Everyone knows that some environmental factors can have adverse effects on health, especially early in development. For example, we know that exposure to cigarette smoke is particularly bad for infants and young children, increasing risk for Sudden Infant Death syndrome, respiratory challenges and middle ear infections. While we are still learning what kinds of environmental factors might impact the intricate process of brain development, and exactly how these impacts occur, we all want to know how environmental factors influence risk for autism.
Last week the Society of Toxicology met in Washington D.C. to discuss not only environmental effects, but how they may interact with our genes to confer autism risk. The most popular topic of this 50th anniversary meeting was epigenetics —literally changes made “above the genome”. Different epigenetic changes have the effect of making the genetic code more or less available for reading and the production of proteins. In other words, the environment can actually turn off the functions of genes, resulting in downstream effects on brain and behavioral development.
During a special symposium organized by autism researcher Isaac Pessah, PhD from the University of California at Davis and Cindy Lawler, PhD at the National Institute of Environmental Health Science, , scientists discussed new data and examples of how environmental factors can lead to changes in autism risk. Animal models of autism are essential for carrying out tests such as these, as different amounts of exposure to a particular substance can be carefully delivered and the outcomes observed with all other variables controlled.
Janine LaSalle, PhD at the University of California at Davis studied the effects of a flame retardant on behavioral development and cognitive function. She and her colleagues showed that these cognitive effects, which are similar to those found in autism, are dependent on both the sex of the animal and proper function of epigenetic mechanisms that turn a collection of other genes “on” or “off”.
Researchers in the Tanguay lab at Oregon State University are using the humble zebrafish to study a newly discovered type of gene expression. The research team is studying the effects of alcohol (ethyl alcohol, both the type found in beverages and and as a biofuel additive to gasoline) and a common acne treatment ingredient (retinoic acid, a metabolite of vitamin A ) on gene expression in the zebrafish. They are finding that disruptions in this new type of gene expression (microRNAs) can have surprisingly large effects on the rest of the genome.
We know from many previous studies that duplications or deletions of collections of genes—called copy number variants or CNVs—can be associated with increased autism risk. Scott Selleck, PhD, from Penn State University reported on his study which looked at the genetic background of children in the CHARGE study at UC Davis (http://beincharge.ucdavis.edu/). Individuals with ASD showed increased lengths of CNVs at certain points in the genome. His lab reasons that these CNVs may be areas of what he calls “genomic instability” where environmental chemicals affect gene expression. We need to know more about these CNVs and whether or not they are the reason some individuals are more susceptible to environmental factors in development.
Genes and environment interact, yes, but another important factor is when. Timing of the environmental insult can be crucial. Studies of neural stem cells are showing us that there exist critical periods in the development of these immature brain cells that include times in which cells divide, and also a later time when the immature cells become either neurons or another type of brain cell known as glia. It is at these times when environmental influences might have their biggest effect.
Pat Levitt, Ph.D. from the University of Southern California spoke on how the combination of genetic vulnerabilities and environmental factors can converge to disrupt brain development and function. One example involves the MET gene, which controls the development of a special class of inhibitory neurons. Previous research showed mutations in MET to be associated with autism, especially in individuals with gastrointestinal dysfunction.
Dr. Levitt and his colleagues demonstrated that exposure to chemicals in diesel fuel exhaust also decreases proper expression of the MET protein. This reduction in expression leads to changes in complexity and length of neurons as they reach to connect with other neurons. These changes may contribute to the previously observed effects on brain development. Interestingly, a recent report notes an increased risk for autism in children whose mothers lived within 1000 feet of a major highway during pregnancy.
Autism Speaks is actively supporting a number of research projects investigating the role of epigenetics in autism, including how environmental factors interact with genetic mechanisms to influence behavior. A primary focus of research invited for submission to Autism Speaks in 2011 is the mechanism of gene/environment interactions, including epigenetics.
To read about all the research Autism Speaks is funding in this area, click here http://www.autismspeaks.org/science/research/initiatives/environmental_factors.php.
Guest blog by Dr. John Vincent, who is a Scientist and Head of the Molecular Neuropsychiatry and Development Laboratory of the Psychiatric Neurogenetics Section in the Neuroscience Research Department and an Associate Scientist of The Centre for Applied Genomics at The Hospital for Sick Children, Toronto.
The group here in the Molecular Neuropsychiatry & Development Lab at CAMH, along with our collaborative partners at Sick Kids and elsewhere report the identification of PTCHD1 as a gene for autism spectrum disorder, as well as intellectual disability, on the X-chromosome. The finding stems from attempts to find differences in specific strands of DNA, called copy number variants (CNVs) in the DNA of autistic individuals that might be linked to the condition. If the human genome can be thought of as a book containing the DNA code written out in words and sentences, we each of us have our own unique book, with many words spelled differently from each other, but mostly without changing the overall meaning of the words and sentences. Traditional genetic studies would try to identify spelling mistakes that compromise the meaning of a sentence, and thus lead to an incorrect message, i.e. resulting in a clinical condition. In our current study we have been looking instead for whole sentences, paragraphs or pages that are either deleted or duplicated (i.e. CNVs), thus altering the meaning of the message, and leading in this case to autism.
