Engineered enzyme reverses age-related protein damage
An engineered enzyme reverses agerelated protein damage by stripping a stubborn advanced glycation end product (AGE) from proteins once thought nearly impossible to fix. Researchers at Calico, Revel Pharmaceuticals, and the University of Colorado report that CMLase cleared CML from aged human tissues in Nature Communications lab tests.
Key Takeaways
- Teams at Calico, Revel Pharmaceuticals, and the University of Colorado engineered CMLase to remove the AGE Nε-carboxymethyl-lysine (CML) from proteins.
- Five rounds of directed evolution yielded an enzyme that restored lysine on model proteins without degrading them.
- In thin sections, CMLase cut CML by about 70% in aged human artery and 55% in aged skin.
- Delivery into living tissue, functional rescue, and immune safety remain untested—this is a lab proof of concept.
What did the engineered enzyme reverse?
Long-lived proteins, especially in the extracellular matrix, can persist for years and accumulate nonenzymatic sugar damage. Some of that damage becomes chemically stable AGEs, which can stiffen tissues, warp protein shape, or trigger inflammatory signaling.
The study targeted CML, an AGE on lysine that can impair proteins and engage the pro-inflammatory RAGE receptor. The body can slow AGE buildup but generally cannot reverse it. Lifespan Research Institute covered the Nature Communications paper describing the breakthrough.
How did scientists build CMLase?
Researchers began with bacterial glycine oxidases that acted weakly on free CML, then screened more than 44,000 structural candidates—including AlphaFold models—for enzymes that could reach peptide-bound CML. A weak hit became the starting point for directed evolution.
They tied enzyme activity to survival in lysine-starved E. coli fed CML and ran five rounds of mutation and selection. The final CMLase variant opened its active site, bound CML more tightly, and worked across more peptide contexts. It removed CML from intact proteins such as modified bovine serum albumin without degrading them.
Did it work on real aged human tissue?
In ex vivo tests, CMLase reduced endogenous CML in soluble lens proteins from a 64-year-old donor. It also sharply lowered CML in aged human artery (by 70%) and skin (by 55%), with post-treatment skin staining below levels seen in 31-year-old skin.
Those experiments used very thin sections and did not show penetration into a living artery or intact skin. The team also did not run functional rescue experiments or immunogenicity tests—key limits for a bacterial-origin enzyme.
Why does this matter for longevity research?
For readers tracking longevity and biohacking science, the win is conceptual: AGE removal ranked among geroscience’s hardest problems, and CML is now experimentally reversible in the lab. Harder targets such as glucosepane collagen crosslinks still await similar tools.
The takeaway is proof of principle, not a clinic-ready therapy. Delivery, safety, and whether repaired proteins restore tissue function are the next questions the field must answer.