DNA methylation signature of autism spectrum disorder (ASD) exists in cord blood


A new study led by UC Davis MIND Institute researchers found a distinct DNA methylation signature in the cord blood of newborns who were eventually diagnosed with autism spectrum disorder (ASD).

This signature mark spanned DNA regions and genes linked to early fetal neurodevelopment. The findings may hold clues for early diagnosis and intervention.

“We found evidence that a DNA methylation signature of ASD exists in cord blood with specific regions consistently differentially methylated,” said Janine LaSalle, lead author on the study and professor of microbiology and immunology at UC Davis.

The study published Oct. 14 in Genome Medicine also identified sex-specific epigenomic signatures that support the developmental and sex-biased roots of ASD.

The U.S. Centers for Disease Control and Prevention (CDC) estimates that one in 54 children are diagnosed with ASD, a complex neurological condition linked to genetic and environmental factors. It is much more prevalent in males than females.

The role of the epigenome in DNA functioning

The epigenome is a set of chemical compounds and proteins that tell the DNA what to do.

These compounds attach to DNA and modify its function. One such compound is CH3 (known as the methyl group) that could lead to DNA methylation.

DNA methylation can change the activity of a DNA segment without changing its sequence. Differentially methylated regions (DMRs) are areas of DNA that have significantly different methylation status.

The epigenome compounds do not change the DNA sequence but affect how cells use the DNA’s instructions. These attachments are sometimes passed on from cell to cell as cells divide. They can also be passed down from one generation to the next.

The neonatal epigenome has the potential to reflect past interactions between genetic and environmental factors during early development. They may also influence future health outcomes.

Finding factors in fetal cord blood that might predict autism

The researchers studied the development of 152 children born to mothers enrolled in the MARBLES and EARLI studies. These mothers had at least one older child with autism and were considered at high risk of having another child with ASD.

When these children were born, the mothers’ umbilical cord blood samples were preserved for analysis. At 36 months, these children got diagnostic and developmental assessments.

Based on these, the researchers grouped the children under “typically developing” (TD) or “with ASD.”

The researchers also analyzed the umbilical cord blood samples taken at birth from the delivering mothers. They performed whole-genome sequencing of these blood samples to identify an epigenomic signature or mark of ASD at birth.

They were checking for any patterns of DNA-epigenome binding that could predict future ASD diagnosis.

They split the samples into discovery and replication sets and stratified them by sex. The discovery set included samples from 74 males (39 TD, 35 ASD) and 32 females (17 TD, 15 ASD). The replication set was obtained from 38 males (17 TD, 21 ASD) and eight females (3TD, 5 ASD).

Using the samples in the discovery set, the researchers looked to identify specific regions in the genomes linked to ASD diagnosis. They tested the DNA methylation profiles for DMRs between ASD and TD cord blood samples.

They mapped the DMRs to genes and assessed them in gene function, tissue expression, chromosome location and overlap with prior ASD studies. They later compared the results between discovery and replication sets and between males and females.

Cord blood to reveal insights into genes related to ASD

The researchers identified DMRs stratified by sex that discriminated ASD from TD cord blood samples in discovery and replication sets. They found that seven regions in males and 31 in females replicated, and 537 DMR genes in males and 1762 DMR genes in females replicated by gene association.

These DMRs identified in cord blood overlapped with binding sites relevant to fetal brain development. They showed brain and embryonic expression and X chromosome location and matched with prior epigenetic studies of ASD.

“Findings from our study provide key insights for early diagnosis and intervention,” LaSalle said. “We were impressed by the ability of cord blood to reveal insights into genes and pathways relevant to the fetal brain.”

The researchers pointed out that these results will require further replication before being used diagnostically. Their study serves as an important proof of principle that the cord blood methylome is informative about future ASD risk.

Autistic disorder or a broader form of autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by difficulties in social interaction, delayed development of communication and language, repeated body movements, and impaired intelligence development, first described by psychiatrist Leo Kanner in 1943 [1,2].

The prevalence of typical autism and ASD is approximately 5.5–20 and 18.7–60 per 10,000 individuals, respectively [3]. Moreover, ASD has increased steadily since the term was coined, with the prevalence of autism worldwide currently at 1%–2% [4,5,6,7].

