Workshop report: 16p11.2 deletions and duplications

Key region: Duplication or deletion of a 593-kilobase stretch of DNA on chromosome 16 is associated with autism and other neuropsychiatric disorders.

On 27 April, SFARI hosted a workshop to discuss the consequences of copy number variations (CNV) — duplications or deletions of DNA — in this region.

The workshop brought together researchers who study clinical features, genetics and neurobiology related to 16p11.2. It also offered an opportunity to share data uncovered so far by the Simons Foundation Variation in Individuals Project, or Simons VIP. This project, funded by’s parent organization, aims to characterize in depth a number of CNVs that contribute to autism, starting with 16p11.2.

Human condition:

Wendy Chung, assistant professor of pediatrics at Columbia Medical Center and director for clinical research for SFARI, described the clinical features of the 201 individuals — 92 with the duplication and 109 with the deletion — recruited as of that date into Simons VIP. The overall aim of the project is to characterize individuals with CNVs in 16p11.2 and see whether they share symptoms and brain characteristics that could suggest targeted treatments.

Overall, the results suggest that the deletion leads to more severe symptoms than the duplication does. More males carry the deletion than females do, whereas equal proportions have the duplication. The deletion is also more likely than the duplication to arise spontaneously, or de novo: About 54 percent of cases of the deletion are de novo, whereas 31 percent are inherited. In contrast, the duplication is inherited in 60 percent of cases and arises de novo 12 percent of the time. A significant proportion of individuals in the study — 8 of 52 with the deletion and 14 of 50 with the duplication — have unknown inheritance, including as a result of adoption.

About 34 percent of individuals with the deletion have a larger-than-typical head size, also called macrocephaly, whereas 16 percent of those with the duplication have smaller heads than controls. The deletion is associated with seizures in 24 percent of individuals and obesity in 19 percent.

Although only some individuals have significantly larger or smaller heads than normal, the overall spectrum of head size among these individuals is shifted from the usual baseline, noted David Ledbetter, chief scientific officer of Geisinger Health System. For example, the smallest head size in individuals with the deletion is larger than the smallest in typical individuals. Atypical head size may be more widespread than the small percentages suggest, he cautioned.

Ellen Hanson, director of developmental medicine and genetics at Children’s Hospital Boston, described in further detail the clinical features and diagnoses of the 55 individuals with the deletion and 47 with the duplication that the researchers had fully characterized at the time of the workshop.

Overall, scores on measures of verbal and nonverbal intelligence quotient (IQ), language development and adaptive behavior are lower in children with 16p11.2 CNVs than in controls. There is no significant difference, however, between the IQ scores of children with the CNV who have autism and those who do not.

Of the 55 with the deletion, 10 meet the criteria for a diagnosis of autism, 20 have language problems and 21 have motor deficits or intellectual disability. Of the 47 with the duplication, 5 have autism.

Mirror effect:

Overall, the preliminary results suggest that individuals with a 16p11.2 CNV have some form of intellectual disability and may have significant repetitive behaviors.

Researchers should pay attention to development and focus on long-term studies of children with 16p11.2 CNVs, said Edwin Cook, professor of psychiatry at the University of Chicago. This is because the data suggest that these children could improve over time, altering their diagnosis.

Of 312 individuals referred to clinics in England and France for intellectual disability and obesity, 9 have a deletion in 16p11.2, reported Sebastien Jacquemont, associate physician at the department of Medical Genetics at the University Hospital of Lausanne, in Switzerland.

Obesity in those with the deletion also increases as they get older, suggesting that environmental factors may influence the onset of weight gain. Obesity does not worsen along with intellectual disability, however, suggesting that poor lifestyle or impulse control is not the primary cause of weight gain, Jacquemont said.

Nouchine Hadjikhani, associate professor of radiology at Harvard Medical School, presented preliminary brain imaging data. Scanning the brains of seven individuals with the deletion, five individuals with the duplication and ten controls using functional magnetic resonance imaging (fMRI), her team found results suggesting that a difference in the brain response to food could underlie the obesity associated with the deletion.

People with the deletion may have a stronger brain response to both healthy and unhealthy food than do controls in regions that process a sense of reward. They also have a stronger brain response in those same regions to food than do individuals with the duplication.

Elliott Sherr, associate professor of neurology at the University of California, San Francisco, also presented preliminary brain imaging data, using fMRI, magnetoencephalography and diffusion tensor imaging, for individuals recruited into the Simons VIP program. His data suggest that deficits in brain regions involved in sensory processing are associated with CNVs in 16p11.2.

A number of individuals with the deletion and the duplication — 23 of 47 and 7 of 36, respectively — also have abnormal spine curvature, suggesting that there may be structural irregularities in the formation of the neural tube.

