Anserine vs beta-alanine:
the carnosinase argument

In 2006, a paper by Roger Harris and Mark Chichester established the foundation that the entire beta-alanine industry now rests on.³ They demonstrated that oral beta-alanine, taken daily over four to ten weeks, raises intramuscular carnosine concentration by between 40 and 80 percent. Beta-alanine is the rate-limiting precursor in the synthesis pathway: the muscle cell has plenty of L-histidine but very little free beta-alanine, so adding beta-alanine to the bloodstream pulls the synthesis equilibrium upward. The carnosine that gets made stays in the cell, because muscle carnosine transporters are inward-facing. There is no efflux problem to worry about.

The follow-on literature is one of the most replicated effect sets in modern sports nutrition. Meta-analyses converge on a 2 to 3 percent improvement in efforts that last between one and four minutes, in subjects who have completed at least four weeks of daily loading at doses between 3.2 and 6.4 grams.⁴ The most-studied form, CarnoSyn, holds an NSF Certified for Sport mark, sits in dozens of off-the-shelf pre-workouts, and is — by any honest reading of the evidence — a settled tool. There is no serious debate that chronic beta-alanine supplementation enlarges the carnosine pool and produces small but real ergogenic effects in the right exercise window.

The tingling sensation, paresthesia, that most users report inside thirty minutes of a dose comes from non-specific binding at MrgprD receptors on cutaneous neurons. It is not pharmacologically meaningful. It does, however, tell you the substance arrived.⁵

Beta-alanine is the gym athlete's tool, the cyclist's tool, the cross-fitter's tool. It is the lifter doing eight-by-three at threshold who feels the last set arriving cleaner in week six. The category has been won. It is a real category.

Beta-alanine has two ceilings that no amount of dosing reform can fix.

The first ceiling is acute timing. The mechanism is chronic by construction. The muscle cell has to synthesize the carnosine, internally, over weeks. There is no race-morning dose that produces a useful effect that afternoon. The athlete who takes their first scoop on Friday before a Saturday marathon is loading a pool that will be ready for some marathon two months from now. This is not a marketing quibble. It is the underlying biochemistry of carnosine synthase, which operates on a slow timescale relative to a race window.

The second ceiling is that beta-alanine cannot raise anserine. Anserine is methylated carnosine — specifically, the methylation occurs on the imidazole ring's tau nitrogen. Carnosine N-methyltransferase, the enzyme that performs this methylation, is expressed in low and variable amounts in human skeletal muscle.⁶ Humans do not, on average, methylate carnosine to anserine in pharmacologically meaningful quantities. Fish do — to extraordinary concentrations. Chickens do. Humans largely do not. The intramuscular dipeptide pool that beta-alanine builds is overwhelmingly carnosine, not the more pharmacokinetically interesting methylated cousin.

For decades the obvious alternative — take carnosine directly, by mouth, and skip the precursor route — failed. The failure was not absorption. Oral carnosine is absorbed efficiently. The failure was downstream: oral carnosine enters circulation and disappears, within minutes, into its constituent amino acids. Beta-alanine and L-histidine. The bloodstream effectively converts the supplement back into one of the precursors it was supposed to bypass.

The disappearing act is performed by an enzyme called serum carnosinase, encoded by the CNDP1 gene, expressed primarily in the liver and circulating in plasma. Plasma half-life of oral carnosine in healthy human adults is on the order of fifteen to thirty minutes. Most mammals — mice, rats, dogs, cats, fish, pigs — do not express CNDP1 in any meaningful concentration and keep their dietary carnosine largely intact in circulation. Humans and a few primates are an evolutionary exception.⁷ The functional reason for this asymmetry is not fully settled. The practical consequence is that, for most humans, taking oral carnosine and expecting it to reach muscle as carnosine is a slow and inefficient way of making beta-alanine.

This is the carnosinase problem. It is the reason beta-alanine won the Western market. It is also the reason anserine — the methylated form — is the molecule worth a second conversation.

Anserine differs from carnosine by a single methyl group on the imidazole ring. That methyl group does almost nothing to the buffering chemistry. The pK_a of the imidazole nitrogen — the actual hydrogen-ion-accepting feature of the molecule — shifts only modestly. The buffering work is essentially the same.

