Skip to content
photos related to autism and publications about it
Summary of Advances Cover 2017

Summary of Advances

In Autism Spectrum Disorder Research


Question 3: What Causes ASD, and Can Disabling Aspects of ASD Be Prevented or Preempted?

Cross-tissue integration of genetic and epigenetic data offers insight into autism spectrum disorder
Andrews SV, Ellis SE, Bakulski KM, Sheppard B, Croen LA, Hertz-Picciotto I, Newschaffer CJ, Feinberg AP, Arking DE, Ladd-Acosta C, Fallin MD. Nat Commun. 2017 Oct 24;8(1):1011. [PMID: 29066808]

Development of ASD is thought to be due in part to genetic factors. Although no single genetic variation has been discovered that is sufficient to cause ASD, current research has provided some insight into variations in gene expression and modifications that are associated with increased ASD risk. In this study, researchers focused on identifying genetic variations that can give rise to altered DNA methylation – a type of gene modification that can change the gene’s expression without altering its DNA sequence. These genetic variations, called methylation quantitative trait loci (meQTLs), can thus indirectly affect the expression of one or more ASD risk genes.

Because previous research has shown that genetic variations associated with ASD can be identified using DNA from blood cells, these researchers looked for meQTLs in cord blood taken from infants at birth (who were later diagnosed with ASD) and in peripheral blood samples from 3- to 5-year old children with ASD. They also examined data from publicly available genetic maps of fetal brain tissue for meQTLs. Lung tissue meQTL data was used as a control because it was not likely to contain ASD-related variations.

The researchers mapped the meQTLs in all four types of tissues and looked for enrichment of known ASD-associated variations. They found enrichment of ASD-associated meQTLs in cord blood, peripheral blood, and brain tissue; importantly, this pattern was not observed in lung tissue. When looking at the ASD-associated meQTL targets, whose expression would be altered by the variation at the meQTL site, the researchers identified 37 different biological processes that would be affected. Of those, three were associated with methylation targets found in all three tissue types, 12 were found in brain tissue but not cord or peripheral blood, and 22 were found in both cord and peripheral blood but not brain tissue. Many of these processes are related to immune system function. This study expands on previous research, which indicates that differences in immune function are associated with ASD. Importantly, the authors of this study suggest that certain types of tissue can provide information about ASD-related immune system processes that may not be evident in the genotype.

The results from this study provide further insight into the genetic variations that are associated with ASD, and the biological consequences of those variations. The fact that ASD-associated variations were apparent in blood samples is also significant because it suggests that future research need not rely solely on the use of brain tissue. Analyzed together, genetic and epigenetic data may provide a fuller and more accurate picture of the interactions among genes, biological factors, and ASD.

Fetal and postnatal metal dysregulation in autism
Arora M, Reichenberg A, Willfors C, Austin C, Gennings C, Berggren S, Lichtenstein P, Anckarsäter H, Tammimies K, Bölte S. Nat Commun. 2017 Jun 1;8:15493. [PMID: 28569757]

ASD is thought to be caused by a combination of genetic and environmental factors. Several behavioral and developmental problems that are common in ASD, such as intellectual, language, and attentional disabilities, are also common in individuals who have been exposed to toxic metals, such as lead, and in those with deficiencies in essential metals, such as zinc and manganese. To better understand potential environmental factors associated with ASD, the authors of this study examined whether toxic metal exposure and/or essential metal deficiencies play a role in the development of ASD.

To study these environmental factors while controlling for genetic variation, the researchers recruited discordant monozygotic (identical) and dizygotic (fraternal) twins — that is, pairs of identical and fraternal twins in whom one has ASD and the other did not — as well as sets of twins that both had ASD (concordant), and non-ASD twin controls. Seventeen pairs of identical twins and 15 pairs of fraternal twins participated in this study.

To study the exposure to toxic and essential metals before and after birth, the researchers examined the twins’ teeth, which provide a developmental record of metal exposure from the second trimester through early childhood stages. By collecting the twins’ lost deciduous teeth (those that naturally fall out during childhood) and analyzing them using mass spectrometry, the researchers were able to determine the concentrations of zinc, manganese, lead, and other metals at multiple developmental time points.

First, the researchers identified typical metal distribution patterns over the course of development by studying the teeth of non-ASD identical twins. They found that in typically developing children, manganese levels decline rapidly during the prenatal period and continue to decline after birth, while zinc levels remain steady during the prenatal period and then decline after birth. Next, they compared these patterns with those found in ASD-discordant identical and fraternal twins. They found that the affected twin’s manganese levels were lower both before and after birth, and zinc levels declined earlier during the prenatal period and increased to levels higher than the non-ASD twin’s after birth. In addition, they found higher levels of lead in the affected twin, particularly after birth. Further testing allowed the researchers to pinpoint the critical time period during which levels of these metals were the most different for the ASD-affected and non-ASD twin. These measurements indicated that lead levels were consistently higher in the ASD-affected twin between 10 and 20 postnatal weeks, while levels of tin in ASD-affected children were most different between 20 and 16 weeks before birth. Finally, ASD-affected children had the lowest levels of manganese – about 2.5 times less than the non-ASD twin – at postnatal week 15.

