The National Association of Science Writers (NASW) and the Council for the Advancement of Science Writing (CASW) held annual meetings together 5-9 November 2010 at Yale University in New Haven, Connecticut. SRF writer Pete Farley helped put together the joint event—ScienceWriters 2010—which marked NASW's seventy-fifth and CASW's fiftieth anniversaries. Victoria Wilcox sends in this report on a special lecture from the meeting.
20 December 2010. High-speed genomic techniques are ushering in an era of relatively consistent findings that could lead to treatment targets for childhood neuropsychiatric disorders, according to Matthew State of Yale University, New Haven, Connecticut. In a lecture on 7 November, at the 2010 New Horizons in Science briefing in New Haven, sponsored by the Council for the Advancement of Science Writing, State described how these newer methods are facilitating the search for rare variants in particular and offer a shortcut to the underlying biology. For instance, his own work unexpectedly fingered a gene that affects levels of histamine in a unique family with Tourette’s syndrome. He has also tied divergent rare brain malformations in different people to mutations in a single gene, challenging beliefs that they involve separate pathways.
After the discovery of the Huntington’s disease gene, said State in a follow-up interview with SRF, the failure of the “one gene, one disease” model to yield similar results for other inherited neuropsychiatric disorders changed the paradigm. The new dogma, that these more complex disorders result from a “conspiracy” of common genetic variations led to efforts to identify the conspirators. Research groups would study their favorite gene and publish their findings, which were hardly ever replicated. “The enterprise became highly suspect because you’d have an association, and it would be one more publication cycle before someone would call that association into question,” State said in his lecture.
In the past three years or so, the availability of high-speed genomic technologies has improved the ability to find common variants of small effect; instead of choosing which genes to study, researchers could study the entire genome in an unbiased way without bankrupting the lab. Yet, State said, “We’ve seen more than 100 associations of common variations in the genome with a whole variety of medical disorders,” such as diabetes, aneurysms, and inflammatory bowel disease; “despite that, we still have not been able to find, as a general proposition, a single common variant in a child psychiatric disorder that has reproducibly been shown to be carrying risk for that disorder.” Taking a different approach, he and others have trained their periscopes on rare variants (see SRF related news story), in part because natural selection would probably limit the population frequency of mutations that cause illness.
A stake in the ground
In State's view, the rare variant approach “has very little to do with how much it explains in the population and everything to do with whether or not it can give you some novel insight into how to begin to treat these disorders.” Even findings that connect a lot of rare mutations to disease might point to a single functional network. One approach to identifying rare variants focuses on outliers, such as families who are inbred, show an extreme phenotype, or are very large. Among psychiatric genetics researchers, State noted to SRF, those who study schizophrenia were among the first to use this approach to explore biology, capitalizing on the discovery of the DISC1 mutation in the Scottish family (see SRF related news story). A second approach involves sequencing the entire coding part of the genome to find rare single-nucleotide variants. Until the recent development of high-throughput sequencing, the cost and difficulty of doing so forced researchers to focus on smaller parts of the genome, he said.
A door to the unexpected
State and colleagues have used the outlier approach to study Tourette’s syndrome, a genetic neurological disorder that causes children to repeatedly twitch, flap their arms, cough, squeak like a mouse, or engage in other semi-voluntary tics. First author A. Gulhan Ercan-Sencicek, also of Yale University, and others studied two generations of a family in which all eight children and the father, but not the mother or her side of the family, developed the syndrome. Unlike most families affected by Tourette’s, this one showed a pure Mendelian inheritance pattern.
As reported in the May 20 New England Journal of Medicine (Ercan-Sencicek et al., 2010), linkage analysis flagged one segment of chromosome 15. When the researchers sequenced the genes there, they found something surprising: a rare stop codon in the histidine decarboxylase, or HDC, gene in all affected but no unaffected family members. HDC makes the rate-limiting enzyme in histamine production. The team further showed that the mutant alters one side of a homodimer, decreasing histamine levels.
Previous research had found that HDC-knockout mice show tic-like behavior, which histamine reverses. Histamine acts throughout the brain, and in the striatum it blocks the release of dopamine, as do the antipsychotic drugs used to treat tics. State noted that drug companies first became interested in compounds that increase brain histamine for their potential as schizophrenia treatments. Because late-stage clinical trials are already testing some of these compounds as treatments for comorbid disorders related to Tourette’s syndrome, State thinks his rare variant finding might quickly lead to a clinical trial of a possible tic remedy.
