Vitamin B12’s Impossible Journey

No other vitamin needs a five-stage relay, two carrier proteins, a working stomach, a working pancreas, and one specific 60 cm of gut just to get in. Vitamin B12 is the most fragile absorption pathway in human nutrition — and that is exactly why deficiency is common, and why it is so often missed. Press play and follow one molecule from a mouthful of food to the two enzymes deep inside your cells that cannot work without it. Then break the chain at any link and watch the whole thing collapse.

Try this: switch to Pernicious anaemia and watch the liver stores drain for three years while the MMA line climbs to meet them — then hit High-dose oral (1000 µg) and watch a 1 % passive-diffusion trickle refill them without a single injection.

Diagram is illustrative — not to scale.
MOUTH STOMACH DUODENUM TERMINAL ILEUM — 60 cm Salivary gland releases haptocorrin (R‑protein) into the saliva B12 arrives still stuck to a food protein Parietal cell makes the acid AND the intrinsic factor — the same cell, both jobs HCl IF pH ~2 · pepsin Pancreas trypsin + bicarbonate → pH rises Enterocyte — the intrinsic factor is destroyed in the lysosome; B12 is reloaded onto transcobalamin II and pushed into the blood unabsorbed → stool CUBAM receptor = cubilin + amnionless. Binds IF–B12 and swallows it whole. It exists nowhere else in the body. Calcium-dependent. PPI — acid pump blocked autoantibodies destroy the parietal cells → no intrinsic factor, and no acid either receptor lost — Crohn’s, resection or bypass metformin — Ca²⁺ step blocked no B12 in the food BLOOD most of your serum B12 rides inert haptocorrin (purple) — only the transcobalamin II fraction (blue) is deliverable CD320 — the TC‑II receptor Target cell nerve, marrow, any cell methyl‑B12 adenosyl‑B12 folate (5‑methyl‑THF) TRAPPED Methionine synthase methyl‑B12 + folate homocysteine methionine → SAM cytosol Methylmalonyl‑CoA mutase mitochondrion — adenosyl‑B12 methylmalonyl‑CoA succinyl‑CoA Myelin sheath B12 → SAM → methylation keeps this intact When methionine synthase stalls, the sheath degrades — this is subacute combined degeneration, and it can begin before any anaemia. 100%

Live blood work

Serum B12
0 pg/mL
deficient <200 · grey zone 200–400 (dashed)
MMA — the specific marker
0.00 µmol/L
normal <0.40 · rises in B12 deficiency and not in folate deficiency
Homocysteine
0.0 µmol/L
normal <15 · sensitive, but folate and B6 raise it too
Of this dose, absorbed
0%
the intrinsic-factor route caps out at ~1.5–2 µg per meal
Liver & body stores
0.00 mg · normal 2–5 mg 5 mg 2 mg 0 3.0 MMA 0 0 y 5 y deficiency floor
Nerve myelin integrity
100%
Simulated time elapsed: 0 mo

What’s happening

Press play. A B12 molecule is about to attempt the hardest absorption route in human nutrition.
B12 (cobalt core) food protein haptocorrin intrinsic factor transcobalamin II stomach acid trypsin / MMA homocysteine

Real numbers: the reference ranges, the ~50 % absorption of a small dietary dose, the ~1.5–2 µg per-meal ceiling on the intrinsic-factor route, the ~1 % passive diffusion of a large oral dose, the 2–5 mg body store and the 2.4 µg/day adult RDA are all real, published figures. Illustrative model: the store-depletion, MMA, homocysteine and myelin curves come from a simple compartment model tuned to reproduce the established timelines (stores drain in ~3–5 years when intrinsic factor or the ileum is lost, far more slowly on low intake with the recycling loop intact). They are not measurements from any individual. Timescale: the molecule’s journey takes hours in reality; the clock runs in simulated years — both are shown together so you can see cause and effect. Passive diffusion is drawn far more often than 1 in 100 so you can see it happen at all.


The Science in Plain Language

Only bacteria can make it — not plants, not animals, not you

Vitamin B12 is the only vitamin that contains a metal atom, and the metal is cobalt. Building the molecule around that cobalt takes roughly thirty enzymes, and the entire toolkit exists only in bacteria and archaea. No plant makes B12. No animal makes B12. A cow does not manufacture the B12 in its liver — the bacteria in its rumen do, and the cow concentrates it. The B12 in beef liver, clams, sardines, salmon, eggs and milk is, without exception, bacterial B12 that an animal collected on your behalf.