The researchers then compared the levels of zinc, manganese, and lead with the severity of ASD traits, as measured by scores on the Autism Diagnostic Observation Schedule Second Edition (ADOS-2) and Social Responsiveness Scale Second Edition (SRS-2). Though they found no significant association between zinc and measures of ASD severity or behavioral deficits, they found that low manganese levels were associated with higher scores on both the ADOS-2 and the SRS-2, and high lead levels were associated with higher ADOS-2 and SRS-2 scores.

The results of this study provide further evidence that environmental factors, particularly at prenatal and early postnatal developmental stages, may contribute to the development of ASD. However, the researchers note this study does not identify the causes of differences in the children’s metal levels. Reduction in the levels of essential elements may not be due to lack of exposure but may instead be caused by biological dysregulation of essential metals, suggesting roles for both genetic and environmental factors in the development of ASD.

Prenatal exposure to fever is associated with autism spectrum disorder in the Boston Birth Cohort
Brucato M, Ladd-Acosta C, Li M, Caruso D, Hong X, Kaczaniuk J, Stuart EA, Fallin MD, Wang X. Autism Res. 2017 Nov;10(11):1878-1890. [PMID: 28799289]

Genetic and environmental factors are thought to contribute to ASD, and research suggests that prenatal exposure to maternal immune activation (MIA; a change in the prenatal environment when an immune response is activated in the mother) is associated with the development of ASD. However, the types of MIA exposure that are risk factors for ASD are not clearly defined. In this study, the researchers aimed to better understand the nature of maternal infections that increase the risk for ASD in U.S. minority populations that are underrepresented in the literature. They focused on MIA due to influenza, genitourinary tract infections, and fever in a population of predominantly low income, urban, minority mothers and subsequent ASD risk for their children.

For this study, the researchers used data from Boston-area mother-child pairs, including 116 children with ASD and 998 children without ASD. Mothers were interviewed 24-72 hours after giving birth and indicated whether they were exposed to a range of infections. The questionnaire also gathered information about covariates (additional variables that may affect the outcome of the study) such as the mother’s race, education level, marital status, and whether the mothers smoked prior to or during pregnancy. The mothers that had been exposed to either fever, genitourinary tract infections, or the flu provided further trimester-specific information about their exposures, and their babies underwent postnatal follow-ups to assess for ASD symptoms.

The researchers compared prenatal infection exposure and ASD risk both before and after accounting for covariates, and they found that no association exists between maternal genitourinary tract infections and risk of developing ASD or between prenatal flu exposure during any trimester and risk of developing ASD. The researchers did, however, identify a significant association between maternal fever at any point during pregnancy and risk for ASD. When broken down by trimester, the researchers additionally found an increased risk of ASD when the maternal fever occurred in the third trimester, but not the first or second.

These findings support and elaborate on past studies that suggest that MIA may increase the risk for ASD. They also note that existing literature does not agree on the role of anti-fever medication (such as acetaminophen) in later ASD diagnoses, meriting future research. The results of this study provide a basis for the development and improvement of clinical strategies to reduce the incidence of maternal fever and provide additional insight about the relationship between prenatal environment and ASD.

Maternal multivitamin intake, plasma folate and vitamin B12 levels and autism spectrum disorder risk in offspring
Raghavan R, Riley AW, Volk H, Caruso D, Hironaka L, Sices L, Hong X, Wang G, Ji Y, Brucato M, Wahl A, Stivers T, Pearson C, Zuckerman B, Stuart EA, Landa R, Fallin MD, Wang X. Paediatr Perinat Epidemiol.. 2018 Jan;32(1):100-111. [Epub 2017 Oct 6] [PMID: 28984369]

Since the discovery of its critical role in preventing neural tube defects during gestation, folate has become an essential supplement for women of reproductive age before and during pregnancy. Though studies show that individual folate levels vary from very low to higher than recommended, the average levels of plasma folate in pregnant women have increased 2.5 times as a result of increased supplementation with multivitamins and consumption of folate-fortified cereal grains. Conflicting findings in the literature make the association between increased maternal folate consumption and development of ASD unclear. Furthermore, vitamin B12 is involved in many of the same biological pathways as folate, but there is limited research on its association with ASD risk. In this study, the researchers sought to clarify the association between increased risk of ASD development and multivitamin use, maternal plasma folate levels, and vitamin B12 levels.

The participants of this study were 1,257 mother-child pairs who were recruited to the Boston Birth Cohort study. The researchers gathered information from the mothers 24-72 hours after birth about their multivitamin supplementation habits during pregnancy along with collecting maternal plasma folate and B12 samples taken 2-3 days after birth and compared with ASD diagnostic information collected from their children’s electronic medical records. They found that during the first trimester, moderate multivitamin supplementation (three to five days per week) was associated with decreased risk of ASD – a finding that is consistent with the current literature. Moreover, both low supplementation (fewer than two times per week) and high supplementation (more than five times per week) were associated with an approximately 2.5 times increased risk of their child developing ASD. These results indicate that over- or under-utilization of multivitamins may both be associated with increased risk of ASD.

