Unraveling the Genetics of Neurodevelopmental Disorders

In order for 85 billion neurons to develop into a healthy human brain, distinct neural cells must proliferate, differentiate, migrate, and integrate successfully.¹ It’s not surprising then that neurodevelopmental disorders (NDDs) affect more than 3% of all children.². Up to 45% of all NDDs, which include autism spectrum disorders (ASDs), intellectual disability syndromes, and syndromes of developmental delay have been linked to specific genes.³

Genetic sequencing has been a quantum leap forward in our ability to gather genetic data, allowing massive amounts of DNA to be sequenced rapidly. Today, tens of thousands of genetic variants can be filtered to identify gene mutations through a process that once took years to identify a small number of disease-causing mutations.¹ “At our specialized site, we can sequence a whole genome in under 50 hours,” said Sarah E. Soden, MD, developmental pediatrician and associate professor at the University of Missouri-Kansas City School of Medicine.

NDDs Still Difficult to Diagnose

Even with state-of-the-art gene sequencing, molecular diagnosis of NDDs is challenging. Although many genetic variations have been identified for many NDDs, variable expressivity, reduced penetrance, and phenotypic complexity make molecular diagnosis difficult. Locus heterogeneity can lead to similar NDDs, and genotypic convergence can lead to an NDD or to a psychiatric disorder.³

“We suspect that at least 30% of ASDs are caused by de novo mutations. Half of the mutations we find may be false positives. They may be silent mutations, or they may code for a different disorder that may show up later,” said Michael Ronemus, PhD, research assistant professor at Cold Spring Harbor Laboratory in New York, and a lead author on a 2014 study on the genetics of autism spectrum disorders. The study, published in Nature, analyzed the genomes of 2,500 families in which a single child had an ASD, but neither parent nor any siblings had ASDs. The study identified about 400 de novo mutations that may contribute to ASDs.⁵

“Some genetic defects seen in ASDs are also seen in schizophrenia. We still have little knowledge of what causes de novo mutations or what triggers mutations to become active. We do know that mutations increase with paternal age,” said Ronemus.

“We all have some de novo mutations. Many of them are silent, and some of them may be advantageous. That’s how evolution works. We still need clinical diagnosis to guide treatment. Fetal alcohol syndrome and prematurity also contribute to NDDs. The epidemiology of these disorders remains a moving target,” said Soden.

Whole Exome Versus Whole Genome Sequencing

One way to cut down on the complexity and the cost of gene sequencing is to sequence only the genes that code for proteins. The exome accounts for 1.5% of the whole genome, and are the genes most likely to cause significant developmental changes. They are the “low-hanging fruit” of molecular diagnosis. “The advantages of whole exome sequencing are less expense and more yield because these are the genes that code for amino acids. The disadvantage is that you may miss something,” said Ronemus.

Soden and her team of researchers used both exome and genome sequencing in 100 families of children with NDDs. The researchers were able to make a molecular diagnosis of an established NDD in 45 out of 100 families. Most importantly, a change in clinical care or clinical diagnosis was reported in 49% of the newly diagnosed families. The study, which was published in Genomics, also concluded that using these technologies at the time when symptoms first appear could shorten the time to diagnosis by 77 months.²

Clinical Utility of Molecular Diagnosis: Now and Going Forward

Despite all the complexities of genetic sequencing, computational methods are being developed to simplify the mystery of genetics in NDDs. An integrated gene network for autism was recently presented by Cold Spring Harbor Laboratory, which was able to define subsets of the genetic network associated with autism to high-functioning autism and more severe forms of autism.⁵

It’s now known that de novo gene-disrupting mutations can be divided into risk classes along the autism spectrum.⁴ Findings like this bring us closer, not only to earlier diagnosis, but also to more targeted interventions. “In five to 10 years we should be able to routinely test for genetic causes of neurodevelopmental disorders with a high probability of getting a diagnosis. Beyond that, there lies the possibility of genetic-targeted drug treatments,” said Ronemus

“Earlier diagnosis can make an impact right now. We have many years of data showing that we can alter the course of autism with early intervention. In some NDDs, molecular diagnosis can lead to earlier treatment for predictable kidney or cardiac defects. We can also use this information to improve genetic counselling. For example, if a de novo genetic defect has been identified, we may be able to reassure parents that the defect is highly unlikely to occur in a subsequent pregnancy,” said Soden.

Chris Iliades, MD, is a full-time freelance writer based in Cape Cod, Massachusetts. 

This article was medically reviewed by Pat F. Bass III, MD, MPH. 

References

1. Hu WF, Chahrour MH, Walsh CA. The diverse genetic landscape of neurodevelopmental disorders. Annu Rev Genomics Hum Genet. 2014;15:195-213.
2. Soden SE, Saunders CJ, Willig LK, et al. Effectiveness of exome and genome sequencing guided by acuity of illness for diagnosis of neurodevelopmental disorders. Sci Transl Med. 2014;6(265):265ra168.
3. Chen ES, Gigek CO, Rosenfeld JA, et al. Molecular convergence of neurodevelopmental disorders. Am J Hum Genet. 2014;95(5):490-508.
4. Cold Spring Harbor Laboratory. New study casts sharpest light yet on genetic mysteries of autism. Available here: http://www.cshl.edu/1257-new-study-casts-sharpest-light-yet-on-genetic-mysteries-of-autism.html
5. Hormozdiari F, Penn O, Borenstein E, Eichler EE. The discovery of integrated gene networks for autism and related disorders. Genome Res. 2014; Available here: http://genome.cshlp.org/content/early/2014/12/04/gr.178855.114.long