We were particularly interested, in our study, to look at CNVs on one of the 2 sex chromosomes, the X, as it might explain some of the bias towards boys having autism over girls. In our initial screen of over 400 autism patients, we identified a large deletion disrupting the PTCHD1 gene. In an analysis of CNVs in genomes from over 1000 more autism individuals, deletions just next to this gene were the most significant finding. These deletions are likely to disrupt DNA sequences that may regulate how the PTCHD1 gene is expressed. In addition, we identified many single letter changes in the PTCHD1 gene that may affect the “meaning”. These changes were not found in the DNA of many control individuals.
The PTCHD1 gene makes a protein with as yet unknown function, however it shows similarities to several known proteins that function as cell-surface receptors for an inter-cellular signaling pathway known as Hedgehog, crucial in determining how brain cells develop and mature. The preliminary data we present in the paper shows that PTCHD1 appears to have similar properties to the two Hedgehog receptors already known, leading us to speculate on a role for Hedgehog-related processes in autism.
These findings give us another important gene that we can screen for in at risk children, and will allow earlier therapeutic interventions, thus increasing the likelihood of success.
For more information, please see the press release.
By Geri Dawson, Chief Science Officer, Autism Speaks
Science moves so slowly and is so labor-intensive that we don’t often have moments to celebrate an achievement or breakthrough that has resulted from our investments. With this week’s announcement of Phase 2 results from the Autism Genome Project, we are celebrating such an achievement.
Several years ago, when I was a professor at the University of Washington, I remember a phone call from Andy Shih, Ph.D. (Autism Speaks VP, Scientific Affairs) who asked if he could take my colleague, Jerry Schellenberg, and me out to breakfast. Over coffee, Andy described to us an idea he had: Would we be willing to collaborate with other scientists around the world and add the genetic data we have been collecting to a combined database? While each of us at that time had been working independently to try to discover autism risk genes, we knew that ultimately we would need much larger samples to deal with the significant heterogeneity that exists in autism spectrum disorder. After a lot of discussion and questions, Andy convinced us that this would be a worthwhile effort and thus we became part of what became known as the “Autism Genome Project,” or the AGP. Eventually, Andy talked with over 50 groups worldwide and cajoled each of them to join the effort. What ensued was a series of monthly conference calls, complex negotiations and agreements that Andy helped broker, the creation of a combined database, and yearly meetings during which the goals for analysis and future data collection would be discussed. Today, the AGP is considered a driving force in autism genetic research.
Meanwhile, Clara Lajonchere, Ph.D. (Autism Speaks VP, Clinical Programs) was spearheading an effort to create a database of multiplex families called the Autism Genetic Resource Exchange (AGRE). She was leaving most of us collecting similar samples in the dust as she quickly assembled the largest private genetic individual data base that exists. Her ability to form partnerships with families, engaging them in the process of scientific discovery, was a model for us all. Not surprisingly, Clara readily agreed to join the AGP since AGRE’s basic premise was “collaboration and data sharing.”
Fast forward to this week when the AGP published the largest and most comprehensive study of copy number variations (CNV) – small deletions or duplications in our genome that can disrupt gene function – in autism families. By comparing CNVs found in 1,000 individuals with autism with those from 1,300 individuals without autism, the AGP reported the following:
- Several novel ASD genes were discovered, and many genes previously implicated by other studies were confirmed. Some of these genes are involved with communication between neurons, while others help regulate cell growth and how they respond to environmental stimuli.
- It was confirmed that autism risk genes are rare variants in our genome that occur very infrequently or not at all in the general population, and each person with ASD may have a unique risk gene or set of risk genes. Some of these genes are “highly penetrant” meaning that, if you carry this risk gene, you very likely will develop ASD, whereas other only raise the risk for ASD and need to combine with other genetic and/or environmental risk factors to cause ASD. Some of these are inherited, but many appear “de novo” meaning that they only exist in the child and not the parents.
- In the not-so-distant future, we will start to see more comprehensive genetic testing being conducted in the clinic to provide parents with information about whether their child may be at risk for ASD, so they can watch for signs or better understand the cause of their child’s ASD. It will be important to consider carefully what tests are appropriate and interpret them in a manner that is responsible and helpful for parents.
- Although the fact that so many rare genes can be related to risk for autism seems to form an overwhelmingly complex picture of autism, there is a path forward: These genes appear to cluster around specific biochemical pathways in the brain and, thus, point to new directions for developing drugs that could potentially help recover function of these pathways. This is good news for families.
