Reflections on the Relationship Between Psychiatric Genetics and Psychiatric Nosology

In a lecture on psychiatric genetics given by a leading academic psychiatrist in 1987, the speaker began by outlining the recently successful efforts at identifying the genes underlying several Mendelian disorders and then went on to say:

With the new advances in molecular genetics and linkage analysis, by the time we get to DSM-V, we will be diagnosing chromosome 4 schizophrenia, chromosome 14 schizophrenia, and chromosome 22 schizophrenia. [His choice of particular chromosomes was illustrative rather than data-based].

In expressing such optimism about the impact of advances in psychiatric genetics on our diagnostic system, this individual was also suggesting something more fundamental. In predicting that our diagnostic manual would contain subtypes of schizophrenia identified by chromosomal location, he was advocating an essentialist model in which abnormal genes would be the defining feature of psychiatric disorders. Such an essentialist perspective assumes that psychiatric conditions are not man-made constructions but rather represent real disorders that exist out there in the world and can be defined by their underlying nature (or essence) (27).

The essentialist model of disease assumes that diseases can be classified in the same manner that atomic elements can be classified. Gold, silver, and lead are all true and independent entities, each with a unique essence—in this case defined by the number of protons in the nucleus.

Mendelian diseases are defined by the genes that cause them. Just as discrete microorganisms cause particular infectious diseases, so discrete genes cause particular genetic diseases. Just as individual types of bacteria or viruses are true entities in the world, so genes could be seen as classic essentialist categories, each with its own essence, roughly analogous to the atomic elements.

Essentialist disease models are very attractive. They are conceptually simple, appealing, and easy to teach. They fit well into the traditional medical model, thereby supporting the status of psychiatry as a medical discipline. Further, if they are linked to biological causes, they provide support for an organic disease model where psychiatric disorders are understood as resulting from pathological processes in the brain. They appear legitimate to third-party payers who can accept them as “real” diseases.

Indeed, the claim that the two great historical success stories in our field—general paresis of the insane and pellagra—are essentialist diseases rests on their both resulting from highly discrete causes: Treponema pallidum and vitamin B₃ deficiency, respectively. In the adjoining discipline of neuropsychiatry, in the last 15 years, the essentialist disease status of a series of disorders has been legitimized by the localization and subsequent identification of the genes that caused them, including Huntington's chorea, early-onset Alzheimer's disease, and Wilson's disease. No wonder some in psychiatry were suffering from “gene envy,” which expressed itself in their advocating what could be called an essentialist gene model (EGM) for psychiatric disorders.

Although appealing, is the EGM, as exemplified by Mendelian diseases, applicable to psychiatric disorders? Back in 1987, things already did not look good for this model, which depends critically on the etiologic link between gene and illness being so strong that the essence of the disorder is explicable by a disruption of gene function (28). Nothing in the twin, family, and adoption studies of psychiatric disorders suggested that they were nearly as genetically simple as Mendelian disorders. Despite much searching, no well-validated pedigrees had been found where a major psychiatric disorder is transmitted through generations as a classical Mendelian trait. The risk of illness in relatives of ill patients never resembled those expected from the simple laws of Mendel. Since 1987, further information on four fronts has all suggested the inapplicability of the EGM for psychiatric disorders. First, several high-profile linkage results in bipolar illness and schizophrenia (29–31) that were in part predicated on Mendelian models of illness could not be replicated. Second, many linkage studies were performed for schizophrenia and bipolar illness, and no replicated evidence was found for genes with a Mendelian-like effect. Instead, it appears that these disorders are the result of at least a moderate number of genes that individually have small to modest effects on disease liability (19,20). Third, supportive of the preliminary findings in psychiatry, work in model organisms demonstrated that genetic influences on the vast majority of behavioral traits were the result of many genes, each of a modest effect size (32,33). Fourth, animal studies consistently showed that, in model organisms, genes that influence behavior typically impact on multiple phenotypes (32). This phenomenon, termed pleiotropy, is inconsistent with one central feature of the EGM—the existence of a one-to-one relationship between gene and disorder (like the one-to-one relationship in the periodic table between the number of protons in the nucleus and other characteristics of the element). Pleiotropy may be so widespread because it reflects the fundamentally opportunistic nature of evolution (34).

One of the hopes expressed by these researchers is that new discoveries in psychiatric genetics will permit us to define the boundaries of psychiatric syndromes. They have expressed a second implicit expectation for the nosologic impact of psychiatric genetics research—that it will support categorical definitions of illness. A categorical view of psychiatric illness—that these disorders are discrete entities with distinct boundaries—can be contrasted with the perspective that these disorders are pathological ends of functional continua.

