Caco-2 Permeability Assay: Papp, TEER, and Efflux Ratio

The Caco-2 monolayer permeability assay is the most widely used in vitro model for predicting how much of an orally dosed compound will cross the human small intestinal epithelium into the bloodstream. It underlies go/no-go decisions in early drug discovery and, when run against validated reference standards, supports permeability-class arguments under the Biopharmaceutics Classification System (BCS). Read correctly, it separates three distinct questions that get conflated in a single “permeability number”: how fast the compound crosses passively, whether the monolayer used to measure that was actually intact, and whether a transporter is actively working against passive diffusion.

What the Caco-2 Monolayer Models

Caco-2 is a human colorectal adenocarcinoma cell line that, when seeded onto a permeable filter support (a Transwell-type insert) and grown to confluence over roughly one to three weeks, spontaneously differentiates into a polarized epithelial monolayer with microvilli, tight junctions between cells, and a functional complement of efflux and uptake transporters. That differentiation is what makes it useful: an undifferentiated, sub-confluent culture does not resemble intestinal epithelium and will not produce a meaningful permeability value no matter how carefully the transport step is run.

The insert splits the well into two compartments. The apical side faces the filter and stands in for the intestinal lumen; the basolateral side stands in for the blood-facing side of the epithelium. A compound is dosed into one compartment (the donor) at a known concentration, and its appearance in the other compartment (the receiver) is tracked over time, typically by LC-MS/MS.

Papp: The Apparent Permeability Coefficient

The core output is Papp, the apparent permeability coefficient, calculated from the standard flux equation:

Papp = (dQ/dt) / (A × C0)

• dQ/dt — the rate at which the compound accumulates in the receiver compartment (amount per unit time), taken from the linear portion of the appearance curve
• A — the surface area of the monolayer/filter
• C0 — the initial donor concentration

Papp is expressed in cm/s and used comparatively, not as a standalone absolute number: a lab runs a panel of reference compounds with known human fraction-absorbed values — a high-permeability marker (compounds such as metoprolol or testosterone are commonly used) and a low-permeability, largely paracellular marker (mannitol or PEG 4000) — in the same experiment, and ranks the test compound’s Papp against that calibration curve rather than trusting the raw coefficient in isolation. Passive transcellular permeability correlates closely with lipophilicity, which is why Papp data is usually read alongside a lipophilicity descriptor; see LogP vs LogD for how those descriptors are defined and why LogD (pH-adjusted) tends to track passive permeability more reliably than LogP for ionizable compounds.

Why Monolayer Integrity Has to Be Verified First

A Papp value is only meaningful if it was generated across an intact monolayer. If the tight junctions between cells are compromised, the compound gets a paracellular shortcut straight through the gaps — and that shortcut inflates the apparent permeability of even a compound that has essentially no real transcellular permeability. A “high permeability” result from a leaky monolayer is not evidence of good absorption; it is an artifact of a failed cell culture. This is why methodology papers on the assay treat integrity verification as a precondition for trusting any Papp number, not an optional add-on.

Two checks are standard, and they catch different failure modes:

• TEER (transepithelial electrical resistance) — measured in Ω·cm² with a pair of electrodes across the monolayer, TEER quantifies the ionic resistance created by intact tight junctions. It is measured before dosing (and often again afterward) as a quick, non-destructive screen. Acceptance thresholds are lab- and protocol-specific — they depend on passage number, insert type, and each lab’s own validation against its reference compounds — so a specific cutoff from one lab’s SOP should not be treated as a universal regulatory number; the point is that every protocol has to define and enforce one.
• Lucifer Yellow — a small, hydrophilic, fluorescent molecule with very low intrinsic transcellular permeability, dosed alongside or immediately after the test compound and quantified by fluorescence detection (see fluorescence spectroscopy for how that quantification works). Because Lucifer Yellow can only cross an intact monolayer through the paracellular route, elevated LY flux is direct evidence of a leak — a real solute actually finding the gap, rather than TEER’s indirect electrical proxy for it. A monolayer can occasionally look marginal on TEER but pass on Lucifer Yellow, or vice versa, which is why well-run protocols use both rather than treating either one as sufficient on its own.

The Efflux Ratio: What a Unidirectional Papp Misses

Everything above describes a one-way experiment: dose apical, measure basolateral appearance (A→B), the direction that models absorption. But Caco-2 monolayers also express active efflux transporters, and a unidirectional assay cannot tell passive permeability apart from passive permeability that is being actively opposed. The fix is to also run the reverse direction — dose basolateral, measure apical appearance (B→A) — and compare the two:

Efflux ratio = Papp(B→A) / Papp(A→B)

A ratio near 1 is consistent with transport dominated by passive diffusion, which is symmetric by definition. A high ratio means net movement is disproportionately basolateral-to-apical — the compound is being pumped back toward the lumen side faster than simple diffusion alone would produce. That asymmetry is the signature of an apical efflux transporter, most commonly P-glycoprotein (P-gp/ABCB1), though BCRP (ABCG2) and MRP2 (ABCC2) can also contribute. A ratio of 2 or higher is the conventional flag used to call a compound a likely efflux substrate — the same threshold referenced in FDA’s transporter-mediated drug interaction guidance for identifying P-gp substrates in vitro.

