicate whether a disorder is genetic or environmentally influenced, and to what extent. Twin studies make several assumptions to analyze gathered data, including that the familial environment is the same for twins raised together but different for twins raised apart, an assumption called the equal environments assumption. Though critics of genetic influence often question this assumption empirical studies have confirmed it (Kendler et al.). The example of schizophrenia may help make some twin results clearer. Employing what are termed concordance studies of twins, Gottesman and his associates have reported over many years that the risk of developing schizophrenia if a twin or sibling has been diagnosed with the condition is about 45 percent for monozygotic (MZ) twins, 17 percent for dizygotic (DZ) twins, and 9 percent for siblings (Gottesman and Erlenmeyer-Kimling). This concordance pattern supports what is called a non-Mendelian polygenic (many genes) quantitative trait etiology for schizophrenia with a major environmental effect (> 50%), i.e., more than half of the differences in liability to schizophrenia among individuals is due to environmental factors. Twin studies can also be used to estimate the heritability of a trait or a disorder, which for schizophrenia is about 80 percent. Heritability is a technical term, one that is often confusing even to experts, and one which only loosely points toward the existence of underlying genetic factors influencing a trait. Investigators note that "it does not describe the quantitative contribution of genes to ... any ... phenotype of interest; it describes the quantitative contribution of genes to interindividual differences in a phenotype studied in a particular population" (Benjamin et al., p. 334). If there are no interindividual differences in a trait, then the heritability of that trait is zero—leading to the paradoxical result that the heritability of a human having a brain is virtually zero. Heritability is also conditional on the environment in which the population is studied, and the heritability value can significantly change if the environment changes.
Keeping these caveats in mind, heritability estimates for many major psychiatric disorders appear to be in the 70 to 80 percent range, and personality studies indicate heritabilities of about 30 to 60 percent for traits such as emotional stability and extraversion, suggesting that these differences among humans are importantly genetically influenced. But even with a heritability of schizophrenia of about 80 percent, it is also wise to keep in mind that approximately 63 percent of all persons suffering from schizophrenia will have neither first- nor second-degree relatives diagnosed with schizophrenia, reinforcing the complex genetic-environmental patterns found in this disorder.
Twin studies were also the basis of a distinction between shared and nonshared environments. The meaning of environment in quantitative genetics is extremely broad, denoting everything that is not genetic (thus environment would include in utero effects). The shared environment comprises all the nongenetic factors that cause family members to be similar, and the nonshared environment is what makes family members different. Remarkably, quantitative genetics studies of normal personality factors, as well as of mental disorders, indicate that of all environmental factors, it is the nonshared ones that have the major effect. A meta-analysis of forty-three studies undertaken by psychologists Eric Turkheimer and Mary Waldron in 2000 indicated that though the nonshared environment is responsible for 50 percent of the total variation of behavioral outcomes, identified and measured nonshared environmental factors accounted for only 2 percent of the total variance. Turkheimer infers that these nonshared differences are nonsystematic and largely accidental, and thus have been, and will continue to be, very difficult to study (Turkheimer, 2000). This possibility had been considered in 1987 by Robert Plomin and Denise Daniels but dismissed as a "gloomy prospect"— though it looks more plausible.
Epidemiological investigations have also identified two important features of how genetic and environmental contributions work together. The first, genotype-environment correlation (GsE), represents possible effects of an individual's genetics on the environment (e.g., via that individual's evoking different responses or selecting environments). Such effects were found for both normal and pathological traits in the large Nonshared Environmental Adolescent Development (NEAD) study, described in detail in the 2000 book The Relationship Code, written by David Reiss and colleagues. Secondly, different genotypes have different sensitivities to environments, collectively called geno-typexenvironmental interaction (GxE). Differential sensitivity is important in many genetic disorders, including the neurodevelopmental models of schizophrenia genetics and in a recent study on the cycle of violence in maltreated children (discussed later).
MOLECULAR METHODS. Classical quantitative or epidemiological studies can indicate the genetic contributions to psychiatric disorders at the population level, but they do not identify any specific genes or how genes might contribute (patho)physiologically to behavioral outcomes. According to psychiatric geneticist Peter McGuffin and his colleagues, "quantitative approaches can no longer be seen as ends in themselves," and the field must move to the study of specific genes, assisted by the completed draft versions of the human genome sequence (McGuffin et al., p. 1232). In point of fact, a review of the recent literature indicates that most research in behavioral genetics, and especially in psychiatric genetics, has taken a "molecular turn."
It is widely acknowledged that most genes playing etiological and/or pathophysiological roles in human behaviors, as well as in psychiatric disorders, will not be single locus genes of large effect following Mendelian patterns of the Huntington's and PKU type discussed earlier. The neurogeneticist Steven Hyman notes that mental disorders will typically be heterogeneous and have multiple contributing genes, and likely have different sets of overlapping genes affecting them. Mental disorders will thus be what are called complex traits, technically defined as conforming to non-Mendelian inheritance patterns.
There are two general methods that are widely used by molecular behavioral and molecular psychiatric geneticists in their search for genes related to mental disorders: (1) linkage analysis, and (2) alleleic association. Linkage analysis is the traditional approach to gene identification, but it only works well when genes have reasonably large effects, which does not appear to be the case in normal human behavior or in psychiatry. Allelic association studies are more sensitive, but they require "candidate genes" to examine familial data. An influential 1996 paper by statisticians Neil Risch and Kethleen Merikangas urged this strategy.
Studies in schizophrenia are again illustrative of these approaches, as are the Alzheimer's disease genetic studies reviewed later. Though there was an erroneous 1988 report of an autosomal dominant gene for schizophrenia on chromosome 5 that is seen as a false positive, evidence has been accumulating for genes or gene regions of small effect related to schizophrenia on many chromosomes, including 1q, 2, 3p, 5q, 6p, 8p, 11q, 13q, 20p, and 22q (Harrison and
Owen). Replication difficulties with these results in different populations of schizophrenics and their families have been a recurring problem, however.
Was this article helpful?