Advocates of a categorical perspective on psychiatric disorders suggest that they are discrete entities that are similar to biological species, such as whales and cows, or to man-made objects such as shirts and pants. Such entities have distinct boundaries. It is clear what is inside and what is outside. With categorical entities, the task of the nosologist becomes finding these boundaries or, as is oft said, “carving nature at its joints.”

Classical Mendelian disorders appear to be such categories. Clinically, these diseases appear to be discrete. In families with multiple affected individuals, there are typically no “spectrum” cases; individuals are either affected or unaffected. Perhaps genetic discoveries would uncover such clear forms of psychiatric illness.

Categorical models of disease, by definition, require discrete boundaries or what has been termed “points of rarity.” For genes to be useful in defining categorical disease entities, the etiologic effect of the gene must be large enough that it produces such a “point of rarity” between those who possess and those who lack the disease gene (35). Classically, this would be represented as bimodality in a distribution of liability—the joint that is to be carved. To successfully ground a categorical diagnostic system in pathogenic genes, the genes need to affect liability strongly enough that their impact is detectable above the background effect of other risk factors.

As reviewed earlier, prior evidence from linkage studies of psychiatric disorders and animal behavior genetics do not provide encouraging news for the EGM. The effect sizes of genes found in these studies have typically been small. In the last decade, individual susceptibility genes have been tentatively identified for psychiatric disorders. Therefore, we can examine the magnitude of their effect more directly. A review of positive meta-analyses of functional candidate genes for psychiatric disorders found odds ratios ranging from 1.07 to 1.57, with a median of ∼1.30 (36). (An odds ratio is the risk for a disorder given the presence of a risk factor—here a particular gene—divided by the risk for that disorder in the absence of exposure to the risk factor.) In schizophrenia, replicated evidence is now emerging for several genes that have been localized under linkage peaks (37), in particular dysbindin 1 and neuregulin 1. A recent review of dysbindin studies suggested that the odds ratio of variants in the gene and risk for schizophrenia average around 1.50 (38). For neuregulin 1, two recent replications reported odds ratios of 1.25 and 1.80 (37).

Figure 1 displays the liability distributions in a putative sample of first-degree relatives of individuals with schizophrenia, one-half of whom possess a high-risk copy (more technically allele) of a gene for schizophrenia. (The term “liability” here reflects the individual's level of risk for the illness, with risk increasing as you move from left to right in each panel in Figure 1.)

Using plausible parameters (see Figure 1 legend for further details), we have varied the magnitude of risk conveyed by the gene to produce odds ratios for the relationship between the high-risk allele and schizophrenia of 1.5, 5, and 10 in Figure 1 panels A, B, and C, respectively. Each panel presents four different distributions of liability. The dark blue line reflects the liability distribution of relatives without the high-risk allele. The purple line reflects the liability distribution of relatives with the high-risk allele. The turquoise line reflects the “reference” liability distribution that would be seen if there were no individual genes of detectable effect and only background genetic and environmental variation that would be predicted to take the shape of a normal distribution. The orange line, which is the most important one, reflects the liability distribution of the population of all relatives and is simply the sum of the blue and purple line. In addition, the green line represents the cutoff point for illness. Individuals with liability above that threshold will develop illness.

The thought experiment we are here conducting is as follows: If we could measure liability directly in these relatives (although we would not know their individual genotype), could we cut cleanly (at nature's joint) between those at high risk (depicted by the purple line) and those at low risk (depicted by the blue line)? That is, in the total population distribution (depicted by the orange line), do we see a clear point of rarity separating the two groups? For a gene with an odds ratio of 1.5 (Figure 1, panel A), there is virtually no deviation in the population distribution from that predicted by a single normal distribution. Even with odds ratios of 5 or 10 (Figure 1, panels B and C, respectively), all that is observed is a slight flattening of the distribution with no evidence for a point of rarity at which to divide those carrying and not carrying the high-risk allele.

Effect sizes are the range of those seen with genes that impact on risk for psychiatric disorder are too small to produce, on their own, syndromes with discrete boundaries.

While psychiatric genetics has sought to understand how genes contribute to psychiatric illness, cracks have appeared in the essentialist image of Mendelian genetic disorders. Examples have emerged that challenge a central feature of the gene-based essentialist disease model: the one-to-one relationship between gene and disease. Numerous instances have been found of mutations in different genes that all cause the same disease or similar diseases. Researchers have identified more than 60 chromosomal regions that lead to Mendelian forms of nonsyndromic hereditary deafness (where deafness is the only symptom) (39). At least 14 genes that produce chronic progressive hereditary ataxia have been identified (40). Three distinct genes produce Mendelian forms of early-onset Alzheimer's disease (41). Many Mendelian human diseases arise from dysfunction in complex biological pathways that involve products from numerous distinct genes. Abnormalities in different genes in these pathways are likely to produce similar diseases.