The researchers also measured maternal levels of plasma folate and vitamin B12 at birth (which are good indicators of maternal levels during the third trimester of pregnancy) and found that very high folate levels (those in the top 10th percentile of the study population) were associated with increased risk of ASD, as were very high levels of B12. The risk was increased whether the mother had elevated levels of folate, B12, or both.

These findings present the possibility that too much or too little folate may be associated with an increased risk of ASD. The researchers present these findings with the hope that future research will clarify the relationship of plasma folate and B12 levels with the development of ASD.

Autism risk following antidepressant medication during pregnancy
Viktorin A, Uher R, Reichenberg A, Levine SZ, Sandin S. Psychol Med. 2017 Dec;47(16):2787-2796. [PMID: 28528584]

When trying to understand the underlying causes of ASD, many researchers have considered the role of the medications taken by the mother during pregnancy and the risk of ASD. Results of previous studies investigating this issue have been equivocal. In this study, researchers sought to determine whether antidepressant use during pregnancy was associated with an increased risk of ASD.

The researchers followed a population-based Swedish cohort of 179,007 children from birth (in 2006-2007) until age 7 or 8 (in 2014). They used clinical diagnoses obtained from the Swedish Patient Register to determine whether the children had been diagnosed with ASD, and whether either parent had at least one psychiatric diagnosis listed. They also used the Swedish Prescription Drug Register to determine whether the mothers had been prescribed antidepressant medication during pregnancy, as evidence of child exposure to antidepressants. Based on these data, 13.2% of the parents in the study had a diagnosed mental illness; 0.9% of the children had an ASD, and among those, 61% were diagnosed with autistic disorder.

The researchers then measured the relative risk of ASD in the children who had been exposed to antidepressant medications during gestation. To minimize the possibility of including children of mothers who stopped taking antidepressants during pregnancy in the exposed group, children were classified as exposed only if their mothers had filled their antidepressant prescription on at least two occasions that overlapped with their pregnancies. The researchers found a 1.3 times increased relative risk of having ASD in the children of mothers that had taken any antidepressant during pregnancy. This risk was reduced to 1.07 after accounting for confounding variables, such as whether the mother had received a psychiatric diagnosis in her lifetime. When determining the risk associated with exposure to individual antidepressant medications in the subset of women diagnosed with a psychiatric disorder, they found that the greatest increased risks were for citalopram and escitalopram (1.47 times) and clomipramine (2.86 times). After controlling for other variables however, the relative risks calculated by the authors throughout the study were reduced to near-negligible levels.

The findings of this study do not indicate that maternal antidepressant use can increase the risk of ASD. Rather, the authors proposed that confounding variables contributed to the relative risks demonstrated. Specifically, ASD and other psychiatric disorders may be caused by similar underlying genetic factors. Thus, the presence of a gene variant that contributes to maternal depression could increase the risk of her child developing ASD, which may confound this and other studies investigating the link between antidepressant use and ASD risk. In addition, although these findings may have implications for clinician recommendations about medication use during pregnancy, untreated mental health conditions are themselves associated with poor health outcomes. Thus, the researchers do not believe these findings support the discontinuation of antidepressant use during pregnancy.

The association between maternal use of folic acid supplements during pregnancy and risk of autism spectrum disorders in children: a meta-analysis
Wang M, Li K, Zhao D, Li L. Mol Autism.. 2017 Oct 2;8:51. [PMID: 29026508]

There are conflicting reports in the literature regarding the association between folic acid supplementation during a mother’s pregnancy and risk for developing ASD in her child. The purpose of this article was to review the current literature on this topic and provide clarification of the nature of this association.

The researchers conducted a meta-analysis, i.e., they combined and re-analyzed data from 16 studies with a total of 4,514 participants with ASD. They excluded any articles that did not present data specific to the relationship between folic acid supplementation during pregnancy and ASD, and articles that did not offer sufficient human subjects data were not considered eligible. Overall, the researchers found that folic acid supplementation during pregnancy was significantly associated with a decreased risk of developing ASD compared to those without folic acid supplementation. They considered whether there were variations in ASD risk based on study type or race and found that the results were consistent across different study types and for Asian, European, and American populations.

Although the researchers identified high heterogeneity (variation in study outcomes) between studies when all the data were pooled, they found no evidence that it was attributable to differences in when, where, or how these studies were conducted. The researchers suggest that the variation may be due instead to unmeasured genetic or environmental factors, which may act as confounding variables. Furthermore, sensitivity analysis showed that no single study had excessive influence on the pooled result while excluding one study at a time. Although it is possible that the mothers may not have accurately remembered their folic acid intake years after pregnancy, future studies can account for this potential recall bias by ensuring that data is collected from mothers who were recently pregnant.

Though this study provides strong evidence to suggest that folic acid supplementation decreases the risk of ASD, the researchers concluded that future studies should address questions such as the timing or dosage of folic acid supplementation during pregnancy.

Question 3

Back to Top