Next-generation sequencing provides novel insights into the mechanisms underlying autism and myotonic muscular dystrophy

In a preprint uploaded to the Research Square* server, researchers used a combination of in vivo mice models and next-generation sequencing techniques to elucidate the pathomechanistic connection between autism and myotonic muscular dystrophy type 1.

Their results suggest that CTG expansion in the DMPK gene and muscleblind-like factor inhibition induces developmental mis-splicing, which in turn cause myotonic muscular dystrophy type 1 and autistic social deficits.

Study: Autistic traits in myotonic dystrophy type 1 due to MBNL inhibition and RNA mis-splicing. Image Credit: SewCreamStudio/Shutterstock.com

*Important notice: Research Square publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.

What are ASD and DM1?

Autism spectrum disorder (ASD) is a neurodevelopmental disability affecting the brain. It is characterized by social communication or interaction problems, and restricted or repetitive behaviors or interests.

The condition affects one in 36 children, almost all of whom (95%) suffer from physical or mental comorbidities. ASD is a clinically heterogeneous condition, and despite hundreds of genes having been associated with the disability, the molecular mechanisms underlying ASD remain unknown.

Previous whole-genome sequencing studies have identified tandem repeat mutations as associated with ASD, contributing ~4% to ASD risk. Tandem repeats are sequences of two or more DNA bases repeated numerous times in a head-to-tail manner on a chromosome.

They are usually present in the non-coding region of DNA, but when these mutations arise in coding genes, they may have severe consequences, including myotonic muscular dystrophy type 1 (DM1).

The myotonic dystrophy protein kinase (DMPK) gene is one of many responsible for proper muscle development and function. Research has shown that mutations in this gene correlate with autism.

One of the most common mutations affecting one in every 2,100 newborns is a CTG expansion (CTG^exp) in the gene’s 3′ untranslated region (3’UTR). These DMPK 3’UTR CTG^exp mutations cause DM1, a variable genetic condition characterized by muscle weakness and wasting, myotonia, cataract, and often cardiac conduction abnormalities.

When present, DMPK 3’UTR CTG^exp  RNAs present in DM1 neurons and muscles result in the inhibition of muscleblind-like (MBNL) RNA-binding proteins through the formation of biomolecular structures called RNA foci.

Under normal conditions, MBNL proteins act as trans-acting factors, regulating alternative splicing (AS) during fetal development. Their inhibition is suggested to result in ‘mis-splicing’ correlating with specific DM1 clinical profiles, especially myotonia.

DM1 and ASD have been identified as comorbidities, with the latter correlating with the age of onset of the former. Like in DM1, developmental mis-splicing occurs in neuronal microexons (miEs), identified in ~30% of idiopathic ASD brains.

These microexons control protein-protein interaction networks and encode post-translational modification sites, which are crucial in normal nervous system development. Mutations in miEs have been shown to trigger ASD-like symptoms in murine models, including social avoidance.

While the processes and mechanisms underlying DM1 mis-splicing have been well documented, ASD mis-splicing remains poorly understood. Further, molecular mechanisms linking ASD and DM1 have hitherto been unstudied.

About the study

In the present preprint, researchers employed integrative genomic- and transcriptomic techniques (next-generation sequencing) in combination with in vivo studies using multiple genetically modified DM1 murine models to elucidate the molecular mechanisms underpinning mis-splicing in DM1-associated ASD.

Their analyses focused on neuronal miEs mis-splicing in ASD-risk genes modulated by MBNL proteins, specifically MBNL1 and MBNL2, during murine and human brain development.

Researchers first analyzed human prefrontal cortex RNA-seq data from DM1-positive (study) and DM1-negative (control) cohorts. They compared the change of percent spliced (DPSI) in 38 ASD-relevant gene sets retrieved from previous studies and found that 76% of their dataset showed significant enrichment of mis-spliced events in study cohorts.

They further identified mis-splicing in the Duchenne muscular dystrophy (DMD) gene, a comorbidity of ASD. Their analysis revealed that CTG^exp size was significantly correlated with the number of mis-spliced events in these genes, verifying that DMPK 3’UTR CTG^exp prefrontal cortex mutations inhibit AS in ASD-relevant genes.

To elucidate the involvement of MBNL protein regulation in the prefrontal cortex, genetically modified mice models’ (Mbnl cDKO) frontal cortex samples were compared to those from wild-type (WT) mice.

Results showed that, similar to humans, 61% of ASD-relevant gene sets were mis-splice-enriched in Mbnl cDKO murine frontal cortexes. Of these, 55 ASD-risk genes overlapped between humans and mice with ASD.

Transcriptomic analyses of neuronal miEs were employed to evaluate the extent of mis-splicing in these microexons. Comparisons between WT and Mbnl cDKO miEs revealed that the former had significantly fewer (4%) mis-spliced events than the latter (10%).

When these analyses focused on ASD-risk genes, these contrasts were as high as 35%. Comparative in silico modeling of miE-encoded protein structure was employed to assess if mis-splices in miEs can impact peptide structures.

Modeling results suggest that internal or C-terminal protein structures might be altered if miEs are mis-spliced.

Gene expression data from five mammalian brains (including humans) was analyzed to infer the role of MBNL proteins and their AS in ASD-risk genes during organismal development.

“Our analysis showed an evolutionarily conserved increase of MBNL2 expression during neonate/P0 to middle childhood/P14 brain development. Although MBNL1 expression increases simultaneously, its expression in the developed brain is approximately 3-fold lower than MBNL2.”

Reverse transcription-polymerase chain reaction (RT-PCR) splicing analysis of MBNL2 in WT and Mbnl cDKO mice revealed that the loss or alteration of MBNL2 caused severe alterations in AS of ASD-risk genes in different parts of Mbnl cDKO brains, especially the hippocampus.

Finally, social interaction in vivo tests were conducted on WT, Mbnl2 knockout, and Dmpk 3’UTR CTG^exp knockin genetically modified mice.

Three-chamber tests which included familiar mice, stranger mice, and novel inanimate objects, were employed to test the sociability of these animals. WT mice spent significantly more time with strange mice than their genetically modified counterparts.

There was no difference in time spent with novel inanimate objects, highlighting that sociability, and not curiosity, was impacted by alterations in MBNL2 and DM1.

Conclusions

In the present preprint, researchers utilized next-generation sequencing technologies to elucidate the pathomechanistic relevance of mis-splicing in DM1-associated ASD.

Their results revealed that tandem CTG repeats in the DMPK gene significantly increase the occurrence of both DM1 and ASD via inhibition of MBNL proteins, especially MBNL2.

The length and number of tandem repeats were positively correlated with the clinical severity of DM1 and ASD. These results suggest that MBNL protein inhibitions alter neuronal development, resulting in these diseases.

In vivo comparisons between genetically modified murine models (altered to reflect these ASD-relevant genotypes) and their regular counterparts to evaluate sociability confirmed the transcriptomic and genomic analyses.

They revealed that alternations or mis-splicing of CTG tandem repeats and MBNL proteins result in significantly reduced sociability in mice.

“Our results provide insights into the molecular mechanism underlying DM1-associated ASD where developmental mis-splicing of ASD-linked genes arises by loss of MBNL activity due to CUG repeat expansions. Understanding this pathomechanistic connection provides an opportunity for greater in-depth investigations of mechanistic threads in autism.”

*Important notice: Research Square publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.