Why an In Vitro Concentration Is Not a Dose
Start with the figure printed in the methods section, because it is shakier than it looks. A study says ten micromolar. That is the nominal value, the one calculated when the stock was made up, and several things have already happened to it. Some of the compound has stuck to the plastic, in the manner described in adsorptive loss to surfaces, so the free concentration is under the target at the start and further under it by the end of the incubation. Some may be aggregated, per peptide aggregation in solution. Some may have degraded while the solution was being prepared. And if the stock was made by weighing a powder, the mass on the balance is not the amount of peptide, as net peptide content explains.
Ten micromolar on paper can therefore be six or seven micromolar of intact, free material in the well. Which is already a substantial error to found anything on — and yet the common next move is to take that same figure, multiply by an assumed volume, divide by a molecular weight, and produce an answer in milligrams. The arithmetic is unobjectionable. The output means nothing, because what went into it was never a property of an organism at all.
Micromolar concentrations where the mechanism runs at nanomolar
In vitro peptide studies commonly operate in the micromolar range. Suppose the receptor-mediated effect under investigation has a potency in the nanomolar range — affinity and functional potency being distinguished in Ki, IC50 and EC50. The exposure then sits orders of magnitude above the concentration at which the specific mechanism actually works.
That far above the potency, whatever is observed may be arising through routes that have nothing to do with the mechanism nominally under study. None of which condemns the experiments: high concentrations are frequently chosen on purpose, to lift a signal into measurable range. It does mean the concentration in the title is often a methodological decision rather than a biologically meaningful quantity, and converting it into anything else treats a decision as a finding.
An experimental setting, like the incubator temperature
Set out plainly, an in vitro concentration is this: the amount of compound in the medium bathing cells in a dish, held roughly constant across the exposure period, with nothing standing between compound and cell.
Each clause matters. The value is imposed and then maintained; it is not something the system arrived at. No tissue distribution takes place. No circulating protein binds anything. Nothing is metabolized, nothing is cleared, no barrier has to be crossed. It belongs in the same category as the temperature the incubator was set to — a condition the experimenter selected, rather than an output the biology generated.
Four reasons the arithmetic cannot travel
| Factor | In the dish | Everywhere else |
| Exposure profile | Flat and sustained for hours | Rises, spreads, falls — often fast |
| Protein binding | Little protein, so nearly all free | A large share bound and unavailable |
| Distribution | One compartment, evenly mixed | Some compartments well reached, some barely, some not at all |
| Stability | Few peptidases present | Peptidase activity the dish does not model |
Take those rows one at a time. A curve and a flat line that share a peak value are not the same exposure, and how quickly the fall happens is the half-life question examined in half-life and analog modifications. Identical nominal concentrations imply different free concentrations once binding differs. Uneven distribution is what makes the “assumed volume” step the weakest joint in the whole conversion. And something that survives six hours in medium may not survive six minutes in a setting rich in peptidases, for reasons covered in how peptides degrade.
What bridging the gap actually requires
Where the question genuinely has to be answered, answering it is a discipline in itself and looks nothing whatsoever like unit conversion. It involves measuring free concentrations rather than nominal ones, characterizing binding, determining clearance and distribution empirically in a living system, constructing a model that links exposure over time to observed effect, and then testing that model against data it was never fitted to.
Every stage of that requires measurements which, for the great majority of research peptides, simply do not exist. Which is the substantive point, and worth stating without hedging: the bridge here is not difficult to construct, it is impossible to construct, because the materials it would be made of were never gathered.
What the experiment can legitimately carry
Mechanistic claims are what in vitro work supports. That a compound engages a target. That a measurable response follows in a given cell type. That the size of the response tracks concentration in some particular way. Those are real conclusions and they are the interesting ones.
Quantitative claims about an organism are not among them. Read “an effect at ten micromolar in cells” honestly and it says an effect occurred at ten micromolar, in those cells, under those conditions — a statement containing no quantity that applies anywhere outside that plate. It is also why the distance between a cell model and an animal model, taken up in the limits of the published literature, carries the weight it does.
Why the research-use restriction is not a formality
Everything supplied here is laboratory research material, and this is among the most concrete reasons that restriction means something. The published record on these compounds consists overwhelmingly of in vitro experiments and animal studies. Neither body of work contains a quantity that survives being moved out of the system where it was measured.
Anyone who runs the calculation anyway will get a number, since arithmetic always obliges. The arithmetic was never the flaw. The flaw is the premise that the figure it started from describes an organism.
