7.1 Absorption & Physiological Barriers (Membrane Transport, Blood-Brain Barrier, GI Absorption)

Why barriers decide the internal dose

Handbook Domain II.2 treats species differences in absorption, distribution, metabolism, and excretion as a mechanistic skill, not a trivia list. The same nominal milligram-per-kilogram dose can produce very different systemic and target-organ exposures if a membrane, a pH-driven precipitation, first-pass enzymes, or a physiologic barrier intervenes. Independent OpenExamPrep material in this section covers how xenobiotics cross membranes, how gastrointestinal, dermal, and respiratory anatomy change the absorbed fraction, and how the blood-brain barrier (BBB) both excludes chemicals and, for some conjugates, assists them. It is not an ABT product and does not claim official approval, review, or partnership with ABT.

Fickian diffusion versus transporters

Passive (Fickian) diffusion moves solute down its concentration gradient. Fick's first law states that the rate is proportional to the diffusion coefficient, surface area, and concentration difference, and inversely proportional to path length (membrane thickness). Toxicologists often collapse this to flux ≈ P × A × ΔC, where P is permeability. Small, uncharged, lipophilic molecules have high P. Over ordinary concentration ranges, passive flux is linear and is not competitively inhibited by a second chemical that merely shares a carrier.

Transporters change that picture. Solute carriers (OATPs, OCTs, OATs, PEPT1, LAT1) and ATP-binding cassette (ABC) efflux pumps (P-gp/MDR1/ABCB1, BCRP, MRP2, BSEP) can concentrate substrate or return it to the donor compartment. Carrier flux is saturable (Michaelis-Menten: v = Vmax·C/(Km+C)), often stereoselective, and inhibitable. Far below Km the curve looks almost linear and can be mistaken for diffusion. At high local concentrations the carrier saturates, the absorbed fraction falls, and a competing substrate can raise or lower net transfer. Distinguishing the two mechanisms matters when you scale a gavage solution to a dietary mixture, or when an inhibitor is co-administered.

Paracellular leak through tight junctions is a third path for small hydrophiles. It is minor in intact small intestine and almost nil at the BBB. Damage, inflammation, or chelators that open junctions can create a path that neither Fickian lipid diffusion nor a named transporter predicted.

pH partition and Henderson-Hasselbalch

Most pharmaceuticals and many industrial organics are weak acids or weak bases. The un-ionized species partitions into the bilayer much more readily than the charged species. The Henderson-Hasselbalch relations are:

• Weak acid: pH = pKa + log([A−]/[HA]); HA is un-ionized.
• Weak base: pH = pKa + log([B]/[BH+]); B is un-ionized.

A weak acid with pKa near 4 is largely un-ionized in fasting gastric fluid (pH about 1–2) and more ionized in the small intestine (pH about 6–7). A weak base with pKa near 8–9 is ionized in the stomach and more un-ionized distally. Do not conclude that weak acids are absorbed only in the stomach. Gastric surface area is small and residence time is limited. Villi and microvilli give the small intestine the area that dominates oral absorption for most xenobiotics, even when the un-ionized fraction is lower there. Ion trapping can concentrate a weak base in acidic intracellular or gastric compartments. For metals and some inorganics, pH controls solubility and precipitation, which is a different mechanism from classic pH-partition of organic electrolytes.

First-pass extraction

Material absorbed from most of the gut enters the portal vein, so it meets enterocyte enzymes and efflux pumps and then hepatocytes before any of it reaches the systemic arterial circulation. Oral bioavailability (F) can be far below the fraction that disappeared from the lumen. Gut-wall CYP3A and P-gp act as a tandem: P-gp cycles substrate back toward the lumen, giving CYP3A more opportunities to oxidize it. Intravenous dosing bypasses that first pass. Comparing oral and intravenous curves is how extraction is quantified. Species differ in intestinal CYP3A/P-gp abundance and in hepatic blood flow relative to body size, so a high-extraction compound in the rat may look more available in the dog or in humans. Rectal, buccal, and dermal routes can spare some hepatic first-pass; they do not magically remove every barrier.

Gastrointestinal anatomy: rat, dog, and human

Interpret the species you actually dosed.

• Rat: a nonglandular forestomach plus a glandular stomach; no gallbladder, so bile trickles continuously into the duodenum rather than dumping as a meal-stimulated bolus; coprophagy can recycle poorly absorbed material; relatively high metabolic rate and often extensive first-pass. Fasted gastric pH is acidic, but feeding and coprophagy change the local environment. Forestomach squamous epithelium can show irritation that has no human gastric analogue.
• Dog: a gallbladder is present; this is the usual nonrodent oral pharmacokinetic species. Fasting gastric pH can be more variable than in humans, including stretches of higher pH that change dissolution of weak bases and some salts. Bile-acid composition differs from the human pool. Gut length and transit are not a scaled-down human.
• Human: gallbladder present; fasting gastric pH typically about 1–2 and intestinal pH about 6–7; enormous absorptive area after villus and microvillus folding.

