Maps of a mouse kidney done with  CytoSPACE and two other methods show the  greater resolution offered by the new method

Most people aren’t exactly the same everywhere they go. People act in ways that are appropriate for their environment, for the people they interact with. At work they have certain sorts of conversations, while at home, they talk about other things. On vacation, they may ride a zipline or lie on the beach, activities they do neither at home nor at work.

Cells in the human body are much the same. How a particular kind of cell behaves, the proteins it produces, the signals it sends out or receives, are highly dependent on the kind of environment that cell finds itself in. Which makes all the difference for understanding biology and treating disease.

“Cells don’t exist in isolation,” said Aaron Newman, PhD, an Assistant Professor of Biomedical Data Science, a member of the Stanford Institute for Stem Cell Biology and Regenerative Medicine, and the senior author of the study. “They exist in communities that are communicating with them and shaping those cells in terms of what they do. Cells have to be understood based on their context and surroundings.”

Newman and his colleagues have now developed a new method for profiling activity of single cells while also identifying exactly where those cells are geographically in a sample. The method, published March 6, 2023, in the journal Nature Biotechnology, is in many ways more powerful than other methods and will provide researchers with a significant new tool for drug discovery and the development of improved diagnostics, Newman said. 

Newman and his colleagues are not alone in realizing the importance of spatial information in biology. “In the last few years there have been really rapid advances in the development of spatial assays,” Newman said. But each method has its own limitations. 10x Visium is a tool that allows researchers to do an analysis of all the genes being expressed by cells, but the location of each gene’s activity can only be narrowed down to within a spot 55 microns wide, an area that might hold dozens of individual cells. The result resembles a crime “heat map” of a city—useful for knowing generally in which areas crimes will occur, but not much use in identifying individuals who will commit crimes. 

Another method, called Vizgen MERSCOPE, very accurately pinpoints the exact location of every cell, but researchers can only look at the activity of 500 genes within each cell. Since there are over 20,000 genes that might be activate in any given cell, MERSCOPE is unable to gather information about differences in cell activity that might be very important. “It’s like coming to a foreign city and trying to get to know someone using only a phrase book,” Newman says. “You can gather enough basic information to get a general idea of the person, but you can’t ask the questions that might reveal, for instance, that this person you are meeting was also born on a farm and shares an interest in modern agricultural technology.” 

What Newman set out to do was combine these features and generate accurate, high-resolution spatial information while also being able to look at the activity of all 20,000 genes in each cell. He and his team accomplished this by creating a new method called CytoSPACE, which uses mathematical techniques to combine spatial assays like Visium and MERSCOPE along with an existing technology for measuring gene activity in individual cells.

As mentioned before, cells will change their behavior depending on how near or far they are from other specific kinds of cells. Some of these differences are well documented. For instance, immune cells that are cheek by jowl with cancer cells experience “immune exhaustion” and put the brakes on their immune activities in defined ways. (This discovery, and the discovery of how to take the brakes off these exhausted cells and revitalize their cancer-fighting abilities, garnered a Nobel Prize in 2018.) By measuring such changes in cell state, the researchers were able to tell how far certain cells were from certain other cells. 

 CytoSPACE starts with a low resolution map of a sample that provides information on the few dozen cells a 55 micron-wide cluster, and then adds information about how near of far some of those cells are from each other. When the researchers gave that information to a computer program that sorts through all possible configurations, the correct configuration pops out. “It’s like puzzling out a map of a foreign country if you only know the distances between the major cities,” Newman says. “Once you get those pieces put in place, the remaining pieces can only fit in certain ways.”

Being able to piece together the exact organization of cells in a sample, and being able to look at the activity level of all the genes in each cell, will open avenues to basic biological discoveries and possible new therapies for disease, Newman said. “For instance, if we didn’t know anything about cell exhaustion and how to reverse that to treat cancer, that information would pop out of this sort of high-resolution, spatial analysis of cancer and immune cells,” he said. “I’m sure there are many other cell behavioral changes, of which we know nothing about now, just waiting to be discovered once investigators start creating these information-rich, high-resolution maps.”