Spatial-ATAC-Hi-C: Spatial 3D Genome Architecture and Chromatin Accessibility

Nature Methods has published a paper on Spatial-ATAC-Hi-C, a method that maps DNA organization and the accessibility of its regulatory regions within the same tissue section

The paper was published on September 1. It describes a method that assigns both DNA contacts and the accessibility of regulatory regions to each position in a tissue section. The authors validated it in mouse and human brain tissue, then applied it to samples of glioma, a type of brain tumor.

Inside the nucleus, DNA is packaged with proteins into chromatin. Chromatin forms loops, which can bring a regulatory region close to the gene it controls even when the two are far apart in the genome sequence. An open chromatin region is accessible to regulatory proteins, providing one mechanism through which a cell controls gene activity. In tissue, both measurements must also retain spatial coordinates because neighboring cells can belong to different cell types and follow different gene expression programs.

In July, a map of 86 thousand cell nuclei also combined three-dimensional chromatin contacts with another layer of information: DNA methylation, which consists of chemical marks at specific genomic regions. Spatial-ATAC-Hi-C measures regions containing several cells while preserving their coordinates in the tissue section and profiling chromatin accessibility. This makes it possible to test whether changes in DNA organization, regulatory region accessibility, and gene expression programs occur within the same tissue region.

To obtain spatial coordinates, the authors fixed the tissue to a glass slide, joined DNA fragments that had been in contact inside the nuclei, and passed two perpendicular sets of barcoded microchannels across the section. Fifty horizontal and fifty vertical channels produce up to 2 500 positions. After DNA sequencing, each barcode assigns both DNA contacts and accessible regulatory regions to the corresponding position. Each position covers a square measuring 50 × 50 micrometers and usually contains from 3 to 21 cells.

Using adjacent sections of mouse brain, the authors compared both maps with standard methods. In situ Hi-C measures DNA contacts, while ATAC-seq identifies accessible chromatin regions. Agreement with both independent measurements showed that the combined protocol preserved both signals. In the brain, the map distinguished chromatin loops associated with different neuronal types and anatomical regions.

In astrocytoma and glioblastoma samples, Spatial-ATAC-Hi-C detected changes in genomic copy number and structural rearrangements, which are large changes in the positions of DNA fragments. The authors compared these results with whole-genome sequencing, which reads all DNA, from adjacent sections. In a separate glioblastoma sample, they identified spatially distinct groups of positions with different DNA copy-number profiles. The chromatin contact and accessibility maps showed the same separation. In aging tissue, this method could test whether age-related changes in DNA accessibility, long-range genomic contacts, and cellular composition occur within the same regions.