This phenomenon is partly due to an increase in awareness and the development of the Mental Disorders Diagnosis and Statistical Manual (DSM) criteria, starting with schizophrenia which began in 1952, and the development of key diagnostics, which currently deal with various mental disorders [8].

In addition, about 31% of ASD patients showed intellectual disabilities [9] and 20%–37% of them were known to have epilepsy [10,11]. Moreover, ASD is often accompanied by psychiatric or other medical problems, including anxiety disorders, depression, attention deficit hyperactivity disorder, sleep disorders, and gastrointestinal problems [12,13,14].

So far, many theories about ASD etiology and pathogenesis have been proposed, but it is said to be related to the interaction of genetic and environmental factors [15,16]. The concordance rate of ASDs in monozygotic twins (92%) was much higher than that in dizygotic twins (10%), indicating that genetic factors are more likely to contribute to ASD than environmental factors [17].

Genome-wide association and microscopy analysis have identified many different loci and genes that are associated with the etiology of ASD. However, although many genetic and epigenetic risk factors have been suggested, no clear pathogenesis and specific diagnostic markers for ASD have been identified.

Accumulated evidence has also demonstrated an important role of epigenetic factors, such as DNA methylation, in ASD etiology [18]. To better understand the molecular basis of ASD, we describe the genetic and epigenetic epidemiology along with the environmental risk factors underlying the etiology of ASD (Figure 1).

An external file that holds a picture, illustration, etc.
Object name is jcm-09-00966-g001.jpg
Figure 1
Comprehensive overview of the diverse etiology of autism spectrum disorder (ASD). Although definitive etiology and pathogenesis underlying ASD have not yet been identified, accumulated evidence has identified various risk factors, including environmental, genetic, and epigenetic factors.

Environmental and Prenatal Factors that Cause ASD

Viral Infection

As ASD is a neurodevelopmental disorder, it is commonly considered a genetic disorder. However, there are many studies that support the idea that environmental factors can be a major cause of ASD. Most of these factors are due to the prenatal period, which can be affected by environmental changes within the parental body [19].

During pregnancy, the maternal body becomes immunosuppressed, which makes the mother and the developing embryo susceptible to many infectious agents [20]. Similarly, it has been consistently suggested that parental viral infections are associated with the development of autism in their offspring [21,22,23].

Among the infectious diseases, some have been specifically pointed out as contributing to infantile autism when infection occurs during the first trimester of pregnancy [24]. These diseases include rubella [21,25,26,27], measles, mumps [21,26,27], chicken pox [21,28], influenza [21,23], herpes simplex virus [29], pneumonia, syphilis, varicella zoster [30], and cytomegalovirus [27,31,32,33].

Of note, cytomegalovirus is known to cause permanent neurological damage in about 10%–20% newborns when the mother is infected [34]. Moreover, bacterial infection during the second trimester of pregnancy has also been suggested to cause autism of infants [21,22]. In some cases, autoimmune diseases of parents were shown to be related to infantile autism [35,36]. Animal model studies have also shown that maternal infections activated the immune system, which eventually affects fetal brain development [37,38].

Parental Age

We have shown that maternal infection is directly related to the fetal pathological status since the baby is developed and nourished within the maternal body [20]. Similarly, the age of pregnant women and paternity were suggested as one of the most plausible contributors to increasing the risk of autism [39,40,41,42,43,44,45].

Meta-analysis for the correlation between maternal age and autism was analyzed by Sandin et al. A maternal age < 20 showed a lower risk (the relative risk for autism was 0.76) for autism compared to a maternal age between 25 and 29. On the other hand, the relative risk for mothers aged 35 or over compared to mothers aged 25 to 29 was 1.52 [46]. Reichenberg et al. reported a population-based study showing that the risk for autism began to increase at the paternal age of 30 and continued to increase after the age of 50 [42].

Paternal ages above 55 had at least twice the risk to have a child with autism, compared with those below 50 [42]. Moreover, it is well known that the older the parents, the higher the chance of miscarriage [47,48,49,50], fetal death [47,51,52], childhood cancers [53,54], and schizophrenia [55,56,57,58]. This is thought to be due to an increase in de novo genetic mutation during germ cell development in the aging process [59,60]. The effect of parental age on various diseases has been supported by many studies, and the correlation between parental age and autism seems to be one of the most acceptable factors causing autism [47,53,55,59].

