We've updated our Privacy Policy to make it clearer how we use your personal data. We use cookies to provide you with a better experience. You can read our Cookie Policy here.

Advertisement

Spatial Biology: The Next Revolution in Understanding Health and Disease

Different components of cells stained in varying colors. A scalebar in the top left corner can be seen.
Credit: iStock
Listen with
Speechify
0:00
Register for free to listen to this article
Thank you. Listen to this article using the player above.

Want to listen to this article for FREE?

Complete the form below to unlock access to ALL audio articles.

Read time: 7 minutes

The conventional tools used by biologists to study the transcriptomes of different cell types were originally limited to immunohistochemistry or in situ hybridization – i.e., labeling RNA, DNA or proteins using one or two different colors – making it possible to distinguish between one or two types of cell. With the advent of powerful single-cell RNA sequencing (scRNAseq) technology, biologists were soon investigating every cell within a tissue in unprecedented detail. Now, by combining these single-cell analysis tools with imaging and microfluidics, spatial biology is adding a new layer of information and is set to transform biomedical science.

Putting single-cell analysis in context

Spatial biology uses any technology that detects the location and biological quantity of cellular contents. This could include the transcriptome, the epigenome or anything you can measure with fluorescent in situ hybridization (FISH) technologies or next-generation sequencing (NGS), but for it to be spatial biology, the location has to be involved.


Dr. Ankur Sharma, laboratory head at the Harry Perkins Institute of Medical Research and Curtin University, studies the interplay between fetal development and cancer. For researchers like Sharma, scRNA-seq provided a step-change in understanding which cell types are disease-specific, which are involved in normal physiological development and which cells are involved in disease. However, a major limitation of scRNAseq is that it requires taking tissue, dissociating it into a single cell suspension and analyzing each cell’s mRNA content. This means researchers lose the context of the cell within its tissue, and how it contacts and communicates with its neighbors.


Spatial transcriptomics can be implemented by using laser capture microdissection technology to procure subpopulations of cells under a microscope, but this still only provides bulk-level transcriptomics. “Most users want to be able to measure the transcriptome of individual cells, and so more advanced technologies were developed to achieve that,” explains Kyoung Jae Won, associate professor at Cedars-Sinai Medical Center, who develops analytical tools for interpreting spatial biology data. Most image-based techniques involve designing probes against target RNA or DNA molecules and then carrying out sequential rounds of FISH and imaging.1 The florescent signals of each probe can be precisely mapped back to their location within the tissue enabling a high-resolution view of the transcriptional landscape. In NGS-based techniques, there are efforts underway to increase the density of spatial barcodes to obtain transcriptomic data at sub-cellular resolution.


About five years ago, spatial transcriptomics technologies began to be developed that make it possible to look into each and every cell type in detail. This provided context about where these cells are sitting in the tissue – how close they are, how this proximity changes during disease progression and how their interactions change when treatments work, versus when they fail. “Before that we could only catalog them,” says Sharma, “whereas now we can catalog them with an understanding of what they are doing in the tissue. Now, we can eavesdrop on the private conversations of cells in tissue.”

Milestones in Spatial Biology

Spatially resolved biology allows researchers to study cells in the context of their tissue microenvironment, enabling a fuller appreciation of cellular function. Download this whitepaper to explore the key applications of spatial profiling methods such as in situ hybridization, in situ sequencing and spatial transcriptomics.

View Whitepaper

A step change for molecular biology

Already, spatial biology is enhancing the depth of our understanding of disease mechanisms and interpreting what cellular changes might be clinically meaningful.


“In understanding biology, it’s essential to distinguish between correlation and causation,” says Sharma. “The presence of a cell or a cellular signal does not always tell us whether it is essential for disease. But if we visualize that two cell types always come together and give signal A when a treatment is working, but transmit signal B when therapy fails, this helps in understanding the biological context of cellular communications in therapy response. This is the power of spatial biology.”


Sharma is using spatial transcriptomics to visualize how immunotherapy is impacting communications between cancer and immune cells to predict the response of immunotherapy in liver cancer patients.2 “We have identified that patients with specific spatial patterns of cancer cell communication do not respond to conventional therapies, but immunotherapy blocks this communication and kills cancer cells in these patients.” Sharma’s team is leading an international study where they profile patients at the start of treatment and can offer adjuvant immunotherapy upfront. The goal is to provide proof of principle that if you profile an interaction between these cell types at the beginning of treatment, you don’t have to wait for patients to relapse before you augment or switch therapy, you can nip the recurrence in the bud.


Current limitations in spatial biology technology


Advertisement

Most of the current spatial biology technologies rely on either single molecule FISH followed by imaging or NGS.1 NGS-based approaches were limited in resolution because this depended on how many spatial barcodes you can fit on a slide, says Won: “When NGS based technology first appeared, the size of each spot on an array was about 100 uM, now it is possible to obtain the grid size even less than 1 uM.”


