Vitamin K2 MK-7: the 7 is not a dose
it is the length of the molecule's tail

Vitamin K2 is not one molecule,
it is a numbered family

Vitamin K1 is called phylloquinone and comes from green leaves. It has a single tail, saturated, always the same. There is nothing to number there.

Vitamin K2 is something else. It is a collective name for the menaquinones, a whole series of molecules running from MK-4 to MK-13. MK stands for menaquinone, and the number counts the isoprene units in the side chain. MK-4 has four, MK-7 has seven. That is all the number means.

The ring at the end of the molecule, the part that actually does the chemical work, is identical in all of them. Only the tail changes. And the tail is not decoration: it is the fatty part of the molecule, so it determines which carrier in the blood the molecule ends up in and how long it circulates there.

Where each form comes from is another detail that tends to get lost. MK-4 occurs in animal tissues. MK-7 is made by bacteria, and the best known is Bacillus subtilis, the strain that ferments soya beans and turns them into natto, the strong-smelling Japanese food that half of Japan eats for breakfast and the other half refuses to touch.

What vitamin K actually does
is a single reaction, repeated on a handful of proteins

Vitamin K is not a mineral that gets deposited somewhere. It is a cofactor. In other words, it sits next to an enzyme and helps it do one single thing, and that thing is surprisingly small.

The enzyme is called gamma-glutamyl carboxylase. It looks for certain glutamic acid residues in a protein and attaches one more carboxyl group to them. The result has a name of its own, gamma-carboxyglutamic acid, abbreviated Gla. Two carboxyl groups side by side form a kind of clamp that can hold a calcium ion.

Without vitamin K, the protein is made anyway. It just comes out without the clamps. That version has a name too, undercarboxylated protein, and it is exactly what laboratories measure in vitamin K studies. They do not measure bones. They measure how many proteins have been left without clamps.

The list of proteins that depend on this reaction is short and mixed. In the blood: prothrombin, that is factor II, plus factors VII, IX and X, and proteins C, S and Z, which put the brakes on the same system. Outside the blood: osteocalcin, produced by bone cells; matrix Gla protein, found in artery walls and cartilage; plus Gas6 and periostin.

This is also why vitamin K always comes up in conversations about bones: osteocalcin, the protein made by bone cells, is one of its substrates. Not because of some vague connection, but because there is a reaction with a name.

The study that put numbers
on K1 and MK-7

It dates from 2007, appeared in the journal Blood and is still the paper everyone goes back to. Four small studies put together, with between 10 and 18 healthy volunteers each, all aged between 25 and 35. The dose was the same in moles for both forms, 0.22 µmol a day, for six weeks, with a twelve-week break between arms.

The first part is identical for both: a peak in the blood at around 4 hours. After that, they look less and less alike.

K1 drops by 86% over the next four hours and returns to baseline within a few hours, which matches its reported half-life of one to two hours. MK-7 remains measurable for at least four days. The authors calculated the half-life from the slope of the second phase of the decline and arrived at 50% lost in 68 hours.

Total exposure, meaning the whole amount that passes through the blood, came out 2.5 times higher for MK-7 in the first 24 hours and 6 times higher over 96 hours. And the ratio of osteocalcin with clamps to osteocalcin without them improved about three times more on MK-7 than on K1.

What this study is not: a study about broken bones, about arteries or about anything that happens over years. It is pharmacokinetics, in small groups of young, healthy people. It tells you what can be seen in the blood and what happens to one protein. That is all, and it is a lot.

And MK-4?
At 420 µg, it did not show up in the blood at all

This is the part you will hardly find anywhere in Romanian, even though it is the easiest result in the whole field to understand.

In 2012, a group published in Nutrition Journal a direct comparison between MK-4 and MK-7 in healthy Japanese women volunteers. The protocol had two parts.

The first: a single dose of 420 µg of each form, with repeated measurements afterwards. MK-7 was well absorbed, reached its maximum serum level at 6 hours and could still be seen up to 48 hours. MK-4 was not detectable in the serum of any participant, at any measurement time.

The second: 60 µg a day of each form, for seven days in a row. MK-7 significantly raised serum levels in every participant. MK-4 did not raise them at all.

It is worth saying clearly what this does not mean. It does not mean that MK-4 does not exist in the body or that it has no role there. MK-4 is found in tissues, and the body makes it locally from other forms of vitamin K. It only means that a dose of MK-4 swallowed as a supplement does not show up in circulating blood, and circulating blood is the only thing the study measured. In practice, this is why the form used in supplements has become MK-7 almost everywhere.

What was measured in bone,
over three years

The longest study on MK-7 was published in 2013 in Osteoporosis International. 244 healthy postmenopausal women, split between 180 µg of MK-7 a day and placebo, for three years.

They measured bone mineral density by DXA at several sites, mineral content, bone strength indices, plus osteocalcin in the blood.

Vitamin K status improved, measured by the ratio of osteocalcin without clamps to osteocalcin with clamps. The age-related decline in mineral content and density was smaller at the lumbar spine and the femoral neck. At the total hip, it was not.

The honest part is that last sentence. A single study, a single population, three years, and one of the sites measured did not move. The authors themselves wrote that the result needs to be confirmed in other groups. It is worth knowing that before you read “clinically proven” on a box.

Where vitamin K2 really does call for
a question to your doctor

Oral anticoagulants of the vitamin K antagonist type, the ones you may know by the names acenocoumarol or warfarin, work precisely by blocking the recycling of vitamin K in the body. A K2 supplement pushes in exactly the opposite direction.

In the 2007 paper, the authors calculated that an MK-7 intake above 50 µg a day may interfere with treatment. Fifty micrograms is not much, well below what an ordinary K2 capsule contains. This is not a theoretical precaution.

What our label says
on the K2 line

For us, this figure sits on Vitamin D3 + K2 + A. A serving of 2 capsules contains 240 µg of vitamin K2 as all-trans MK-7, obtained by fermentation with Bacillus subtilis, alongside 100 µg of vitamin D3, which is 4,000 IU, and 800 µg retinol equivalent of vitamin A. All three are fat-soluble and come dissolved in oil in a softgel, in an amber bottle with an aluminium cap, because vitamins A and D do not get on with light. One bottle lasts 30 days.

Vitamin K contributes to the maintenance of normal bones and to normal blood clotting.

Why K2 sits in the same capsule as D3 is a discussion we have written about separately, in the article on vitamin D3 and K2 taken together, and we wrote about the figure on the D3 line in how many units of vitamin D a day. From this article, the only thing to remember about our label is the form: MK-7, so the seven-unit tail, so what you take today can still be seen in your blood the day after tomorrow.

A practical tip, which does not depend on us: since all three are fat-soluble, the capsule is taken with your main meal, not on an empty stomach in the morning. That is the most common way to waste a fat-soluble vitamin, and we have put the rest of the timetable together in which supplements should not be taken together.

Frequently asked
questions