Researchers Analyze CD8+ T Cell Types in Unprecedented Detail
By Andreas von Bubnoff
A new technology that can simultaneously measure about twice as many different cellular markers than flow cytometry has allowed researchers to analyze the different types of CD8+ T cells in healthy individuals in unprecedented detail. Evan Newell, a research associate in Mark Davis’ lab at Stanford University, and colleagues analyzed CD8+ T cells from six healthy individuals for 16 cell surface markers, nine functional markers such as cytokines, and six antigen specificities for three viruses: Epstein-Barr virus (EBV), flu virus, and cytomegalovirus (CMV). This showed, the authors say, that CD8+ T cells exhibit a much greater degree of complexity than previously appreciated (1).
Simultaneous analysis of this many markers has only become possible recently with the development of a new method called cytometry by time-of-flight (CyTOF), or mass cytometry, which allows for the simultaneous analysis of up to 40 different markers in living cells, Newell says. With flow cytometry, researchers can simultaneously analyze up to about 20 different markers, but according to Newell, even that is difficult with that technology. “Above 10 it gets really dicey,” he says.
Both methods use antibodies that bind to different markers, but CyTOF differs from flow cytometry in that the antibodies are tagged with heavy metal atoms that are usually not found in living cells, rather than with fluorescent markers. The labeled cells are then analyzed by mass spectrometry, which can distinguish between more markers at the same time than flow cytometry because their signals overlap less than fluorescent signals.
A study last year used CyTOF to analyze human bone marrow immune cells in unprecedented detail (2), but Newell and colleagues used this approach to take a detailed look at CD8+ T cells. To visualize the many different marker combinations, the researchers projected their 25-dimensional data set into a three-dimensional data set, choosing as the three dimensions the parts of the data that accounted for most of the variation. In other words, they projected the data in such a way that they would cast the biggest shadows in each of the three dimensions.
In the resulting data cloud, cells that are closer to each other are more similar in the levels and kinds of markers they express than cells that are further apart. Although the analysis is unbiased, the shape of the cloud was similar in all six healthy individuals analyzed in the study. Newell compares it to a Y with a stem that’s bent (see image, below).
Many of the clusters of cells in the cloud correspond to known types of CD8+ T cells. For the first time, the analysis showed that some of these previously known cell types, such as the central and effector memory CD8+ T cells, are not separate cell types, but connected with each other, suggesting the existence of intermediate cell types between them, according to Newell. “The memory cells are all one big continuum,” he says. The cloud might also include the possible location of stem cell-like CD8+ T cells, a new type of CD8+ memory T cell recently identified that can multiply and regenerate better than other CD8+ memory T cells (see Research Briefs, IAVI Report, Sep-Oct. 2011). These “stem cell memory” cells might be located between the cluster of naive and central memory CD8+ T cells in the cloud, Newell says.
Cells specific for EBV, CMV, or flu virus occupied different parts of the cloud, but the clusters they occupied were surprisingly large, Newell says, suggesting that the CD8+ T-cell response to a certain type of virus is surprisingly variable.
Even though CyTOF can analyze more markers than flow cytometry, traditional flow cytometry still has its place. CyTOF is much slower and kills the cells during the analysis, while flow cytometry keeps the cells alive, making it possible to sort the cells for additional experiments, Newell says.
Nicholas Haining, an assistant professor of pediatrics at the Dana-Farber Cancer Institute and Harvard Medical School who wrote a commentary on the study in the same issue of Immunity (3), says it shows that the CD8+ T cell compartment is even more complex than previously thought. The study is also one of the most compelling pieces of evidence yet, he says, that there is a continuum of CD8+ T cell types rather than separate “buckets” of cell types. “That was very clearly demonstrated in this paper,” says Haining, who wasn’t connected to the study.
At the same time, he adds, the paper shows that the CD8+ T cells don’t just randomly express all possible combinations of functional markers. “There is an enormous amount of heterogeneity, but the fact that fewer than all possible permutations of those markers [are expressed], implies that there is some programming that drives the heterogeneity, and that it’s not just a random arrangement of cell functions.”
One question now is whether all the cells that can be differentiated from each other using CyTOF truly have different biological functions. “It will be even more challenging to try and disentangle what all of these cells are doing than we previously thought,” Haining says.
1. Immunity 36, 142, 2012
2. Science 332, 687, 2011
3. Immunity 36,10, 2012