- Author
- Laura C. Hinte, Daniel Castellano-Castillo and colleagues, ending with Ferdinand von Meyenn, ETH Zurich
- Published
- 18 November 2024
- Kind
- Peer-reviewed research article in Nature, with human biopsies and mouse experiments
- Original
- www.nature.com/articles/s41586-024-08165-7
- Subject
- Science and medicine
- Themes
- What lasts
What it is
A team from Switzerland, Sweden and Germany studied adipose tissue before and after weight loss. They used single-nucleus RNA sequencing in humans and mice. In mice they also profiled histone marks and chromatin accessibility in fat cells.
What it says
- Human fat biopsies taken two years after bariatric surgery kept many obesity-linked expression changes.
- Persisting human adipocyte changes included lower metabolic pathways and higher fibrosis and apoptosis pathways.
- Obese mice returned to normal weight in 4 to 8 weeks. Adipocyte epigenomes did not fully normalize.
- Over 1,000 promoters showed altered H3K4me3 marks. 848 enhancers stayed altered after weight loss.
- Epigenetic marks explained 57 to 62% of lasting decreases and 68 to 75% of increases.
- Formerly obese mice gained weight faster than controls when fed a high-fat diet.
- Epigenetic signatures explained three to six times more refed-mouse gene changes than earlier transcript status.
What evidence it gives
The human data are biopsies from three bariatric surgery studies. Groups had five to eight people. The mouse work adds paired sequencing of translation, chromatin and four histone marks. It also tests glucose and palmitate uptake in cultured fat cells, and a diet challenge.
Interests and limits
The authors say the results do not prove cause, and epigenetics was mouse-only. The mice were male, most metabolic measures normalized, and experiments were not randomized or blinded. Two authors report consultant fees from drug firms, including Novo Nordisk. The main text and discussion were read. Most Methods detail, the Extended Data legends and the references were not.
Connections
These links are this wiki's own reading, not claims the sources make.
- Agents Don't Need Memory. They Need Documentation.. Liao says stale snippets survive because retrieval ranks by similarity, not currency. Hinte and colleagues find fat cells keeping obesity-linked marks after weight loss. Memory can outlast the reason for it. Neither system seems able to mark an entry as outdated.