Unlocking the mysteries of cellular senescence with single cell and spatial tools
In his novel, The Marble Faun, Nathaniel Hawthorne wrote, “Time flies over us, but leaves its shadow behind.” This powerful reminder about how the passage of time shapes our experiences is poignant even at the cellular and molecular level.
There comes a point where the ultimate “shadow” of time causes cells to stop dividing, entering into a state termed senescence. This process may result from the normal course of development or can be the result of age-related insults, such as telomere erosion, DNA damage, mitochondrial dysfunction, and beyond (1).
Paul Robson, PhD, a professor at The Jackson Laboratory and member of the Cellular Senescence Network (SenNet) Program, is at the forefront of understanding developmental cellular senescence. Using advanced tools like single cell RNA-sequencing and spatial transcriptomics from 10x Genomics, Robson's research aims to define age-related changes in both human and mouse tissues. Keep reading to discover how Dr. Robson leverages these cutting-edge technologies to illuminate the molecular underpinnings of senescence and advance our understanding of cellular aging.
Tell us about the primary area of research that you're working on in the lab.
There are two major projects. One is a tissue mapping effort for the SenNet project. We're mapping out adult tissues in both mice and humans to understand and describe senescent cells across the lifespan (2). A component of that is development, and developmental biology is my main area of focus. Specifically, we're looking at developmental senescence in human and mice placenta. There are specific differentiation processes that occur when trophoblast stem cells differentiate into trophoblast populations that model developmental senescence. If we can capture that molecularly, it gives us a tool to get at the mechanisms and temporal dynamics of senescent cell formation.
Related to that, we extensively use differentiation strategies for human induced pluripotent stem cells (hiPSCs) to model this process in a dish as well. We knock out genes and phenotype across developmental time points to not only capture senescence and developmental senescence in the placenta, but, in other related projects, to extensively and comprehensively phenotype null alleles of various hiPS-derived cells.
How do you source your iPSCs? Are these obtained from commercial sources or are you collecting them from clinical research subjects for a specific reason?
For the project where we use hiPSCs most extensively, the cells originate primarily from a single course, a healthy white European male. The long-term goal of this project, which is referred to as MorPhiC (3), is to knock out and phenotype all human protein-coding genes. It is in the proof-of-concept phase as NHGRI-funded work, where the consortium as a whole is doing 1,000 genes and our center, specifically, is focusing on 250 genes.
But we're also trying to introduce genetic diversity, representative of the human population within our hiPSC sources, so we're actually deriving new iPS cells from the local community. The neighboring clinical lab draws blood [from research volunteers], and then we derive the iPS cells. We're working to establish a bank of 96 genetically diverse lines to model variation in iPSC differentiation.
Similarly, in the SenNet project, we're sourcing from 20 different individuals, so that we have genetic variability represented there. In our mouse work, we're also specifically using genetically diverse mice to model phenotypic variation.
What is the significance of ensuring genetic diversity for these projects?
The focus is not necessarily to investigate race-based or ethnicity-based factors. We're collecting samples from research donors from all major ethnic groups—Africans, Indigenous Americans, Asians, Europeans, etc.—to truly grasp phenotypic diversity. We’ll be sequencing and identifying known single nucleotide polymorphisms of functional variation, then doing the experiments to show that there is phenotypic variation as a consequence of the underlying genetic variation.
We're not doing genome-wide association studies in a dish because I don't think we have enough sampling. However, we do want to make it so that when we create null alleles we can assess if the phenotype we get across genetic diversity is robust or not.
How long have you been using 10x products in your research?
We've been using 10x products from the beginning. Prior to starting with 10x assays, we were doing single cell experiments manually. Our first publication was in 2010.
At first, we started using Fluidigm products, and I ran a Fluidigm single cell omics center that started in 2012. We got our first Chromium assay in March 2016. So we've been a long-time user of the Chromium platform since then. The consistency and quality of the product has kept us an avid user of 10x products for the last decade.
How have 10x products impacted your research and the questions you can ask?
Oh, tremendously. Many of the 10x technologies are major pillars in our grants, including the senescence grants. We were actually proposing Visium HD [experiments] in the SenNet grant application before the product’s release. I think we managed to get that grant because of the technologies, like Visium HD, that we were proposing to apply. Similarly, there's multiple publications we have using the Chromium system because it clarifies the underlying biology of any system.
Where do you see the future of this field in the next 2 to 5 years?
Many of the experiments can be quite expensive. Anything that reduces the cost and enables more experimentation, makes it more efficient, we look forward to since that’s what we are always trying to achieve.
For example, single cell RNA-sequencing is becoming a routine assay for us. The new [Universal] on-chip multiplexing combined with the increased number of unique molecular identifiers detected per cell with the GEM-X technology, we think is great. It's just getting better and more cost effective, letting us do more experiments for similar costs. So I think anything that improves cost efficiency, but also increases the richness of the data you get, will be important. On the single cell side, that could be barcoding more cells.
On the spatial side, it's always driven by the quality of the data. That’s why Xenium [profiling] is even more attractive than Visium HD [analysis], which is not as sensitive but has its utility for discovery. It's going to be a little bit of a trade off there: How much information do you need? How many Xenium probes do you need to answer the questions you want to ask? I think that's still an open question. I don't think we need a Xenium platform that profiles all genes. Maybe in a tissue-specific context that might be 1,000 genes, but anything that allows us to continue to get spatial orientation, which is super important in understanding underlying biology, and that leads to more cost-effective analysis will be what will be incredibly useful going forward (4).
Do you see any role for spatial multiomics?
Yes, definitely. We currently run Xenium assays followed by Codex on the very same tissue section. So any advances that integrate it [multiomics] into the imaging of the Xenium run will presumably improve the signal, and we look forward to growth in that area. In particular, proteins and RNA. We are less enthusiastic about spatial epigenetics because of the inefficiency or the lack of sensitivity in that signal, but anything that measures proteins, RNA, and potentially metabolites and lipids will be extremely valuable for future research efforts.
This interview was edited for length and clarity. We’d like to thank Dr. Robson for his insights. You can read about his lab’s latest research on his website.
To discover more about the technologies Dr. Robson discussed, check out the Chromium, Visium, and Xenium platform pages.
References:
- McHugh D, Gil J. Senescence and aging: Causes, consequences, and therapeutic avenues. J Cell Biol 217: 65–77 (2018). doi: 10.1083/jcb.201708092
- Consortium, S. et al. NIH SenNet Consortium to map senescent cells throughout the human lifespan to understand physiological health. Nat Aging 2: 1090–1100 (2022). doi: 10.1038/s43587-022-00326-5
- Adli, M. et al. MorPhiC Consortium: towards functional characterization of all human genes. Nature 638: 351–359 (2025). doi: 10.1038/s41586-024-08243-w
- Li, S. et al. Advancing biological understanding of cellular senescence with computational multiomics. Nat Genet 57: 2381–2394 (2025) doi: 10.1038/s41588-025-02314-y