The methyl group does almost everything to the pharmacokinetics. Serum carnosinase cleaves carnosine readily. It does not cleave anserine at meaningful velocity in vivo. The bond geometry around the methylated nitrogen sits outside the enzyme's active-site preference. Plasma half-life of orally administered anserine is materially longer than that of oral carnosine. The molecule has a chance of reaching the working muscle intact and entering the buffer pool from outside.

The methyl group does almost nothing to the chemistry, and almost everything to the half-life.

This is the only reason anserine — in a category dominated, justifiably, by beta-alanine — is interesting at all. The single methyl group is the entire commercial premise of the molecule. Without it, oral anserine would be just another way to make beta-alanine.

The cleanest independent work on acute anserine supplementation in humans comes from Wim Derave's laboratory at Ghent University in Belgium. The Ghent group has spent close to two decades on the carnosine-anserine system: first establishing the chronic beta-alanine loading curves, then probing the carnosinase bottleneck, and then turning, in the last few years, to anserine itself.

The 2021 paper in the Journal of Applied Physiology⁸ remains the foundation of every defensible acute claim. Fifteen male volunteers, crossover design (each subject as their own control), 20 mg/kg of carnosine plus 20 mg/kg of anserine taken thirty minutes before a Wingate-style sprint protocol. Outcome: a six-percent improvement in initial five-second sprint power. Plasma sampling tracked the effect to anserine, not carnosine. Carnosine was cleaved as expected; anserine survived. The methyl group did exactly what its structural geometry predicted it would.

The 2022 follow-up in the Journal of the International Society of Sports Nutrition⁹ mapped the dose-response curve across three doses (10, 20, 30 mg/kg) and two timing windows (30 and 60 minutes pre-effort). The best protocol — 30 mg/kg, 60 minutes pre — produced a 3 percent increase in peak power and a 4.5 percent increase in peak torque on repeated-sprint testing. Anserine bioavailability ran approximately 2.5 times higher than carnosine bioavailability across all conditions.

The Ghent group also reported what the supplement-industry summary tends to omit: effect size was inversely correlated with the individual subject's serum carnosinase activity. High-carnosinase subjects were partial responders. Low-carnosinase subjects were the larger responders. There is no over-the-counter assay for serum carnosinase activity. Until there is, the honest read of every anserine effect-size claim is responder-dependent.

Two tools, two clocks, one buffer pool.

Yes. The two molecules operate on non-overlapping clocks, and the buffer pool they enlarge is the same pool. Chronic beta-alanine raises the standing intramuscular carnosine baseline. Acute anserine adds, on the day, a circulating dipeptide pool that the working muscle can draw on through the effort window. The two contributions are additive at the level of total buffering capacity. There is no documented antagonism between them at any dose either is administered at.

The stacked athlete looks like this: a marathoner who has been on a chronic beta-alanine loading regimen for the build phase of their training block, who then takes anserine 40 to 60 minutes before the gun on race morning. The standing pool is as enlarged as biology will allow it to be. The circulating dose is timed to the window. Both buffer mechanisms are operating at peak when the lactate threshold gets crossed in mile 22.

What you do not need to do, and should not do, is reach for beta-alanine on race morning. The mechanism does not exist there. The paresthesia is the only thing that arrives.

The Ghent results are small numbers attached to a real mechanism. Three percent, four-and-a-half percent, six percent on the highest-responding protocol. These are the kinds of numbers that should make the honest reader pause and ask: what does three percent mean, in real race terms?

For a 2:40 marathoner, three percent is roughly four minutes. For a sub-3:00 athlete it is closer to five. These are not negligible margins in a sport where qualifying-time gaps are measured in seconds. They are also not the kind of effect that converts an untrained athlete into a trained one. They are tail-end effects: visible at the margin, in the athletes already pushing the buffer system as a limiting factor. The well-fueled, well-trained marathoner has the strongest mechanistic case for adding either molecule. The novice athlete whose limits are elsewhere — aerobic capacity, glycogen storage, gait economy — should fix those first.

Effect sizes are also responder-dependent. The serum-carnosinase activity result from the Ghent group is the most important hedge in this whole conversation. Until there is a commercial assay, every claim about anserine in particular should be read with the silent footnote: in the population subset whose carnosinase activity is in the responsive range.

Both molecules work. Both have published, peer-reviewed mechanism and effect. Neither is magic. The honest pitch on either of them is that they each contribute a small amount of marginal buffering capacity through a different pharmacokinetic pathway, and that the athletes who get the most from them are the ones who have already done the larger, harder, non-supplemental work.