Sulfur & sulfate

"Bone broth is rich in sulfur" is one of the more common claims made for collagen-rich food, and it is usually deployed to imply something about MSM. The MSM version is wrong. What sits underneath it, though, is a more interesting idea than the claim it gets attached to — and one nobody has tested.

Why sulfate matters to cartilage at all

Cartilage resists compression because its proteoglycans are heavily sulfated. Chondroitin sulfate and keratan sulfate carry negative charges that pull water into the matrix; the water is what bears the load. Undersulfate the proteoglycans and the tissue fails mechanically.

Every one of those sulfate groups is transferred by a sulfotransferase using PAPS as the donor, and PAPS is regenerated from inorganic sulfate that comes largely from the breakdown of methionine and cysteine. That the pathway is essential is not in doubt — genetic disorders of sulfate handling produce skeletal dysplasias, and sulfatase deficiency in Morquio syndrome causes keratan sulfate accumulation with a characteristic bone and cartilage phenotype.

The open question is whether the supply ever runs short in an ordinary person.

The study that makes it interesting

One small paper from 2001 supplies three findings that, together, form the hypothesis.

Sulfate is the part that moves. Oral glucosamine sulfate raised serum sulfate from 331 to 375 µmol/L three hours after a 1 g dose — while having no detectable effect on serum glucosamine. Even large oral doses do not raise blood glucosamine. If glucosamine sulfate works, the authors argue, the sulfate may be what is doing it.

The pool is depletable, and by something people take for joint pain. When the same dose was taken with 1 g of acetaminophen — a drug cleared largely by sulfation — serum sulfate fell, from 325 to 290 µmol/L. Sulfate is a consumable resource that drugs compete for.

And it reaches the joint. In 15 patients having a knee effusion aspirated, synovial fluid sulfate tracked serum sulfate almost perfectly: r = 0.99, slope 0.97. Whatever your systemic sulfate status is, that is what your cartilage is working with.

What that does and doesn't establish

It establishes that sulfate availability is modifiable, competable, and transmitted to the joint. Those are three real facts and they make a testable prediction — one the authors state: non-sulfate salts of glucosamine should be ineffective. That prediction has arguably borne out, since the trials using glucosamine hydrochloride are consistently the null ones.

It does not establish that sulfate is rate-limiting for proteoglycan synthesis in anyone eating a normal diet. That is the load-bearing assumption and it is untested. The study was small, not randomized, and its authors say plainly that their results "do not prove that glucosamine sulfate improves osteoarthritis."

No trial of methionine, cysteine, N-acetylcysteine, or inorganic sulfate with a cartilage or joint endpoint exists. For a hypothesis this specific and this cheap to test, that is a striking gap.

What bone stock actually contributes

Sulfur, but not the kind usually claimed. A defined bone stock measured 9.29 mg of methionine and 3.46 mg of cysteine per 100 mL, plus whatever sulfated glycosaminoglycans came out of the bones. Those are genuine sulfur sources feeding the same pool.

They are also unremarkable. Any protein-containing meal supplies sulfur amino acids, and stock is a poor protein source by volume in most preparations. There is nothing special about stock here except that the claim gets made about it.

MSM is a different molecule

The claim that bone stock supplies MSM is the specific thing this entry exists to correct, and it does not survive contact with the literature.

Methylsulfonylmethane is dimethyl sulfone. Its biosynthesis is marine — algae make dimethylsulfoniopropionate, which becomes dimethyl sulfide, which oxidises in the atmosphere to DMSO and then to MSM. It is not a structural component of connective tissue. The canonical survey of its natural occurrence covers fruits, vegetables, grains and beverages; the only animal product it has ever been isolated from is cow's milk.

There is no measurement of MSM in any meat, bone, cartilage or stock. And even granting presence, natural food concentrations run in hundredths of a part per million against trial doses of 1.5–6 grams a day — a gap no quantity of soup closes.

Sulfur-containing is not the same as containing MSM. That substitution is almost certainly where the folk claim originates, and it is worth separating cleanly: the sulfate hypothesis above is interesting and untested, while the MSM-in-stock claim is simply unsupported.

A practical note that follows from the biochemistry

If the sulfate account is right, then the most actionable thing in it is not what to eat but what competes. Acetaminophen is cleared by sulfation and measurably lowered serum sulfate in the study above. Anyone taking it routinely for joint pain is, on this hypothesis, drawing down the pool their cartilage sulfates proteoglycans from.

That is a hypothesis, not advice, and it has never been tested against any clinical outcome. It is flagged here because it is the kind of interaction that would be easy to miss and cheap to study.