If the genome is an instruction manual, then the chemical modifications that comprise the epigenome are like highlighters and sticky notes. Each epigenetic mark tells cells which parts of the manual to read to carry out their functions. In complex organs like the brain, where thousands of cell types are carefully organized into intricate neural structures with essential functions, the spatial patterns of these epigenetic marks matter. Alterations in these patterns can be a sign of disease.
New technologies measure epigenetic marks on histones and DNA at nearly single-cell resolution in the brain.1,2 However, these tools can be expensive for labs to use, especially when studying diseases across large groups of patients and controls.
“Because of the cost and labor, the single-cell approach is not practical,” said Chang Lu, a chemical engineer at Virginia Tech. “We thought there has to be a lower-cost, simpler approach to do that.”
To address this, Lu’s team came up with an alternative approach to scan spatial patterns of epigenetic marks in the brain by dissecting thin slices and running lower-cost bulk epigenomic methods.3 In doing so, they avoid reading millions of cellular instruction manuals cover to cover and instead read a single summary of all the manuals on a shelf. They published their work in Cell Reports Methods and dubbed the approach epigenomic tomography, inspired by a previously published method called RNA tomography that itself references the family of methods to image large objects by iterating over slices.4
“Although we sacrifice spatial resolution compared to a single-cell approach, we can look at bigger regions, which is going to be very costly and hard to do with single-cell approaches,” Lu said. “There are processes that happen in a larger region of the brain. Even if you sacrifice a little bit of spatial resolution, you’re still going to be able to see them.”
To showcase their method, the researchers isolated the neocortex region of the mouse brain. After thinly cutting each brain sample into 0.5-millimeter-thick slices, they separated the neurons from the glial cells and amalgamated thousands of cells in each cell type together. With so few cells, the researchers turned to a method called MOWChIP-seq, which is short for microfluidic oscillatory washing–based chromatin immunoprecipitation followed by sequencing, that they had previously developed to identify epigenetic marks from small numbers of cells.5
In each brain slice, the researchers measured specific histone modifications—similar to only focusing on text highlighted in a particular color in a book. But they didn’t have a way to track how these marks varied across space; that is, to see whether the text highlighted in the manuals on one shelf differed from the text highlighted on other shelves.
To do this, the team devised an approach to cluster the epigenetic marks across slices, figure out which marks cooccured spatially, and link them to neural processes mediated by nearby genes. This provided the basis for a neocortical map of where certain processes may occur or become dysregulated in disease.
To test their approach, Lu and his team turned to a model for studying seizures, which can disrupt the brain’s function in spatially defined ways. When they used epigenomic tomography to study the brains of mice with and without induced seizures, they found spatial differences in the epigenetic marks between the two groups. “That certainly gives us some confidence that we can use this type of approach to study processes in the brain,” Lu said. His team is working on extending the method to even thinner slices that will help them achieve even higher resolution. They hope to use their new method to study human brains to identify epigenetic changes linked to diseases such as schizophrenia and addiction.
The main advantage of this method is the reduction in cost, said Marc Beyer, an immunologist at the German Centre for Neurodegenerative Diseases who was not involved in this study. “If you were to do what the paper has done—this cross-sectional analysis throughout the brain—on a single-cell level, currently we’re not able to do it cost-wise,” he said. “I think it’s a worthwhile effort.”
Although Lu estimates that epigenomic tomography is one to two orders of magnitude cheaper than applying single-cell-resolution methods, Gonçalo Castelo-Branco, a biochemist at the Karolinska Institute who was not involved in this study, wanted to see a rigorous side-by-side comparison to confirm the cost reduction that comes with epigenomic tomography is worth the loss in resolution. “The single-cell aspect is quite important,” he said. Aggregating all the cells in a brain slice may obscure many different cell types with varied functions. For example, there are different types of neurons even in a single layer of the cortex, Beyer notes, which may play disparate roles in disease.
Beyer pointed to Alzheimer’s disease as a potential application because there are large-scale spatial changes in the brain that may correlate with epigenomic features. Castelo-Branco also imagines that epigenomic tomography could be useful for exploratory studies of an organ where researchers don’t know which region to focus on. “[With single-cell methods] we cannot study a large number of samples, and we are limited to one region,” he said.
Lu’s team is still exploring the optimal applications for their method, and by studying many different brain disorders, he hopes they can figure out the right level of spatial resolution required to capture the patterns in each condition. “We feel that you need to have the right technology for the specific question that you’re asking,” he said. “You don’t necessarily always need the most powerful one, because there is a price to pay for that.”
References
1. Lu T, et al. Spatially resolved epigenomic profiling of single cells in complex tissues. Cell. 2022;185(23):4448-4464.e17.
2. Deng Y, et al. Spatial-CUT&Tag: Spatially resolved chromatin modification profiling at the cellular level. Science. 2022;375(6581):681-686.
3. Liu Z, et al. Epigenomic tomography for probing spatially defined chromatin state in the brain. Cell Rep Methods. 2024;4(3):100738.
4. Junker JP, et al. Genome-wide RNA tomography in the zebrafish embryo. Cell. 2014;159(3):662-675.
5. Zhu B, et al. MOWChIP-seq for low-input and multiplexed profiling of genome-wide histone modifications. Nat Protoc. 2019;14(12):3366-3394.
Discussion about this post