Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Sep 3;7(36):eabh1663.
doi: 10.1126/sciadv.abh1663. Epub 2021 Sep 3.

Atypical genomic cortical patterning in autism with poor early language outcome

Michael V Lombardo  1   2 , Lisa Eyler  3   4 , Tiziano Pramparo  5 , Vahid H Gazestani  5 , Donald J Hagler Jr  6   7 , Chi-Hua Chen  3 , Anders M Dale  6   7   8 , Jakob Seidlitz  9   10 , Richard A I Bethlehem  2   11 , Natasha Bertelsen  1   12 , Cynthia Carter Barnes  5 , Linda Lopez  5 , Kathleen Campbell  5   13 , Nathan E Lewis  14   15   16 , Karen Pierce  5 , Eric Courchesne  5
Affiliations

Atypical genomic cortical patterning in autism with poor early language outcome

Michael V Lombardo et al. Sci Adv. .

Abstract

Cortical regionalization develops via genomic patterning along anterior-posterior (A-P) and dorsal-ventral (D-V) gradients. Here, we find that normative A-P and D-V genomic patterning of cortical surface area (SA) and thickness (CT), present in typically developing and autistic toddlers with good early language outcome, is absent in autistic toddlers with poor early language outcome. Autistic toddlers with poor early language outcome are instead specifically characterized by a secondary and independent genomic patterning effect on CT. Genes involved in these effects can be traced back to midgestational A-P and D-V gene expression gradients and different prenatal cell types (e.g., progenitor cells and excitatory neurons), are functionally important for vocal learning and human-specific evolution, and are prominent in prenatal coexpression networks enriched for high-penetrance autism risk genes. Autism with poor early language outcome may be explained by atypical genomic cortical patterning starting in prenatal development, which may detrimentally affect later regional functional specialization and circuit formation.