This is why the vegan question is not a matter of opinion. Spirulina, chlorella, tempeh, barley grass and most seaweeds contain corrinoid analogues — molecules shaped enough like B12 to bind the carrier proteins and to register on some older lab assays, but useless to your enzymes. Worse, they can occupy the carriers that real B12 needs. A B12 supplement or fortified food is not a concession by a plant-based diet; it is simply where the bacteria are, in a bottle instead of a cow.

Five gates, and any one of them can close

Watch the animation and count the failure points. One: stomach acid and pepsin have to prise B12 off the food protein it arrived bound to. Two: haptocorrin, released in saliva, has to grab the freed B12 — which it does happily, because haptocorrin binds B12 best in acid. Three: pancreatic trypsin has to digest that haptocorrin in the duodenum, where the pH rises and haptocorrin loses its grip, so that B12 can be handed to intrinsic factor. Four: the IF–B12 complex has to reach the terminal ileum and be recognised by the CUBAM receptor — cubilin plus amnionless — in a calcium-dependent step. That receptor is found in roughly the last 60 cm of small intestine and nowhere else in the body. Five: inside the absorbing cell, the intrinsic factor is destroyed and B12 is reloaded onto transcobalamin II, the only carrier your tissues can actually take up.

Two numbers matter here. About 50 % of a small dietary dose is absorbed by this route — which is respectable. But the route saturates at roughly 1.5–2 µg per meal. You cannot bank B12 by eating a huge serving of liver once a week: the receptors take what they can hold and the rest goes past. Spreading intake across meals absorbs more than the same amount in one sitting. The adult RDA is 2.4 µg/day; a typical omnivore eats 4–7 µg/day.

Why the serum B12 test is weak, and what to order instead

Here is the most useful thing on this page. A serum B12 test measures total B12 in your blood. But 70–90 % of that B12 is riding haptocorrin, and most cells cannot take haptocorrin up at all. Only the 10–30 % bound to transcobalamin II — called holotranscobalamin, or “active B12” — is deliverable. Your serum B12 can read a comfortable 350 pg/mL while the fraction that can actually enter a nerve cell is on the floor.

That is why the number alone lies. Below 200 pg/mL (about 148 pmol/L) is deficient; 200–400 pg/mL is a grey zone where the test simply cannot tell you the answer — and a very large number of symptomatic people live in that grey zone. The way out is to stop measuring the vitamin and start measuring whether the enzymes are working:

If your serum B12 is in the grey zone and you have symptoms, an MMA is the test that settles it. In the animation, watch for the moment the alert fires: serum reading “normal” while MMA is high. That combination is the single commonest way B12 deficiency gets missed.

Inside the cell: the two jobs, and the folate trap

B12 does exactly two things in a human cell. Job one is in the cytosol: methionine synthase takes a methyl group from 5-methyl-THF (the circulating form of folate) and hands it to homocysteine, producing methionine. Methionine becomes SAM, the body’s universal methyl donor — used to methylate DNA, neurotransmitters and, critically, the proteins of the myelin sheath. Job two is in the mitochondria: a different form of B12, adenosylcobalamin, runs methylmalonyl-CoA mutase, converting methylmalonyl-CoA (from odd-chain fatty acids and the amino acids valine, isoleucine, methionine and threonine) into succinyl-CoA so it can enter the Krebs cycle. Starve that enzyme and methylmalonyl-CoA backs up, is hydrolysed to methylmalonic acid, and leaks into blood and urine — which is where your MMA test comes from, and why nothing else in normal metabolism raises it the same way.

Now the trap. Methionine synthase is the only enzyme in the body that can regenerate plain THF from 5-methyl-THF. Take away B12 and it stalls, and all your folate piles up as 5-methyl-THF with nowhere to go. The cell becomes functionally folate-deficient even though it is full of folate. That is the folate trap, and it explains why B12 deficiency produces exactly the same big, immature red cells — megaloblastic anaemia — as folate deficiency.

And here is the dangerous part. If you give folic acid alone to someone whose real problem is B12, you flood the cell with enough folate to force DNA synthesis through anyway. The anaemia improves. The blood count normalises. The patient looks treated. But you have done nothing whatsoever for job two in the mitochondria, and nothing for the methylation that maintains myelin — so the neurological damage marches on, now with the anaemia removed as a warning sign. This is not a theoretical concern. It is the reason B12 must be excluded before treating anyone with folate, and it is why folic-acid fortification of flour made the question sharper rather than softer.

