Radioligand Binding Assays: Measuring Receptor Affinity (Kd and Ki)

When a paper or a specification sheet says a compound binds its receptor "tightly," there's almost always a number behind that word — and a radioligand binding assay is usually how the number was found. These experiments are the classic way to put a figure on receptor affinity in vitro. The figures they produce, Kd, Ki, Bmax, and IC50, are the shared vocabulary of receptor pharmacology. One note before we start: radioligand binding is a laboratory research method, and the compounds discussed here are for research use only. This article is about how the measurement works, not about use of any kind.

The idea is simple, even when the math looks intimidating. You attach a radioactive label to a molecule that binds your receptor, let it find that receptor in a test tube, wash away what didn't stick, and count what remains. Done cleverly across a range of concentrations, that count tells you how tightly the compound holds on and how many binding sites are present. Below, we walk through what the four numbers mean, the three experimental formats that generate them, and how to read the values without being misled.

What Kd, Ki, Bmax, and IC50 actually measure

These four numbers answer two plain questions: how tightly? and how many? Getting them straight is most of the battle.

The Kd, or equilibrium dissociation constant, is the concentration of the labeled compound at which half of the available receptors are occupied at equilibrium. That half-occupancy definition anchors everything else — when the free ligand concentration equals the Kd, exactly half the sites are filled (Binding Curve Viewer, PMC). The Ki applies the same concept to an unlabeled competitor: the concentration at which it would half-occupy the receptor, worked out indirectly by watching it push the labeled compound off.

Bmax is a different kind of number altogether. It measures the total count of binding sites in the sample — the receptor density, usually reported per milligram of protein — and says nothing about affinity. IC50, finally, is the midpoint of a competition experiment: the competitor concentration that halves the specific signal under one particular set of conditions. Think of it as a stepping-stone to Ki, not a destination.

One intuition carries all the way through. Because Kd and Ki are the concentrations needed to reach half-occupancy, a lower value means a tighter binder. A compound with a nanomolar Kd grabs the receptor at a thousandth of the concentration a micromolar binder needs. Affinity and these constants run in opposite directions — which is exactly why researchers prize "sub-nanomolar" numbers.

Saturation binding: deriving Kd and Bmax

The most direct experiment measures Kd and Bmax at the same time. You take a fixed amount of receptor preparation and expose it to a range of labeled-ligand concentrations, typically from about one-tenth to ten times the Kd you expect (Assay Guidance Manual, NCBI).

The one-site hyperbola

As the labeled-ligand concentration climbs, specific binding rises steeply at first, then flattens as the sites fill up. That curve is a rectangular hyperbola, described by a compact equation: Bound = Bmax × [L] / ([L] + Kd). Fit the measured points to this model by nonlinear regression and you get both parameters at once — the plateau gives Bmax, and the concentration at half-plateau gives Kd (Assay Guidance Manual, NCBI).

Staying in "Zone A"

One experimental discipline separates a clean saturation curve from a misleading one: keep less than about 10% of the added label bound at every concentration. This "Zone A" rule guards against ligand depletion, where so much free label disappears onto the receptor that the concentration axis no longer means what the equation assumes (Assay Guidance Manual, NCBI). It's also why modern practice fits the untransformed data directly instead of reaching for the old Scatchard plot; the linearized version distorts the error structure and can bias the fitted numbers.

Specific vs nonspecific binding

Here's the practical catch that trips up newcomers: the raw signal is never all real. Some of the label genuinely sits on the target receptor. Some of it sticks to filters, tubes, membrane fragments, and off-target sites. That second component is nonspecific binding, and it has to come out before any affinity number is calculated.

The trick for isolating it is elegant. You run a parallel sample containing a large excess of unlabeled competitor — enough to occupy essentially all the real receptor sites. In that sample, the label can no longer reach the receptor, so whatever signal remains is the incidental sticking. Total binding minus that nonspecific value leaves specific binding, and specific binding is the quantity every Kd and Ki calculation actually uses (Assay Guidance Manual, NCBI).

This signal-minus-background thinking shows up across quantitative in-vitro methods. Anyone who has worked through an ELISA has met the same logic of subtracting a blank before trusting a number. In binding assays, it's simply built into the experimental design from the start.

Competition binding and the Cheng-Prusoff equation

Most compounds a lab wants to profile aren't radioactively labeled, and labeling every candidate would be impractical. Competition binding gets around that by letting an unlabeled compound reveal its affinity indirectly.

Displacing the tracer

You hold the labeled ligand at a fixed, low concentration and add increasing amounts of the unlabeled compound. As the competitor concentration rises, it occupies more sites and pushes the label off, so specific binding falls along a sigmoidal curve. The midpoint of that curve — where specific binding is halved — is the IC50 (Assay Guidance Manual, NCBI).

From IC50 to Ki

The IC50 on its own is slippery, because it shifts depending on how much labeled ligand you happened to use. The Cheng-Prusoff equation corrects for that and turns the IC50 into a Ki: Ki = IC50 / (1 + [L]/Kd), where [L] is the labeled-ligand concentration and Kd is its own affinity for the receptor (Cheng-Prusoff analysis, ScienceDirect). Because the correction leans on the label, you can't skip measuring the tracer's Kd first.