The image-based approaches require preselection of genes to design probes for mRNA species, which limits the number of RNA species that can be detected. “Although some labs say they can measure thousands of mRNA species, it’s still laborious and expensive,” says Won. “And because image-based technologies rely on measuring fluorescent signals that can overlap with each other, it’s not easy to distinguish between mRNA species by looking at the images.”


Advances in the barcoded chip-based spatial technology is equivalent to the semiconductor arms race: who can make the smallest semiconductor and therefore the most powerful computer within a small space. “The smallest grid you can print is the smallest detail you can capture from a cell,” says Sharma. “You are talking about the difference between visualizing two or three neighborhoods in a city versus a house within a neighborhood. The latest technology has enabled us to visualize these cellular neighborhoods to see which neighbors are there and what they are communicating to each other.”

New ways of analyzing the data

Advertisement

A further challenge in spatial biology is developing the algorithms required to stitch back together each piece of the puzzle from every barcode on a slide. Won’s research is focused on doing just that. He has already developed an algorithm to detect tissue architecture from spatial transcriptomics or spatial epigenomics data, and another to understand cell-cell interactions. 3,4*


“Although there are some tools that attempt to merge neighboring locations if they are transcriptionally similar, we had the idea to treat the transcriptomes as an image and apply image processing technology to it,” explains Won. Before they could do this, they had to convert the transcriptome into an image using dimension reduction algorithms such as principal components analysis (PCA). This allowed them to represent 20,000 genes in a three-dimensional (3D) space, with each dimension transferred to a color channel. Applying image processing software to this visual data allowed them to detect tissue architecture much more accurately and quickly from publicly available brain data and mouse embryo data. 3,4*


The next opportunity, says Sharma, is to turn two-dimensional (2D) spatial biology into 3D. “We’ve started to scratch the surface of studying 3D interaction of cells, but this is partly limited by sequencing cost. When sequencing becomes cheaper, we can do these experiments in different tissue layers. But the second component is how are you going to stitch all this together? I think for this, the community needs to develop more computational tools to understand [the] 3D dynamics of these cell types.”

Human Whole Transcriptome Atlas

Traditional gene expression technologies are unable to capture heterogeneity of the transcriptome with spatial context. Download this whitepaper to discover a solution that is designed for comprehensive profiling of spatial biology, provides superior sensitivity to detect thousands of unique human genes in <50 μm regions and delivers spatial analysis of any target in any tissue.

View Whitepaper

Another advance will be to apply the technology to live cells, as spatial biology is currently limited to fixed cells, such as tumor biopsies. Recently, a new method for scRNA-seq in single, live cells was published,5 which makes it possible to extract mRNA from single cells without killing the cell and could be a step towards spatial biology in live tissues.


Sharma believes we are already in a second phase for spatial transcriptomics, one where it is entering clinical translation: “I think it will have a huge impact on precision medicine because clinicians and pathologists have traditionally been trained to look at the broad morphological information from slides, but now they can look into the slides with this profound information about all the different cell types. I think that's something that will revolutionize the field,” he says.


Although spatial biology is starting to show clinical potential in theory, the field will have even broader potential for biology as scientists increase its resolution, of the number of genes that can be detected, and the availability of computational algorithms. “We are starting to study a new biology, putting another layer of knowledge to conventional molecular biology, says Won. “What kind of answer can we get out of it? We don't yet know; we are only just opening the lid of the box. But it will be a powerful technology and people will find use for it everywhere.”


*This article is a preprint that is yet to be peer-reviewed. Results are therefore regarded as preliminary and should be interpreted as such. Find out about the role of the peer review process in research here. For further information, please contact the cited source.


References


1.       Lewis SM, Asselin-Labat ML, Nguyen Q, et al. Spatial omics and multiplexed imaging to explore cancer biology. Nat Methods. 2021;18(9):997-1012. doi: 10.1038/s41592-021-01203-6


2.       Sharma A, Blériot C, Currenti J, Ginhoux F. Oncofetal reprogramming in tumour development and progression. Nat Rev Cancer. 2022;22(10):593-602. doi: 10.1038/s41568-022-00497-8


3.       Martin PCN, Kim H, Lövkvist C, Hong BW, Won KJ. Vesalius: high-resolution in silico anatomization of spatial transcriptomic data using image analysis. Mol Syst Biol. 2022;18(9):e11080. doi: 10.15252/msb.202211080


4.     Kim H, Lövkvist C, Martin P, Kim J, Won KJ. Detecting cell contact-dependent gene expression from spatial transcriptomics data. bioRxiv. 2022.02.16.480673; doi: 10.1101/2022.02.16.480673 (This article is a preprint and has not been certified by peer review)*


5.     Chen W, Guillaume-Gentil O, Rainer PY, et al. Live-seq enables temporal transcriptomic recording of single cells. Nature. 2022;608(7924):733-740. doi: 10.1038/s41586-022-05046-9