The neurological picture: it can arrive with a perfectly normal blood count

B12 deficiency demyelinates the dorsal columns and the lateral corticospinal tracts of the spinal cord — hence the old name, subacute combined degeneration. The pattern is characteristic: symmetrical tingling, burning or numbness in both feet and both hands; loss of vibration sense and of the ability to tell, eyes closed, which way a toe is pointing; a gait that is fine in daylight and unsteady in the dark or in the shower with eyes shut; later, stiffness and weakness. Cognitive slowing, low mood, irritability and, occasionally, frank confusion belong on the same list.

The point that matters clinically: none of this requires anaemia. A substantial minority of people with neurological B12 deficiency have a completely normal haemoglobin and a normal MCV. If you are waiting for big red cells before you take the numbness seriously, you will be too late. And time matters — damage treated within weeks to a few months usually reverses; damage that has been running for a year or more often only partly reverses. The neurological picture, not the blood count, sets the urgency.

Who is actually at risk

Why stores last years — and why that is a trap

You carry 2–5 mg of B12, roughly half of it in the liver. Against a daily requirement measured in micrograms, that is three to five years’ supply — and considerably longer if the only problem is low intake, because the enterohepatic loop keeps reclaiming most of the B12 your bile secretes. This is why B12 deficiency does not announce itself. It creeps. Someone can start a PPI, or a vegan diet, or have twenty centimetres of ileum removed, and feel entirely fine for years while the tank quietly empties.

It also explains a bitter asymmetry the model in this page reproduces. Lose intrinsic factor or the ileal receptor and you lose the recycling loop as well as the food — stores drain in about three to five years. Lose only the dietary supply, with the loop intact, and the same stores can last a decade or more. Same empty tank at the end; very different clocks.

Oral versus injection, honestly — and what good treatment looks like

Here is the fact the “High-dose oral” button exists to show you: about 1 % of a large oral dose of B12 crosses the gut wall by passive diffusion — no acid, no haptocorrin, no intrinsic factor, no CUBAM receptor. It simply goes through, all along the small intestine, on a concentration gradient. One per cent of 1000 µg is 10 µg a day, which is four times the adult RDA.

That single mechanism is why high-dose oral B12 can treat even pernicious anaemia — a disease that, for most of the twentieth century, meant injections for life. A Cochrane systematic review comparing oral with intramuscular B12 found that high-dose oral treatment produced comparable serum B12 levels and comparable haematological recovery in most patients. For a great many people, a daily 1000 µg tablet is as good as a needle, at a fraction of the cost and inconvenience.

Injections still earn their place: when there is severe neurological disease (you want the tank refilled fast and with certainty, because the clock on nerve recovery is running); when malabsorption is severe or the bowel is very short (passive diffusion needs gut surface to happen across); and when daily adherence is genuinely doubtful — which is not a small point, because oral B12 only works if it is taken every day. You cannot skip a fortnight of tablets the way you can stretch an injection. Choose the route you will actually follow.

Either way, repletion is fast and cheap. Reticulocytes — new red cells — surge within about a week, which is itself confirmation the diagnosis was right; the blood count typically normalises within about eight weeks; MMA and homocysteine fall within days. Nerve recovery is the slow one: months, and incomplete if the damage has been running too long. Two things to watch as the marrow restarts — repleting a severe megaloblastic anaemia can unmask iron deficiency (the marrow suddenly builds red cells in bulk and consumes the available iron) and, in severe cases, can drop potassium sharply as it is pulled into all those new cells. Neither is a reason not to treat; both are reasons to check the iron studies and the potassium while you do. And finally: find out why. A low B12 is a symptom, not a diagnosis — think about the parietal cells, the ileum, the metformin, the PPI and the nitrous, because the deficiency will come straight back if the broken link in the relay is never identified.

Nitrous oxide oxidises the cobalt atom

Nitrous oxide — N₂O, whether from an anaesthetic machine or a whipped-cream charger — irreversibly oxidises the cobalt atom at the heart of methylcobalamin, flipping it out of the reactive state that methionine synthase depends on. The enzyme is inactivated. The vitamin is still in your blood; it just no longer works.

The consequences are exactly what you would predict from the animation. A long anaesthetic in someone whose B12 was already marginal can precipitate acute subacute combined degeneration. Heavy recreational nitrous use — canisters, balloons — has become a recognised cause of B12-related myeloneuropathy in young, otherwise healthy people, and it can develop over weeks. The diagnostic trap is that the serum B12 is often normal, because the vitamin is present and merely chemically disabled: it is MMA and homocysteine that reveal it. If someone using nitrous heavily develops numb, tingling feet and an unsteady walk, a normal B12 level does not clear them — and this is one of the few situations where you treat immediately rather than wait.

↑ Back to the animation

Connections