Most of all, I see the publication of this report as a celebration of the fruitful partnership between the families and the scientific community. While Autism Speaks staff like Andy and Clara helped create and implement unique and productive scientific endeavors like the AGP, ultimately, it is the families who contributed their time and literally a part of themselves that is helping us put together this puzzle called autism piece by piece.
Beyond genetics: What the new fields of functional genomics and epigenetics are revealing about autism
Today’s guest post comes to us from Valerie W. Hu, Ph.D. Dr. Valerie Hu is a Professor of Biochemistry and Molecular Biology at The George Washington University Medical Center as well as a mother of a son with ASD. She redirected her research focus towards autism about 5 years ago and has since published 6 papers on the genes and biological pathways associated with ASD. Dr. Hu received her Ph.D. in Chemistry from Caltech and did her postdoctoral research in Membrane Biochemistry and Immunology at UCLA. More information about her research and papers can be obtained at: http://www.gwumc.edu/biochem/faculty_vhu.html
For many years, genetics has followed the traditional approach to identifying genes associated with various disorders, including autism. However, the wide diversity of symptoms and behaviors associated with autism spectrum disorders (ASD) has posed a significant challenge to identifying mutations in one or a few genes that are reliably associated with ASD. More recent attention has focused on copy number variants (CNVs) which are submicroscopic regions of chromosomes that have been found to be lost or duplicated in some individuals with ASD. Together, these studies are making progress in identifying a number of genes that function at the synapse between nerve cells. Nevertheless, the combination of all known genetic mutations still does not account for the majority of autism cases or for the association of non-neurological symptoms observed in autistic individuals.
My laboratory at The George Washington University Medical Center has taken a functional genomics approach to studying genes that may be deregulated in autism. Rather than identify mutations in DNA, our goal has been to identify genes whose activity, as indicated by gene expression level, is altered in order to identify dysfunctional biochemical pathways and impaired cellular functions in autism. Thus, using a method known as DNA microarray analysis to profile gene expression in cells from identical twin pairs who are differentially diagnosed with autism (one autistic, the other not reaching the threshold for an autism diagnosis), sibling pairs where only one sibling is autistic, and unrelated case-controls, we identified many genes whose expression levels (activities) are different from non-autistic individuals. Furthermore, by subtyping the unrelated autistic individuals according to symptomatic severity across over 63 symptoms probed by a commonly used diagnostic assessment instrument (Autism Diagnostic Interview-Revised), we were able to identify gene “signatures” for each of 3 ASD subtypes studied. These signatures not only revealed genes unique to a given subtype of autism, but also overlapping genes that presumably control common symptoms of autism across subtypes. Interestingly, a set of genes that regulate the biological clock (that is, circadian rhythm) were found to be disrupted only in the subtype of ASD exhibiting severe language impairment.
These studies demonstrating different gene expression signatures between identical but differentially diagnosed twins as well as revealing differential expression of hundreds to thousands genes depending on ASD subtype suggest the involvement of “master switches” (or epigenetic mechanisms) in the control of gene expression. Two new studies by our laboratory, published in two different journals on Apr. 7, 2010, suggest that epigenetics may play a significant role in the regulation of gene expression in autism (read a blog from AS staff on epigenetics). In one study, published in the FASEB Journal, we identified chemical tags (called “methyl groups”) on the DNA of individuals with autism that led to gene silencing. This mode of turning off a gene is potentially reversible with the proper drugs if the specific gene can be properly targeted. The second study, which was published in the journal Genome Medicine, reports on the differential expression of microRNA in autism. MicroRNA are recently discovered snippets of RNA (ribonucleic acid), each of which can inhibit the expression (and thus activity) of hundreds of genes. The effects of microRNA are also reversible by treatment with complementary “anti-sense” RNA. While methylation inhibits gene expression at the level of DNA, microRNA inhibits at the level of RNA. These two studies together illustrate two different “epigenetic” mechanisms controlling gene activity in autism that lie beyond genetic mutations.
In the study published in the FASEB Journal, we again used cell lines derived from identical twins and sibling pairs in which only one of the twins or siblings was diagnosed with autism to identify chemical changes on DNA. We then compared the genes that showed changes in DNA tagging (methylation) with a list of genes that showed different levels of expression from these same individuals. The amounts of protein produced by two of the genes that appear on both lists were then investigated in brain tissues from the cerebellum and frontal cortex of autistic and control subjects which were obtained through the Autism Tissue Program. We found that both selected proteins, RORA (retinoic acid-related orphan receptor-alpha) and BCL-2, as predicted by the observed increase in methylation, were reduced in the autistic brain. Although BCL-2 has previously been reported to be reduced in autistic brain, RORA is a novel gene which is relevant to many of the observed deficits in autism. Specifically, RORA is involved in several key processes negatively affect by autism, including Purkinje cell differentiation, cerebellar development, protection of neurons against oxidative stress, suppression of inflammation, and regulation of circadian rhythm.