Furthermore, mutations in the same disease gene can produce distinct phenotypes. Cystic fibrosis results from dysfunction in the cystic fibrosis transmembrane conductance regulator. However, in addition to leading to classic cystic fibrosis, mutations in the this gene can solely cause congenital bilateral absence of the vas deferens, isolated idiopathic chronic pancreatitis, mild late-onset pulmonary disease, sinusitis, allergic bronchopulmonary aspergillosis, and possibly asthma (42). Dysbindin has been associated with schizophrenia in more than 10 independent samples (38,43). However, mutations in this gene also cause Hermansky-Pudlak syndrome type 7, which is characterized by symptoms—oculocutaneous albinism, prolonged bleeding, and pulmonary fibrosis—that are apparently unrelated to those of schizophrenia (44). Mutations in fibroblast growth factor receptor 3 produce a range of skeletal dysplasias with quite distinct phenotypes (45).

Even within medical genetics, essential disease models based on “the pathogenic gene” have become more difficult to sustain as evidence has mounted that different genes can cause the same disorder and abnormalities in one gene can cause different disorders.

Few concepts in biology have generated as much controversy as the nature of “the gene” (46–48). Debates have focused on the validity of several different conceptualizations, including 1) a statistical definition as seen in population genetics or genetic epidemiology, 2) a latent “unit” controlling phenotypic inheritance as conceptualized by Mendel and Morgan, 3) the template for production of a unique protein, and 4) a discrete physical entity that is a specific piece of DNA with a particular chromosomal location. When psychiatrists think about grounding their essentialist diagnostic concepts on the firm foundation of genes, they focus on the third and fourth definitions—a specific “hunk” of DNA with a discrete biological function. We see the gene as a clear “natural kind”—a material entity that exists as a real, discrete unit in the world. In basing our “messy” diagnostic concepts on this natural kind—the gene—we hope that nosologic clarity will follow.

However, advances in molecular biology have undermined these simple definitions of the gene. The “one gene=one enzyme” hypothesis has been falsified. In the human genome, 75% of multiexon human genes are alternatively spliced with approximately 3.5 alternative forms of each gene (49). Neuregulin 1, one of the best supported susceptibility genes for schizophrenia, produces at least 15 distinct protein products (50). That is, with alternative splicing, the same gene, defined at the level of nucleotide sequence, produces different mRNA transcripts, which are in turn transcribed into different proteins. (So each transcript produces a unique protein, but one gene produces multiple transcripts.) If a multiply spliced gene contains a variant sequence in one of its alternately spliced exons, that variant will be present in some but not other proteins produced from the gene.

Many of these alternatively spliced genes have multiple promoters, with the result that different protein variants of a single gene are expressed at distinct times in different tissues. For example, the gene α-tropomyosin in the rat produces seven distinct proteins, two of which are expressed in striated muscle and one each in smooth muscle, myoblasts, fibroblasts, brain, and hepatomas (48).

The functional boundaries of the “gene” concept have been blurred by a phenomenon termed “gene sharing” whereby the same gene product serves dramatically different biological functions. For example, Piatigorsky (51) has documented instances in which several metabolic enzymes have been “recruited” to also function as crystallins in the vertebrate lens.

A further uncertainty in the function of “the gene” arises from RNA editing—the posttranscriptional alteration of RNA sequence from that encoded in DNA (52). In some cases, such editing alters the structure of the expressed protein.

The physical boundaries of a “gene” are also becoming blurred. Key to the functioning of classic protein-transcribing genes is a series of control regions that influence the rate of transcription. Although such regions—termed promoters—exist immediately upstream of the coding region, researchers have found other control regions (enhancers and repressors) that are up to a million base pairs upstream or downstream and sometimes even in the introns of neighboring genes of unrelated function (53).

New variants of noncoding (nc) RNA have been discovered that further obscure the boundaries of what is meant by a “gene” (54). These ncRNAs can be classified into two broad groups: housekeeping ncRNAs and regulatory ncRNAs. Housekeeping ncRNAs are involved in RNA splicing and translation. Regulatory ncRNAs, including short interfering (si) RNA, can play an important role in gene expression through both transcriptional and posttranscriptional mechanisms as well as through alteration of higher-order chromatin structure.

Advances in our knowledge have indicated that the concept of the “gene” as an essentialist biological entity with an unambiguous nature and clean boundaries is unsustainable. Genes are not discrete entities like atoms of gold and silver. They are dynamic parts of biological systems of immense complexity. The discovery of specific genes that are involved in the etiology of psychopathology will not likely prove to be the basis on which to build an essentialist and categorical model of psychiatric diagnosis.