Caco-2 permeability: the gatekeeper for oral drugs

A pill only works if the drug gets out of the gut and into the blood. Potency at the target is irrelevant if the molecule cannot cross the single layer of cells lining the intestine. The standard in vitro proxy for that crossing is the Caco-2 assay, and its readout — apparent permeability, or Papp — is one of the first numbers a project tracks once a series shows on-target activity.

What the Caco-2 assay measures

Caco-2 is a human colorectal adenocarcinoma cell line that, when grown on a porous membrane for about three weeks, differentiates into a polarized monolayer resembling the small-intestinal epithelium. It forms tight junctions, expresses brush-border enzymes, and displays the efflux transporters (P-glycoprotein, BCRP, MRP2) you find in the real gut wall. You add compound to the apical (gut-lumen) side and measure how much appears on the basolateral (blood) side over time.

That rate is expressed as the apparent permeability coefficient, Papp, in units of cm/s. It captures everything that governs absorption: passive transcellular diffusion, paracellular leak between cells, active uptake, and active efflux pumping the drug back out.

Reading the numbers

Regulators and labs use rough Papp bins to map onto the fraction of an oral dose a human will absorb:

• High permeability: Papp greater than ~10 x 10⁻⁶ cm/s, corresponding to high human absorption (fraction absorbed ≥ 85%).
• Moderate: roughly 1 to 10 x 10⁻⁶ cm/s, mapping to partial absorption (fa ~50-84%).
• Low permeability: below ~1 x 10⁻⁶ cm/s, a warning that oral absorption will be poor (fa < 50%).

The exact cutoffs are lab-specific and depend on the reference compounds, which is why every assay is calibrated with known standards spanning the absorption range (for example mannitol or atenolol for low, propranolol or metoprolol for high). The Caco-2 model is formally recognized by the FDA, EMA, and WHO as a permeability surrogate for Biopharmaceutics Classification System (BCS) biowaivers.

The efflux ratio tells a second story

Because Caco-2 cells express transporters, you can run the assay in both directions and compute an efflux ratio (Papp basolateral-to-apical divided by Papp apical-to-basolateral). A ratio well above 2 usually means the compound is a substrate of an efflux pump, most often P-glycoprotein. That matters far beyond the gut: P-gp substrates struggle to cross the blood-brain barrier, so a high efflux ratio is an early flag for both oral-absorption trouble and poor CNS penetration. Co-dosing with a P-gp inhibitor in the assay confirms whether efflux is the culprit.

How it connects to docking

Permeability and potency pull a molecule in opposite directions. The same features that improve passive permeability — moderate lipophilicity, few hydrogen-bond donors, low polar surface area — often trade against the polar contacts that drive tight binding. A docking score tells you the molecule can grip the target; a permeability estimate tells you it can reach the target in the first place. You want both numbers in front of you before you commit to a synthesis.

Try the prediction yourself

Liganx’s ADMET panel reports a Caco-2 permeability estimate alongside the cardiac, hepatic, and metabolic risk readouts, computed by the admet-ai ensemble on every docked pose. Open Studio, dock a candidate, and open the ADMET pill on the result row to see the permeability estimate next to the binding pose. Running molecular docking online and the absorption readout together is how you catch a low-permeability liability while it is still just a docking pose, not a failed pharmacokinetic study.

Primary sources

• Artursson P, Karlsson J. Correlation between oral drug absorption in humans and apparent drug permeability coefficients in human intestinal epithelial (Caco-2) cells. Biochem Biophys Res Commun 175, 880-885 (1991). doi:10.1016/0006-291X(91)91647-U
• Hubatsch I, Ragnarsson EGE, Artursson P. Determination of drug permeability and prediction of drug absorption in Caco-2 monolayers. Nat Protoc 2, 2111-2119 (2007). doi:10.1038/nprot.2007.303
• ICH M9 Guideline. Biopharmaceutics Classification System-Based Biowaivers. FDA / ICH (2019). fda.gov/media/148472