That's why IC50 isn't treated as a fundamental constant. The same compound shows a different IC50 if you change the labeled-ligand concentration, even though its true affinity hasn't budged — so IC50 varies between studies while Ki stays put (IC50 overview, ScienceDirect). A well-behaved competition curve also has a Hill slope near −1.0. A slope that departs noticeably from that flags a mixed receptor population, cooperativity, or an artifact worth chasing down (Assay Guidance Manual, NCBI).

Kinetic assays: how fast, not just how tight

Two compounds can share an identical Kd and still behave very differently, because affinity is a balance of two rates rather than a single stickiness. Kinetic assays open up that balance.

The two rate constants are kon, the association rate, which describes how quickly the compound and receptor form a complex, and koff, the dissociation rate, which describes how quickly that complex falls apart. The equilibrium constant is simply their ratio: Kd = koff / kon (real-time binding kinetics, PMC). A compound can reach a low Kd either by latching on fast or by letting go slowly — and those two routes to the same number can matter a great deal.

Real-time methods, including ones that follow labeled ligand on living cells, track association and dissociation continuously rather than at a single endpoint. That makes it possible to tell apart compounds with matching Kd values but different residence times on the receptor (real-time binding kinetics, PMC). The label-free cousin of this approach, surface plasmon resonance, reads the same kon and koff without any radioactive tracer at all — which is why kinetic thinking now sits alongside the classic equilibrium numbers.

What a binding assay does and does not tell you

It's worth being clear about the limits of the measurement. A binding assay reports that a compound occupies a receptor site, and how strongly — it doesn't report what happens after the compound binds. Occupancy and function are separate questions. A high-affinity binder might act as an activator, a blocker, or something in between.

That's why affinity data is normally paired with functional readouts. In cell-based work, researchers follow what the receptor does downstream once it's engaged — for GPCRs, that often means watching second-messenger changes with a cAMP accumulation assay, or tracking recruitment steps with beta-arrestin recruitment assays. Binding tells you the compound reaches the door; functional assays tell you what it does once it's through. Read together, and always in observed-in-research terms, they give a far more complete picture than either one alone.

Frequently Asked Questions

What is the difference between Kd and Ki?

Kd (the equilibrium dissociation constant) describes how tightly the labeled tracer itself binds the receptor, measured directly in a saturation experiment. Ki describes how tightly an unlabeled compound binds, measured indirectly by watching it compete the tracer off. Both are concentrations, and for both a lower number means higher affinity. The practical link between them is the Cheng-Prusoff equation, which converts a competition IC50 into a Ki using the tracer's concentration and its Kd.

What does a low Kd or Ki value mean?

A low value means high affinity. Because Kd and Ki are the concentrations needed to occupy half the receptors, a compound that reaches half-occupancy at a very small concentration is binding tightly. A Kd in the nanomolar range is a strong binder; a Kd in the micromolar range is a weak one. The numbers are inverses of affinity, which is why researchers speak of "sub-nanomolar" binders as the tightest.

Why report Ki instead of IC50?

IC50 is the competitor concentration that halves specific binding under one particular set of conditions, and it shifts with the tracer concentration you happened to use. That makes it hard to compare across experiments. Ki corrects for the tracer using the Cheng-Prusoff equation, so it is an intrinsic property of the receptor-compound pair. Assay-reporting guidance recommends Ki for competitive binding and reserves IC50 for mechanisms where the correction does not apply.

What is specific versus nonspecific binding?

Total signal in a binding assay includes tracer genuinely bound to the target receptor plus tracer stuck to filters, tubes, membranes, and other sites. Nonspecific binding is measured in a parallel sample containing a large excess of unlabeled competitor, which blocks the real receptor sites while the incidental sticking continues. Subtracting that nonspecific value from the total leaves specific binding, the number every affinity calculation actually uses.

The Bottom Line

Radioligand binding assays turn "binds tightly" into a measured number. Kd and Ki are both half-occupancy concentrations, so a lower value always means a tighter grip; Bmax counts the sites; and IC50 is the raw competition midpoint that the Cheng-Prusoff equation refines into a Ki. Saturation experiments hand you Kd and Bmax together, competition experiments give you Ki, and kinetic experiments break affinity into the association and dissociation rates behind it. Keep the specific-minus-nonspecific discipline and the Zone A rule in mind, and the numbers become readable rather than mysterious. Affinity is the entry point to receptor pharmacology, not the whole story — pairing these values with functional readouts is where the fuller picture comes into focus.

For research use only. Not for human or animal consumption of any kind. The information in this article is for educational purposes only and is not intended to diagnose, treat, cure, or prevent any disease. The statements made have not been evaluated by the U.S. Food and Drug Administration. These products are NOT FDA APPROVED. Please consult with a licensed healthcare professional before making any decisions regarding your health or research.

Optides LLC is a chemical supplier. Optides LLC is not a compounding pharmacy or chemical compounding facility as defined under 503A of the Federal Food, Drug, and Cosmetic Act. Optides LLC is not an outsourcing facility as defined under 503B of the Federal Food, Drug, and Cosmetic Act.