We used #MALDI imaging mass spectrometry to measure the single-cell lipidomes of hundreds of individual human dermal #fibroblasts. We identify specific lipid metabolic pathways that display cell-to-cell variability. Unexpectedly, single cells clustered by lipid composition!
It took a lot of experiments to convince us, but now we can say it confidently: there is such a thing as a #lipotype!! First of all, toxin-based lipid staining validated the existence of the different populations of dermal fibroblasts both in vitro and in vivo.
Second, by coupling #scRNAseq with #scLipid analysis (yes, a new flavor of #multiomics !), we found that there is a correspondence between discrete lipid metabolic conformations and characteristic fibroblast transcriptional states.
However, #lipotypes are not just the “metabolic shadow” of transcriptional heterogeneity. Enzymes that synthesize lipids are expressed homogeneously across cells and do not correlate with cell-state markers. The correspondence must come from post-transcriptional regulation.
If one wipes out the class of lipids that make up lipotypes, fibroblasts change cell state! This happens through an imbalance of the FGF-TGFb pathways. Manipulating specifically single metabolic reactions, we saw again the same result: fibroblasts change their state.
More experiments and analyses confirmed that FGF/TGF signaling contributes to determining cell lipotypes. Interesting to see how signaling and metabolism can be wired-up in self-contained circuits that control cell identity.
We have been thrilled to observe a biologically-relevant function for cell-to-cell metabolic heterogeneity and are excited to see how #multiomics single-cell analysis can drive us towards uncharted domains of biology.
If proof was needed to the few still #singlecellSkeptics out there, here you have it, and I can say it out loud: #singlecell analyses can be used to learn mechanistic aspects of biology!
Of course, @GiO__DAngelo is the right Twitter handle 😅
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Excited to release the first preprint from my lab: a new framework to convert existing NGS tube-reactions into spatial techniques using compressed sensing principles and a probabilistic algorithm that we call #Tomographerbiorxiv.org/cgi/content/sh…
In the lab we are thrilled about possibilities offered by #spatialomics methods, and a bit impatient... it will take a bit until methods for the different omics are available. That's when we had the idea of using already available low-input protocols to obtain spatial information
The core idea is sampling a tissue by cutting thin slices at several orientations so as to obtain a high throughput readout from each of the strips. The measurements are then used to reconstruct the original localization of the signal by a maximum a posteriori estimation