PubMed Disclaimer

Figures

Fig. 1.
Fig. 1.. Subtype differences in total CV and total SA.
Standardized effect sizes (Cohen’s d) are shown for each pairwise group comparison for total CV (A), total SA (B), and mean CT (C). Asterisks indicate statistically significant pairwise group comparisons that survive false discovery rate (FDR) q < 0.05. Good, ASD Good language subtype; Poor, ASD Poor language subtype; TD, typically developing.
Fig. 2.
Fig. 2.. Normative associations between gene expression and SA or CT are absent in ASD Poor.
(A) PLS correlations for each gene coexpression module (rows) and each group (columns). Modules with a black outline are non-zero modules where the correlation between gene expression and SA is significantly non-zero, as indicated by 95% bootstrap CIs not encompassing a correlation of 0. These non-zero modules are the strongest contributors to the PLS relationship. All nonoutlined cells are zero modules that are not sufficiently correlated to SA in a non-zero way (e.g., 95% bootstrap CIs include a correlation of 0). (B) BSRs for each brain region in the GCLUST parcellation. Regions with red BSRs have correlations that manifest in the directionality shown in heatmaps in (A). Brain regions with blue BSRs have correlations where the directionality is flipped relative to the heatmaps in (A). Stronger BSRs indicate regions that are more important in driving the SA LV1 relationship. (C) Similarity in PLS correlations between groups. In these scatterplots, each dot is a coexpression module, and the x and y axes indicate the PLS correlations for different groups. Dots colored in dark red and dark blue indicate the non-zero modules (red for positive correlations and blue for negative correlations), while gray dots indicate zero modules. The scatterplots with the orange outline indicate that the PLS SA LV1 relationship manifests similarly for TD and ASD Good. (D to F) The same as (A) to (C) except that they show the association between gene coexpression modules and CT (CT LV1).
Fig. 3.
Fig. 3.. Atypical association between gene expression and CT that is specific to the ASD Poor subtype.
Atypical association specific to ASD Poor between gene coexpression modules and CT, as described in CT LV2. (A) CT LV2 PLS correlations for each gene coexpression module (rows) and each group (columns). Modules with a black outline are non-zero modules where the correlation between gene expression and SA is significantly non-zero, as indicated by 95% bootstrap CIs not encompassing a correlation of 0. Nearly all of the non-zero modules for CT LV2 are present for the ASD Poor subtype. (B) Lack of similarity in CT LV2 PLS correlations between groups. In these scatterplots, each dot is a coexpression module, and the x and y axes indicate the PLS correlations for different groups. Dots colored in dark red and dark blue indicate the non-zero modules (red for positive correlations and blue for negative correlations), while gray dots indicate zero modules. (C) CT LV2 BSRs for each brain region in the GCLUST parcellation. Regions with red BSRs have correlations that manifest in the directionality shown in heatmaps in (A). Brain regions with blue BSRs have correlations where the directionality is flipped relative to the heatmaps in (A). Stronger BSRs indicate regions that are more important in driving the CT LV2 relationship.
Fig. 4.
Fig. 4.. Cortical patterning along genetic similarity gradients.
(A) Brain BSR maps for SA LV1 (top), CT LV1 (middle), and CT LV2 (bottom). (B) Coarse two-cluster A-P and D-V genetic similarity partitions identified by Chen and colleagues (13, 14). (C) Rank ordering of regions by hierarchical genetic similarity found by Chen and colleagues (13, 14). The rank ordering here defines a genetic similarity gradient for how SA or CT varies across brain regions. Areas rank numbered close together are more genetically similar than regions numbered farther apart. (D) BSRs are differentiated along A-P and D-V axes (SA LV1, top; CT LV1, middle; CT LV2, bottom). (E) BSRs co-vary with genetic similarity gradients (SA LV1, top; CT LV1, middle; CT LV2, bottom).
Fig. 5.
Fig. 5.. Enrichment between PLS non-zero modules and genes involved in prenatal A-P and D-V expression gradients and prenatal cell types.
(A) Cortical brain areas sampled from 12 to 24 weeks after conception from the Development PsychENCODE RNA-seq dataset from Li and colleagues (17). Adjustment-for-confounds PCA (30) was used to isolate (B) A-P (PC1) and (C) D-V (PC2) expression gradients. (D) −log10 P values for enrichment tests of non-zero and zero modules for SA LV1, CT LV1, and CT LV2 for genes isolated from PC1 and PC2. (E to G) Enrichments in prenatal cell types for SA LV1 (E), CT LV1 (F), and CT LV2 (G). Asterisks mark enrichments at FDR q < 0.01. MFC, medial prefrontal cortex; V1C, primary visual cortex; OFC, orbitofrontal cortex; DFC, dorsolateral refrontal cortex; M1C, primary motor cortex; VFC, ventrolateral prefrontal cortex; A1C, primary auditory cortex; ITC, inferior temporal cortex; STC, superior temporal cortex; IPC, posterior inferior parietal cortex; S1C, primary somatosensory cortex.
Fig. 6.
Fig. 6.. Enrichments between PLS non-zero modules and songbird vocal learning or human-specific genes.
(A to C) Enrichments between DE songbird vocal learning genes and non-zero and zero modules for SA LV1 (A), CT LV1 (B), and CT LV2 (C). (D to F) Enrichments between human-specific genes and non-zero and zero modules for SA LV1 (D), CT LV1 (E), and CT LV2 (F). Asterisks mark enrichments at FDR q < 0.01. HS, human-specific.
Fig. 7.
Fig. 7.. Enrichment between PLS non-zero modules and autism-associated genes.
(A to C) Enrichments between different autism-associated gene lists and non-zero and zero modules for SA LV1 (A), CT LV1 (B), and CT LV2 (C). (D to F) Enrichments between DE genes in specific cell types in autism and non-zero and zero modules for SA LV1 (D), CT LV1 (E), and CT LV2 (F). Asterisks mark enrichments at FDR q < 0.01. SCZ DE, DE genes in schizophrenia; BD DE, DE genes in bipolar disorder.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Lombardo M. V., Lai M.-C., Baron-Cohen S., Big data approaches to decomposing heterogeneity across the autism spectrum. Mol. Psychiatry 24, 1435–1450 (2019). - PMC - PubMed
    1. Parikshak N. N., Gandal M. J., Geschwind D. H., Systems biology and gene networks in neurodevelopmental and neurodegenerative disorders. Nat. Rev. Genet. 16, 441–458 (2015). - PMC - PubMed
    1. Collins F. S., Varmus H., A new initiative on precision medicine. N. Engl. J. Med. 372, 793–795 (2015). - PMC - PubMed
    1. Courchesne E., Pramparo T., Gazestani V. H., Lombardo M. V., Pierce K., Lewis N. E., The ASD living biology: From cell proliferation to clinical phenotype. Mol. Psychiatry 24, 88–107 (2019). - PMC - PubMed
    1. Tager-Flusberg H., Kasari C., Minimally verbal school-aged children with autism spectrum disorder: The neglected end of the spectrum. Autism Res. 6, 468–478 (2013). - PMC - PubMed