These results suggest that blocking the chemical tagging of these genes may reverse some symptoms of autism if targeted removal of methyl groups from specific genes can be accomplished. Furthermore, this study, which links molecular alterations in blood-derived cells to brain pathobiology, demonstrates the feasibility of using more easily accessible cells from blood (or other non-brain tissues) for diagnostic screening.
This research is reported in the study, titled “Global methylation profiling of lymphoblastoid cell lines reveals epigenetic contributions to autism spectrum disorders and a novel autism candidate gene, RORA, whose protein product is reduced in autistic brain,” which was recently published in the Federation of American Societies for Experimental Biology (FASEB) Journal, and is available online at: http://www.fasebj.org.
In the study published in Genome Medicine, we identified changes in the profile of microRNAs between twins and sibling pairs, again discordant for diagnosis of autism. We discovered that, despite using cells derived originally from blood, brain-specific and brain-related microRNAs were found to be differentially expressed in the autistic samples, and that these microRNAs could potentially regulate genes that control many processes known to be disrupted in autism. For example, differentially expressed miRNAs were found to target genes highly involved in neurological functions and disorders in addition to genes involved in gastrointestinal diseases, circadian rhythm signaling, and steroid hormone metabolism. The study further shows that by treating the cells with antisense RNA antagonists (inhibitors) to specific microRNA or by employing mimics of a particular microRNA, one can reverse the pattern of expression of a given target gene regulated by that microRNA.
This study, titled “Investigation of post-transcriptional gene regulatory networks associated with autism spectrum disorders by microRNA expression profiling of lymphoblastoid cell lines” was highlighted as an “Editor’s pick” in the journal Genome Medicine. It is available online at: http://genomemedicine.com/content/2/4/23.
By integrating both DNA methylation and miRNA expression studies with gene expression data, Dr. Hu and colleagues are applying a systems biology approach to understanding this complex disorder.
In honor of the anniversary of Autism Speaks’ founding on Feb 25, for the next 25 days we will be sharing stories about the many significant scientific advances that have occurred during our first five years together. Our ninth item, Copy Number Variations, is from Autism Speaks’ Top 10 Autism Research Events of 2007.
Three studies published in 2007 pointed autism researchers toward the importance not only of mutations within the DNA code of specific genes, but also of variations in the number of copies of genes, known as Copy Number Variations (CNV). Created by submicroscopic deletions or duplications of DNA sequences, recent advances have only now allowed CNV to become routinely detectable, establishing an entirely new avenue of genetic research.
Armed with the latest technology and three different collections of patient DNA, including Autism Speaks’ Autism Genetic Resource Exchange (AGRE), researchers at Cold Spring Harbor scanned the genome for the presence of CNV in autism. In February 2007 they reported that not only do individuals with autism have more CNV than individuals without autism, but that CNV in autism occur more often as “de novo” or spontaneous mutations (mutations not found in the DNA of either parent). They also found that these spontaneous CNV appear to be more common in families with only one child with autism (simplex) than those with multiple affected children (multiplex).
With their data suggesting that genetic mechanisms may be different in different types of autism, the researchers then carefully studied the inheritance pattern of autism in the many families in Autism Speaks’ AGRE and IAN databases. In July they published that they could fit the data to a model in which there are at least two distinct ways that genes may play a role in the development of autism: spontaneous CNV might help to explain autism in simplex families, whereas inherited gene mutations may be at the root of autism spectrum disorders in a greater portion of multiplex families. Such a model is significant because although autism has been thought to have a strong genetic component, so far it has not been shown to be as clearly inherited as other simple genetic disorders.
The team from Cold Spring Harbor has provided a new theory of autism risk that stands to influence how future autism genetic research is conceptualized. Although their results require replication, it also now leads to the question of what causes this increase in spontaneous CNV, opening the door to an exploration of the interplay between genetics and environmental factors. Possible risk factors include age of the parents, specific toxicological factors and accumulated exposures, as well as genetic predisposition.
Update since this story was first run: Improvement in DNA technology since the publication of this first major autism CNV study has meant that new insights into the role of CNV are actively being pursued. The largest and most comprehensive autism CNV study to date was published in April 2009 in the top research journal, Nature. The new study focused on inherited rather than spontaneous CNV, with the researchers finding CNV variations in genes important for neural development [for more details, see http://www.autismspeaks.org/science/science_news/top_ten_autism_research_events_2009_cnv.php]. Researchers in autism are continuing to lead geneticists in other fields with an intense focus on CNV and early recognition of their importance as potential disease risk factors.