Not so different after all
The new techniques enable researchers to use smaller pedigrees to hunt for variants over the genome. For instance, State and colleagues sequenced every exome in five inbred families from Turkey that would have been too small to study using older methods. In a paper published in Nature on September 9, first author Kaya Bilgüvar of Yale University and others (Bilgüvar et al., 2010) tied recessive mutations in WD repeat domain 62 (WDR62) to severe brain malformations in one family with two affected siblings, a finding supported in six other families. Although the mutations involved the same gene, they seem to have produced a range of phenotypes, such as microcephaly (small brain), pachygyria (cortical thickening), lissencephaly (smooth cortex), and others, which were previously thought to stem from distinct molecular mechanisms.
These results complement past work suggesting genetic overlap among different neuropsychiatric disorders. For instance, other studies have implicated duplications at 16p11.2 with increased risk for autism, schizophrenia, and other developmental abnormalities; deletions of that region to autism spectrum disorders and intellectual disability; and deletions in 22q11 to autism, velocardiofacial syndrome, and perhaps schizophrenia (see SRF related news story).
Opportunities in autism
About 20 years ago, State and colleagues identified an outlier group that they had no way to study genetically: children with childhood disintegrative disorder, an illness loosely within the autism spectrum. These youngsters develop normally for about two years and then regress, losing skills that they had previously acquired in areas such as language and social interaction. To determine whether de novo mutations play a role, the researchers are doing whole-exome and whole-genome sequencing and neuroimaging of these children and their families, a comprehensive approach made feasible by the new techniques. They hope the findings will not only help families affected by this rare illness, but also generate leads as to the biology behind social behavior in children.
The latest sequencing techniques are also helping researchers find rare variants in typical families instead of having to track down extreme phenotypes or pedigrees. State and colleagues are currently studying common-variety families that include one child with autism. So far, he told SRF, they are finding “a lot of rare variation in the genome; there are many, many things in the genome that we’ve never seen before because we haven’t had the ability to see them.” Before they publish the results, he and his colleagues want to see if the de novo mutations they have found relate to disease, a step that other studies, even those in top journals, sometimes overlook.
Compared to other neuropsychiatric research communities, State told SRF, the schizophrenia field is setting the pace of discovery and offers a model for others to follow. He cites the work connecting 22q11 deletions to schizophrenia as a good example of moving from identifying an association to identifying interesting genes and fleshing out the mechanisms by which they could lead to the human phenotype (see SRF related news story; SRF news story; SRF news story; SRF news story).
In his lecture, State said that all treatments for childhood neuropsychiatric disorders have arisen serendipitously, but the new technologies could lead to more rational treatment development. He sees the day dawning when psychiatrists will be able to talk about targeting pathways and mechanisms the same way that their colleagues who treat heart disease, cancer, and diabetes do.—Victoria L. Wilcox.
Bilgüvar K, Öztürk AK, Louvi A, Kwan KY, Choi M, Tatli B, Yalnizoğlu D, Tüysüz B, Çağlayan AO, Gökben S, Kaymakçalan H, Barak T, Bakircioğlu M, Yasuno K, Ho W, Sanders S, Zhu Y, Yilmaz S, Dinçer A, Johnson MH, Bronen RA, Koçer N, Per H, Mane S, Pamir MN, Yalçinkaya C, Kumandaş S, Topçu M, Ozmen M, Sestan N, Lifton RP, State MW, Günel M. Whole-exome sequencing identifies recessive WDR62 mutations in severe brain malformations. Nature. 2010 Sep 9;467(7312):207-10. Abstract
Ercan-Sencicek AG, Stillman AA, Ghosh AK, Bilguvar K, O'Roak BJ, Mason CE, Abbott T, Gupta A, King RA, Pauls DL, Tischfield JA, Heiman GA, Singer HS, Gilbert DL, Hoekstra PJ, Morgan TM, Loring E, Yasuno K, Fernandez T, Sanders S, Louvi A, Cho JH, Mane S, Colangelo CM, Biederer T, Lifton RP, Gunel M, State MW. L-histidine decarboxylase and Tourette's syndrome. N Engl J Med. 2010 May 20;362(20):1901-8. Abstract