Zinc Deficiency

The physiological function of zinc was first identified in the study of carbonic anhydrase [61]. Currently, more than 300 zinc-related enzymes have been discovered, including isoenzymes [62]. Zinc, as a cofactor in metalloenzymes, plays a catalytic role mostly in the transformation of substrates by aiding the formation of hydroxide ions at neutral pH or through Lewis acid catalysis [63,64].

Zinc is now known to be an essential trace element that plays a role in the immune system, protein synthesis, and wound healing [65]. Moreover, zinc has been known to play a role in forming the zinc finger motif of proteins and binding them to DNA, suggesting that zinc is also involved in regulating gene expression [66]. Zinc also supports fetal growth and development during pregnancy and the development of children [67,68].

Therefore, prolonged deficiency of zinc during pregnancy might lead to diverse dysfunction of embryonic growth, especially neurodevelopment [69,70,71]. Research on the relationship between zinc and autism began with reports of the metal ion’s involvement in neurodegeneration and dysfunction [72,73,74,75,76,77]. Since metal toxicity was shown to cause damage to the central nervous system [71,72], it was expected that an excess of zinc could cause damage to the nervous system [78,79,80,81,82,83,84,85].

A recent study also suggests that a toxic metal uptake and deficiency of essential elements increase the risk of ASD [86]. It has been noted that zinc interacts with β-amyloid and its precursors, which are crucial factors for the degenerative process of the brain [73,85,87,88,89,90,91].

Synaptic morphology and function were associated with autism, schizophrenia, and Alzheimer’s disease. The normal function of synapses depends largely on the molecular setting of the synaptic proteins, including ProSAP/Shank proteins, which function as scaffolding molecules for protein–protein interaction at postsynaptic density. ProSAP/Shank localization to postsynaptic density is induced by increased levels of zinc [92,93,94].

Thus, zinc deficiency was shown to dysregulate ProSAP/Shank and postsynaptic density in vivo and in vitro [94]. Several reports suggested that mutation in ProSAP/Shank could lead to ASD [95,96,97]. Moreover, ProSAP/Shank proteins, including ProSAP1/Shank2 and ProSAP2/Shank3, have a C-terminal sterile alpha motif, to which zinc can bind [98]. Thus, a lack of zinc prevents the zinc-dependent ProSAP/Shank proteins from playing a normal role in the formation of the scaffold structure. This leads to synaptic defects and can also lead to autism [94,99].

Moreover, the relation of zinc uptake and the expression of Shank3 regarding autism has been studied recently [100]. This study only included participants with genetically confirmed diagnosis of Phelan McDermid Syndrome (PMDS) with deletion of Shank3 gene [100]. The study showed that low Shank3 levels resulted in abnormally low zinc transporter, which led to low zinc concentration [100].

Statistical data also suggests the close relationship between zinc deficiency and infantile autism [101]. Of children between 0 and 3 years of age with autism, 43.5% (251/577) were zinc deficient in males and 52.5% (62/118) in females [101]. Among autism children from 4 to 9 years old, high rates of zinc deficiency were still found in males (28.1%) and females (28.7%) [101]. An animal-based study of Shank3+/- and Shank-/- transgenic mouse compared with prenatal zinc-deficient autism mouse model, which are offspring from zinc-deficient diet fed mice, showed diverse brain region abnormalities in different models of ASD [102]. However, the role of zinc deficiency on autism is still controversial.

Sweetman et al. tested blood sample of 74 ASD children and claimed that zinc deficiency was not related to ASD [103,104]. Another recent report also suggested that zinc deficiency may not be micronutrient deficiency during pregnancy but may be a compensatory mechanism to prevent exposure to air pollutants during fetal development [103,104].

reference link : https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7230567/

More information: Charles E. Mordaunt et al, Cord blood DNA methylome in newborns later diagnosed with autism spectrum disorder reflects early dysregulation of neurodevelopmental and X-linked genes, Genome Medicine (2020). DOI: 10.1186/s13073-020-00785-8


Please enter your comment!
Please enter your name here

Questo sito usa Akismet per ridurre lo spam. Scopri come i tuoi dati vengono elaborati.