What is sulforaphane?

Where it comes from

Cruciferous vegetables store a compound called glucoraphanin, a glucosinolate. They also produce an enzyme called myrosinase. In an undamaged plant the two are kept apart, in separate cellular compartments.

When the plant is cut, chewed, crushed or blended, those compartments break and the enzyme meets the precursor. Myrosinase strips the sugar off glucoraphanin, the remaining fragment loses a sulfate group and rearranges, and the result is sulforaphane.

This is a defense mechanism. The plant is not making a supplement ingredient; it is making something unpleasant for whatever is eating it, at the moment it is being eaten.

Which plants can do it is narrower than the usual lists suggest. Glucoraphanin is the only common glucosinolate that yields sulforaphane, and among the five groups of Brassica oleracea that Kushad and colleagues grew side by side it predominated in broccoli alone. Kale, cabbage, cauliflower and Brussels sprouts are led by sinigrin, which yields a different isothiocyanate. The measurements for each are set out under foods high in sulforaphane.

Why that matters in practice

Because the compound is made on demand rather than stored, how you handle the food changes how much you get. Heat destroys myrosinase. Chopping and waiting before cooking gives the enzyme time to work. Chewing thoroughly does more than swallowing quickly. These are not marginal effects: a crossover trial in eight men measured 37 percent bioavailability from raw broccoli and 3.4 percent from the same amount cooked. That study and two others on sulforaphane absorption and timing are set out with their assays on the dosage page. The broccoli sprouts section applies the same point to the food where it matters most.

It also explains the confusion around supplements. A product containing glucoraphanin has the precursor but may have no working enzyme, in which case conversion depends on gut bacteria, which vary widely between people. Measured in urine, that difference came to about 10 percent of the dose against almost 40 percent when active enzyme was supplied with it. What else a sulforaphane supplement label leaves out is covered in the supplements section.

What does it do once it is absorbed?

The usual answer is that it activates Nrf2, and that answer comes from cells and animals rather than from people. Nrf2 is a transcription factor that switches on a group of antioxidant and detoxification genes, and sulforaphane induces it. The laboratory literature on that mechanism is large and consistent, and nearly every health claim made for the compound routes through it.

In people it has been checked once. Long and colleagues gave 300 mg of a stabilized synthetic sulforaphane daily for 14 days to 65 patients hospitalized with pneumonia, against 68 on placebo, then sequenced messenger RNA from their blood leukocytes and ran a 45-cytokine serum panel. Expression of Nrf2 target genes was not significantly changed. Neither were interleukin-6, interleukin-1 beta or tumor necrosis factor alpha. The sequencing did find 310 genes differentially expressed between the groups at day 8 and 42 at day 15, at a false discovery rate below 0.05, so the treatment was altering gene expression without touching the pathway everybody names.

A single trial in acutely ill patients cannot settle a mechanism, and the dose was far above any food portion. What it does establish is that the mechanism sentence you will read on every other page about this compound has been checked in humans once, and the check came back negative. The sulforaphane benefits page sets that trial out alongside the other ten.

Is it an essential nutrient?

No. There is no deficiency state, no recommended intake and no established requirement. Sulforaphane is a compound found in some foods that has been studied for possible effects, which is a different thing from a nutrient the body needs.

What has it been shown to do in people?

Less than the popular coverage implies, and the gap is worth stating precisely. Eleven trials have given sulforaphane to people, for conditions from prediabetes to Helicobacter pylori gastritis to autism, and eight of the eleven ended in a laboratory measurement rather than a change anybody would notice. The two that followed patients to recovery, length of stay and death are the largest on the list, at 105 and 133 patients, and neither found an effect. In prediabetes, a randomized placebo-controlled trial of 74 adults missed its prespecified primary outcome, finding fasting blood glucose 0.2 mmol/l lower than placebo over 12 weeks against the 0.3 mmol/l it had set as the target. Each trial is set out with its sample size and its outcome type under sulforaphane benefits.