The practical stakes: a compound can show a deceptively low apical-to-basolateral Papp not because it diffuses poorly, but because efflux is actively pumping it back out as fast as it diffuses in. Unidirectional data alone cannot distinguish “genuinely low passive permeability” from “high passive permeability masked by efflux,” and those two cases call for different responses — a genuinely impermeable scaffold may need reformulation or a different chemical series, while an efflux-limited one may respond to a small structural change that reduces P-gp affinity, or may simply carry a drug-interaction liability worth flagging (co-administered P-gp inhibitors can meaningfully increase its exposure). Confirming which transporter is responsible typically requires a follow-up step beyond the ratio itself, such as re-running the assay with a known P-gp inhibitor co-incubated and checking whether the ratio collapses toward 1.

How This Assay Fits the BCS Permeability Axis

The Biopharmaceutics Classification System sorts compounds along two independent axes — solubility and permeability — into four classes; the full framework, its class boundaries, and how it supports biowaiver arguments are covered in the BCS guide. Caco-2 Papp, benchmarked against the reference-standard calibration curve described above, is one of the accepted lines of evidence sponsors and regulators use to argue high- versus low-permeability classification, alongside artificial-membrane assays like PAMPA and, at the top of the evidentiary hierarchy, direct human intestinal perfusion or mass-balance fraction-absorbed data.

The efflux ratio matters specifically here because a compound with genuinely high passive permeability but a high efflux ratio is a harder BCS argument to make cleanly than the ratio-near-1 case: transporter-mediated efflux is exactly the kind of confound permeability-classification guidance expects sponsors to characterize and account for, not one that a single unidirectional Papp value can wave away. Downstream, that same Papp/efflux dataset is also a direct input to whole-body PBPK absorption models and sits alongside other ADME parameters — see blood-to-plasma ratio and toxicokinetics in nonclinical safety studies — that together describe a compound’s disposition beyond the gut wall.

Practical Limitations

A few caveats keep the assay honest:

• Cross-lab variability. Absolute Papp values drift with passage number, culture duration, and insert supplier. Comparing a test compound’s raw Papp against a published value from a different lab is unreliable; comparing it against reference standards run in the same plate, same day, is the standard the assay is actually designed around.
• Transporter expression isn’t a perfect match to native intestine. Some uptake transporters are expressed at different levels in Caco-2 than in native human jejunum, which is one reason PAMPA (a transporter-free artificial membrane, useful for isolating the purely passive component) and transporter-transfected lines such as MDCK-MDR1 exist as complements rather than replacements. Newer microphysiological models, including organ-on-a-chip gut models, aim to close some of that gap with multi-cell-type architecture and flow.
• It’s still a live-cell-culture assay. Reproducible results depend on routine, careful subculturing of an adherent monolayer line — see trypsinization for how that detachment and passaging step is actually done — and throughput/cost considerations that don’t apply to acellular methods like PAMPA.

Frequently Asked Questions

What does Papp actually measure in a Caco-2 assay?

Papp (the apparent permeability coefficient) is the rate at which a compound crosses an intact Caco-2 monolayer, normalized for the monolayer’s surface area and the dosing concentration. It’s calculated as Papp = (dQ/dt) / (A × C0) and is read comparatively, against reference compounds of known human absorption run in the same experiment, rather than as a standalone absolute figure.

Why measure TEER before trusting a Caco-2 permeability result?

TEER screens for intact tight junctions. A monolayer with compromised junctions lets compounds through a paracellular shortcut that inflates Papp regardless of the compound’s real transcellular permeability, so a Papp value from a monolayer that failed its TEER check isn’t a usable measurement of anything.

Is Lucifer Yellow the same check as TEER?

No. TEER is an electrical proxy for tight-junction integrity; Lucifer Yellow is a real, low-permeability solute whose actual paracellular flux is measured directly by fluorescence detection. They catch overlapping but not identical failure modes, which is why many protocols require both rather than either alone.

What does a high efflux ratio mean?

The efflux ratio compares basolateral-to-apical transport against apical-to-basolateral transport [Papp(B→A) / Papp(A→B)]. A ratio near 1 is consistent with passive diffusion alone. A ratio of roughly 2 or higher indicates an active efflux transporter — most often P-glycoprotein, sometimes BCRP or MRP2 — is pumping the compound back toward the apical side faster than passive diffusion would predict on its own.

Can a Caco-2 efflux ratio alone confirm P-glycoprotein is responsible?

Not definitively on its own. A high ratio flags likely transporter involvement, but confirming which transporter requires a follow-up step, typically re-running the assay with a selective inhibitor (a known P-gp inhibitor, for instance) co-incubated and checking whether the ratio collapses toward 1.

How does a Caco-2 result get used in BCS permeability classification?

Caco-2 Papp, calibrated against high- and low-permeability reference standards, is one of the accepted lines of evidence used to argue a compound’s permeability class under the Biopharmaceutics Classification System, which can in turn support a biowaiver argument for an immediate-release oral product. See the BCS guide linked above for the full classification framework and its regulatory use.