The Cartilage Guide
PreclinicalFoods & Nutrition · What the data say to limit

Dietary AGEs & cooking method

Preclinical · 7 studies cited · 3 min · Updated 2026-08-15

In short: Advanced glycation end-products crosslink the collagen in cartilage and make it stiffer — measurably, in human tissue, with the effect tracking AGE concentration — and cartilage accumulates them essentially for life because its collagen barely turns over. Separately, dry-heat cooking raises the AGE content of food ten- to a hundredfold over the raw state. Both halves are well measured. No study in any species has put dietary AGE intake against cartilage glycation or any joint outcome.

This entry is two well-measured things and the missing study between them.

What glycation does to cartilage

Sugars react non-enzymatically with proteins and leave crosslinks behind. In human articular cartilage incubated with threose to raise glycation end-product levels, the collagen network stiffened in proportion: instantaneous deformation fell by up to 40% at the highest concentration, correlating tightly with AGE fluorescence (r = −0.81, p < 0.0001), confirmed by an independent osmotic-stress measurement, and attenuated when glycation inhibitors were added. The authors propose this as the molecular reason age is the single largest risk factor for osteoarthritis.

Cartilage is the tissue where it matters most, and the reason is turnover. AGE accumulation is governed by how long a collagen molecule stays in place, and articular cartilage collagen has one of the longest half-lives in the human body. Crosslinks laid down in your twenties are still there in your sixties; there is no meaningful clearance route.

The picture is not one-directional, and the paper that shows it best is the one that measured mechanics rather than deformation. Bovine cartilage brought to the AGE levels of aged human tissue with ribose became 60% stiffer in dynamic modulus and 35% stronger — and reached failure at 25% less strain. Stiffer, stronger, more brittle, all at once. The authors note the strength gain may partly offset degenerative changes, while the loss of extensibility makes the tissue prone to stress concentration and fracture. Which of those dominates in a living joint is not something a materials test can answer.

What cooking does to food

Dry heat is the variable. In a validated database of AGE content across food categories and preparation methods, dry-heat cooking raised AGE content ten- to a hundredfold above the uncooked state. Animal foods high in fat and protein are AGE-rich to begin with and gain the most on cooking; vegetables, fruit, wholegrains and milk stay comparatively low even after cooking. New formation was reduced by moist heat, by shorter cooking times, by lower temperatures, and by acidic ingredients — lemon juice, vinegar — in the pan.

That is a large, controllable difference in exposure, and a portion of ingested AGEs is absorbed: dietary restriction lowers circulating AGEs and markers of oxidative stress and inflammation in human studies. Those studies measured metabolic and renal endpoints. None of them looked at a joint.

The study nobody has done

Put those two paragraphs next to each other and the missing experiment is obvious. No study, in any species, has measured dietary AGE intake against cartilage AGE content. None has measured it against cartilage on imaging. None has measured it against a joint symptom. The cartilage biochemistry and the food chemistry are both good and they have never appeared in the same paper.

There is also a quantitative question underneath. Most of the crosslinking in cartilage is endogenous — a function of blood glucose over decades acting on a protein that does not turn over. How much the dietary route adds to that pool is unmeasured, and it may be small. Until somebody measures it, "grill less" is a plausible inference from two solid literatures rather than a finding.

The one experiment that touches both

Human cartilage explants exposed to advanced glycation end-products bleed glycosaminoglycan and uronic acid out of the matrix, and anthocyanins reduce that loss while suppressing the NF-κB and MAPK signalling the AGEs switch on. That is the only study in this area to bring a food compound and cartilage glycation into the same preparation, and it is a cell and explant experiment.

Practical notes

If the argument appeals, the levers the food-chemistry work actually supports are: moist heat over dry, lower temperatures, shorter cooking, and acid in the marinade. They cost nothing and they overlap with a good deal of ordinary cooking. What they do not have behind them is a joint outcome.

Note the useful mirror with what cooking does to collagen: there, long moist heat is what makes a food work. Here, dry heat is what loads it with glycation products. Both are the same chemistry seen from opposite ends of the pot.

What would change this entry

One animal study. High-AGE versus low-AGE diet, matched for energy and macronutrients, with cartilage pentosidine and histology at the end. By the standards of this field it is cheap, and it would settle whether the dietary route contributes to the cartilage pool at all — which is the entire question this entry is currently unable to answer.

Why this tier? The cartilage biochemistry is solid explant work in human tissue, and the food chemistry is a validated composition database — but they are separate literatures. Nothing measures dietary AGE intake against cartilage AGE content, imaging or symptoms, and the human AGE-restriction trials use metabolic and renal endpoints. The same experiment that shows AGE crosslinking makes cartilage brittle also shows it makes cartilage stronger. Preclinical.

Key studies