Ruminant rumen fermentation is a specialized reducing and metabolic compartment. It is not the default comparison when the laboratory species are rat, dog, and human, and it is not required to interpret most nonclinical TK packages.

Dermal absorption

Intact stratum corneum—anucleate corneocytes in a lipid matrix, the bricks-and-mortar barrier—rate-limits percutaneous absorption. Flux rises with applied concentration (until the vehicle saturates), exposed area, contact time, and damage or hydration of the barrier. Occlusion (impermeable wrap, tight gloves, clothing that traps sweat) hydrates the stratum corneum, increases partitioning from many vehicles, and can raise the absorbed dose several-fold versus the same load unoccluded. Lipophilic small molecules penetrate better than large polar ones. Ions and proteins barely move unless the barrier is broken.

Rodent skin (rat, mouse, rabbit) is generally more permeable than human skin: thinner stratum corneum and higher hair-follicle density provide shunt paths. Pig skin is closer to human. A rat dermal milligram-per-kilogram is not a human dermal milligram-per-kilogram without a permeability discussion. OECD 427 (in vivo) and 428 (in vitro) are the usual chemical methods. Industrial hygiene still has to ask whether the worker’s skin was occluded.

Respiratory absorption and particle size

Gases and vapors absorb according to blood:air partition, alveolar ventilation, and whether uptake is perfusion- or ventilation-limited. Rodents are obligate nasal breathers with complex turbinates and metabolically active olfactory epithelium, so a vapor that a human inhales partly by mouth can be scrubbed or locally bioactivated in the rat nose.

For aerosols, mass median aerodynamic diameter (MMAD) and geometric standard deviation predict regional deposition. Human mouth-breathing teaching ranges are approximate; hygroscopicity, density, and minute ventilation move the curves:

• Greater than about 10 μm: mostly extrathoracic (nasopharyngeal) impaction; much of that deposit is swallowed and becomes an oral dose.
• About 5–10 μm: tracheobronchial impaction and sedimentation.
• About 1–5 μm: greater pulmonary/alveolar deposition (the classic respirable window for many workplace aerosols).
• Submicron and ultrafine: diffusion; some alveolar deposition; a fraction may be exhaled or, for the smallest particles, translocate.

The same MMAD does not deposit in the same region in a nose-only rat and a mouth-breathing worker.

Blood-brain barrier

Brain capillary endothelial cells are joined by tight junctions (claudins, occludin), with few fenestrae, a continuous basement membrane, and astrocytic end-feet. Luminal P-gp effluxes many xenobiotics back into blood; BCRP and MRPs contribute. Small lipophilic neutrals can still diffuse. Polar nutrients use specific uptake carriers.

A high-yield way to increase brain entry is molecular mimicry. Methylmercury bound to L-cysteine resembles methionine and is a substrate for the large neutral amino acid transporter LAT1. Excess methionine can compete. D-cysteine conjugates are poor LAT1 substrates. This is the opposite of the slogan that binding always keeps a metal out of brain. Teach the conjugate and the carrier; do not treat every metal-thiol pair as equivalent, and do not invent a beryllium-BBB story to replace this mechanism.

Limited oral absorption by intestinal precipitation

Soluble beryllium salts can ionize in acidic gastric fluid. As contents reach intestinal pH (about 6–8), beryllium precipitates as poorly soluble phosphates (and related hydroxides). Little dissolved Be2+ remains for absorption, so oral bioavailability is very low and most of an ingested dose appears in feces. Occupational risk is inhalation of poorly soluble particles that reach the deep lung, not swallowing the same mass. The physicochemical lesson is general: pH-driven precipitation can dominate absorption even when the salt looks soluble on the bottle.

Realistic scenario

A rat nose-only inhalation of a 12 μm MMAD aerosol is compared with a human workplace cloud of 2 μm MMAD particles of the same chemistry, and a gavage arm is added “to cover ingestion of deposited dust.” The gavage arm understates inhalation risk if the chemical precipitates or is poorly absorbed in the intestine (beryllium salts are the teaching example). The 12 μm rat exposure deposits largely in the nose and is swallowed, so nasal metabolism and a secondary oral dose dominate. The 2 μm human cloud reaches alveoli. Treating those three internal doses as equal because the chamber and gavage concentrations were “matched” fails Domain II.2.

Traps

• Equating un-ionized gastric fraction with the site of most oral absorption.
• Treating rodent dermal milligram-per-kilogram as human milligram-per-kilogram.
• Ignoring occlusion when comparing in vitro unoccluded flux with gloved workers.
• Assuming every thiol-bound metal is excluded from brain.
• Using ruminant anatomy to explain a rat-dog-human oral package.