Featured Profile

For the last 20 years, scientists have been looking feverishly for specific genes, or mutant versions of genes, whose presence in a person can be linked with greater statistical risk for developing major psychiatric illnesses. Genes called DISC-1 (for “disrupted in schizophrenia”) and
neuregulin-1, for instance, are among several that have shown strong statistical correlations with risk for schizophrenia, and are widely presumed to have some causative role in at least some people who develop the illness.

But this is not to say that DISC-1 or neuregulin-1 causes schizophrenia. Like every other major neuropsychiatric illness, schizophrenia is a “complex disease,” not caused by inheritance of a single errant copy of a single gene. Many of the best scientists in the world find it reasonable to believe that schizophrenia, bipolar disorder, autism, and other psychiatric illnesses with a known genetic component are the product of some as yet unknown interaction of several genes and the environment (the latter encompassing a vast range of physical, social, and cultural exposures and influences). That view, however, is coming under scrutiny, as other contributing factors, including chemical alterations of DNA, are studied.

What Twins Teach Us About Schizophrenia

Arturas Petronis, M.D., Ph.D., is among the leaders in the search for alternative accounts of causality. He has been thinking about the problem of causation in mental illness throughout his career in medicine. When NARSAD first funded his lab — and it has done so on three subsequent occasions — it enabled him to embark on a variation of a classic study. He wanted to look at sets of identical twins and examine the incidence of schizophrenia through a special conceptual lens, an approach called epigenetics.

“The remarkable fact about monozygotic, or identical, twins, which has been known for many decades, is that in spite of their identical genomes, one sibling can develop a serious illness like schizophrenia while the other does not,” says Dr. Petronis, who is head of the Krembil Family
Epigenetics Laboratory at the Centre for Addiction and Mental Health and Associate Professor at the University of Toronto.

If the cause of schizophrenia is genetic, how can two people with precisely the same sequence of 3 billion DNA “letters” — the same letters, placed in the precisely the same order — suffer such different fates? “In science, we say that in such cases, the genotype, or genetic code, is identical; but the phenotype — the way that the genome expresses itself in each individual — is different,” Dr. Petronis explains.

The question that has driven his career, since he arrived in Canada from his native Lithuania in 1991, when the Soviet Union collapsed, is: what happens to a given genome, or even a small section of one, that makes it issue different directions to cells of the body — directions which have at least some bearing on whether that body is in a state of health or illness?

“NADSAD came into my life when I began to study this question, and the situation was very difficult,” Dr. Petronis remembers. “I didn’t have any dedicated sponsors initially, and then, every few years I would come to a point where I’d say, ‘Either I get a grant to continue, or I wrap up and change my focus. It was phenomenal. Each one of those grants was like an injection of enthusiasm into our work, for which I cannot thank NARSAD enough.”

Changes That Turn Genes ‘Up’ and ‘Down’

There are all sorts of things that can happen to two identical stretches of DNA in two individuals that will result in differences in the way genes in each person direct cells to manufacture proteins — the basis of all structural and functional aspects of life. Many of the changes affect the process by which genes are “transcribed” into RNA “messenger” molecules that command the ribosomes in cells to make proteins. Many things can happen to these “messenger-RNAs” while they are being spliced together and after they are produced, which alter them, and which can alter gene expression.



Many other things can happen to molecules in the DNA double helix that can have the effect of “turning up” or “turning down” genes, somewhat like the volume on a TV remote control. One is the attachment of a molecule called a methyl group to one of the “bases” in the chain of linked molecules in the spiraling DNA “staircase” (see illustration, page 5). This is called methylation, and it is extremely commonplace. Methylation does not change the sequence of DNA “letters,” but does affect the structural and chemical configuration of the DNA, and also the intensity with which genes are expressed. The total set of such changes, over the entire genome, is called the genome’s epigenetic profile, or epigenome.

Some epigenetic events are normal, while others are not. But in either case, epigenetics adds a remarkable dimension of complexity to the problem of disease causation. At any given moment, some of the genes in our bodies are undergoing epigenetic changes. Some of these are fleeting, and some are with us always, having been inherited from our parents. As opposed to the human genome — which is very stable throughout our lives — the epigenome varies within individuals over time, and from individual to individual across the population. Indeed, even within families. Hence, the phenomenon of the identical twins, only one of whom has schizophrenia: genetically the same, but expressing their genes differently.



A Pioneering Epigenetics Study

Genes like DISC-1 are important because they have been associated with schizophrenia risk; yet they have “not given us a single solid finding that we are able to use today in clinical practice,” Dr. Petronis notes. For this reason, he is not devoting his research to the search for additional “risk-genes,” which depend on identification of variations in DNA sequence. Rather, he, with NARSAD’s backing, is exploring the alternative hypothesis that “some part of the heritability” of schizophrenia, bipolar disorder and other psychiatric illnesses “is due to epigenetic factors.”

Working with 150 postmortem brain slices and other tissues taken from people affected by schizophrenia and from controls, the Petronis lab has just completed a grueling, multiyear study. “It’s the first, we believe, that looks at epigenetic changes across the human genome in the context of a psychiatric illness.”

The difficulties of the project are mind-boggling. For each of the 150 samples, 12,000 spots on the epigenome were examined at the molecular level — the level at which it was possible to see where methyl molecules have attached to the DNA double-helix; and where proteins called histones are chemically modified, also altering gene expression. The histones, assembled in groups of eight, are the structural “spools” around which strands of DNA wrap themselves for dense packaging into the genome’s 23 chromosomes.

“We ended up with several dozen loci in the genome where we were able to find epigenetic differences between the two groups,” Dr. Petronis reports. “This suggests to us that DNA methylation changes are important to the etiology of schizophrenia and bipolar disorder, which we have known for years, clinically, overlaps with schizophrenia in terms of family inheritance patterns. We can now try this over the full length of the human genome — a project that would involve comparisons at some 40 million spots. In this way we can test the hypothesis that a small number of key elements, which are epigenetically inherited, as well as acquired epigenetic misregulation, can explain a large number of clinical, epidemiological and molecular findings — the kind of features that make ‘complex’ genetic diseases complex.”

“The point is not about the presence or absence of epigenetic factors in the context of disease,” Dr. Petronis stresses. “Epigenetic changes are critical for normal function of the cell. We are talking specifically about misregulation, and about measuring the quantities of changes between affected individuals and controls, to get to the bottom of this.”

Top of page