Genes to symptoms:

One key research goal is to investigate the role of each gene within the 16p11.2 region.

Rui Luo, a graduate student in the laboratory of Daniel Geschwind at the University of California, Los Angeles, presented a study of gene expression levels in blood cells, which can mirror changes that also take place in neurons. The researchers looked at overall changes in gene expression in individuals with the deletion or the duplication and found that the two affect different pathways. For example, the deletion alters the expression of 291 genes and the duplication affects 665, with very little overlap — only 6 genes — between the two.

To home in on the role of individual genes, researchers are also using zebrafish, which are easy to manipulate genetically and which have transparent embryos that allow researchers to visualize the brain and nervous system.

Hazel Sive, professor of biology at the Massachusetts Institute of Technology (MIT), has identified zebrafish versions of 22 of the 29 genes within 16p11.2. Her team has engineered zebrafish lacking one of each of these genes. All but one of the fish have abnormal brain structure and show a number of other defects, including abnormal tail shape and motor deficits.

Sive’s team also developed a system to cut the expression of each gene by half, to mimic the loss of one copy of 16p11.2. This approach highlighted two genes, ALDOA and KIF22, which are important for the generation of neurons and muscle function.

Her team is also looking to see how loss of activity of multiple genes in the region might work together to lead to autism symptoms. The researchers have already identified 4 of 20 investigated gene pairs that have strong effects when their expression is decreased together but not individually.

Nicholas Katsanis, director of the center for human disease modeling at Duke University, also presented research using the zebrafish model. Katsanis and his team injected the genetic messages for human 16p11.2 genes into zebrafish, to mimic 16p11.2 duplication, and then lowered the expression of the zebrafish version of the gene to look at the role of deletion.

This approach identified one gene, KCTD13, that may be involved in the alterations in head size seen in people with 16p11.2 CNVs. An extra copy of KCTD13 leads to zebrafish brains that are about 20 percent larger than the brains of controls, whereas fish lacking the gene have brains that are 18 percent smaller. KCTD13 plays a role in the generation of new neurons.

Although these experiments aim to home in on the role of genes within 16p11.2, the genes may function in a larger network of autism candidates, according to preliminary research presented by Lauren Weiss, assistant professor of psychiatry at the University of California, San Francisco.

Weiss and her team found that variations in single DNA base pairs, or single nucleotide polymorphisms, in a number of known autism candidates outside 16p11.2 influence the expression of genes within the region. This suggests that the 16p11.2 genes may form part of a larger functional network, the understanding of which could help researchers narrow down the mechanisms involved in autism.

Mouse models:

Several researchers at the meeting presented data from mouse models with deletions and duplications of the 16p11.2 region.

Alea Mills, professor at Cold Spring Harbor Laboratory, in New York, has created mice that have either one or three copies of the 16p11.2 region. The deletion is partially lethal, leading to the death of about half of the pups with the mutation. These mice do not have milk in their stomachs, suggesting that they have difficulty suckling. This could be the result of problems in the hypothalamus, as disruptions to this region affect feeding in newborn rodents, says Mills.

Mice with the deletion have defects in motor skills that are so obvious that researchers can identify them just by watching them for a few days after birth, Mills said. They are also hyperactive, have difficulty sleeping and obsessively climb on the lids of their cages, suggesting the repetitive behaviors seen in individuals with autism. Mice with the duplication show opposite symptoms.

Whereas control mice lose interest in another mouse over time, mice with the deletion continue to show the same level of interest, suggesting defects in social memory.

Dolmetsch also described a number of neurological deficits in the brains of the mice. In particular, they have 30 to 50 percent more neurons that produce dopamine, a chemical messenger that regulates motor function and reward, than do controls, in the striatum — a brain region involved in processing reward.

However, there are also pronounced differences between the mice and people, Dolmetsch noted. For example, the mice are thinner than controls, whereas people with the deletion tend to be obese. The mice are also deaf, a characteristic that has not been associated with the disorder in people, although people with autism are known to have some hearing-related problems.

Mark Bear, professor of neuroscience at MIT, presented preliminary data suggesting that mice lacking one copy of 16p11.2 have neuronal deficits similar to those seen in mouse models of fragile X syndrome. Treatments being tested for fragile X syndrome may also benefit individuals with a 16p11.2 deletion, Bear said.

The workshop concluded by acknowledging the large amount of work required to characterize and understand just one of the many genetic regions that underlie autism.

However, the fact that studying this one region has already highlighted candidate genes for autism suggests that this approach is well worth it, said Gerald Fischbach, scientific director of SFARI. In fact, the studies have highlighted genes both within and outside the 16p11.2 region, he said, and can help parse out which neurobiological pathways contribute to autism.

Recent Workshop and Meeting Reports