What Can Cross the Blood-Brain Barrier?

Small, fat-loving molecules with low molecular weight cross the blood-brain barrier most easily, slipping between lipid-rich cell membranes by passive diffusion. Oxygen, carbon dioxide, ethanol, caffeine, and nicotine all pass through with relatively little resistance. Beyond those free travelers, the brain selectively imports essential nutrients through dedicated transport proteins, while actively pumping many unwanted compounds back out. The full picture involves multiple transport routes, deliberate gaps in the barrier, and a list of uninvited guests that exploit or damage the system to get through.

What Makes the Barrier a Barrier

The blood-brain barrier is not a single wall. It is a system built from specialized endothelial cells that line the brain’s capillaries, sealed together by structures called tight junctions. These junctions are dynamic protein complexes that lock neighboring cells so closely together that most water-soluble molecules in the bloodstream simply cannot squeeze between them. Supporting cells, including pericytes embedded in the vessel wall and star-shaped astrocytes wrapping the outside, help regulate how tight those seals stay and how the endothelial cells behave. Together, all of these cell types form what researchers call the neurovascular unit.¹

The practical effect is that the brain’s capillaries are far more selective than capillaries elsewhere in your body. In your muscles or your gut, small gaps between endothelial cells allow nutrients and immune cells to pass freely. In the brain, the tight junctions block that route almost completely. Anything that wants to get in has to pass through the endothelial cells themselves, not between them, which means the cell’s own membrane acts as a chemical filter.

What Slips Through by Passive Diffusion

Because the endothelial membrane is made of lipids, molecules that dissolve easily in fat tend to pass right through it. Gases like oxygen and carbon dioxide cross almost instantly. Small, lipid-soluble drugs such as ethanol, nicotine, and many anesthetics also diffuse in without needing any help from transport proteins. The key properties that predict whether a small molecule can passively cross the barrier are its lipophilicity (how fat-loving it is), its polar surface area (how much of the molecule carries an electrical charge), its molecular weight, and the number of hydrogen-bond donors it has.²

Of those properties, polar surface area turns out to be the single strongest predictor of whether a drug reaches the brain. Molecules with a large polar surface have lots of charged or hydrogen-bonding atoms exposed, which makes them stick to water rather than sliding through fatty membranes. Research on drugs that are active in the central nervous system shows that the vast majority have a polar surface area below roughly 90 square angstroms. Go much above that and the molecule is effectively locked out by passive routes alone.²

This is why water-soluble molecules, charged ions, and most proteins in the bloodstream are excluded. They are too polar, too large, or both. Even many small drugs that are technically lipophilic enough to enter the membrane get pumped back out before they accumulate in brain tissue, as the next section explains.

The Pump That Kicks Drugs Back Out

Getting into the endothelial membrane and getting into the brain are not the same thing. The barrier’s endothelial cells express a protein called P-glycoprotein on the surface facing the blood. P-glycoprotein is an energy-driven pump that recognizes an enormous variety of lipophilic compounds and actively shuttles them back into the bloodstream before they can complete their journey into brain tissue.³ It is one of the main reasons that so many otherwise promising drug candidates fail to treat brain diseases: they dissolve into the membrane just fine, but P-glycoprotein catches them and spits them back out.

Studies in mice engineered to lack P-glycoprotein showed dramatically higher brain levels of many drugs, confirming that this single pump accounts for much of the barrier’s resistance to larger lipophilic compounds, particularly those over about 400 daltons in molecular weight.⁴ P-glycoprotein is not the only efflux pump at work, but it handles the widest range of substrates and is expressed at the highest levels, making it the primary obstacle for pharmaceutical development.⁵

Nutrients the Brain Imports on Purpose

The brain consumes a huge share of the body’s glucose but cannot store much of it, so there has to be a reliable way to pull glucose out of the blood continuously. Dedicated transporter proteins, especially one called GLUT1, sit in the endothelial membrane and ferry glucose molecules across. This is carrier-mediated transport: the molecule binds to a specific protein, the protein changes shape, and the molecule is released on the other side. Amino acids, vitamins, and other essential nutrients each have their own family of carriers.

Glucose transport across the barrier is not a simple one-way pipeline. In vitro studies using brain endothelial cell models show that the direction and rate of glucose movement depend on the concentration gradient between the blood-facing and brain-facing sides of the endothelium.⁶ This keeps supply matched to demand while preventing the brain from being flooded when blood sugar spikes.

Large Molecules and Receptor-Mediated Transcytosis

Proteins like insulin, transferrin (the protein that carries iron), and certain growth factors are far too large and polar to diffuse through the membrane. Instead, they bind to specific receptors on the blood-facing side of the endothelial cell, which triggers the cell to engulf the molecule in a small bubble of membrane and carry it across to the brain side. This process, called receptor-mediated transcytosis, is highly selective: only molecules that fit the receptor get ferried through.

Transferrin receptor transcytosis has attracted intense interest from drug developers because it offers a way to sneak large therapeutic molecules into the brain. The idea is to attach a drug to an antibody or nanoparticle that binds the transferrin receptor, hijacking the brain’s own iron-delivery system. Research has shown that getting the binding strength right is critical. Antibodies that grip the receptor too tightly get trapped inside the endothelial cell instead of being released into the brain. Lowering the binding affinity, or using single-armed antibodies, actually improves transport across the barrier.⁷ Work on another receptor target, CD98 heavy chain, is expanding the toolkit for brain-targeted delivery of large molecules.⁸

Where the Barrier Has Gaps

Not every part of the brain has a blood-brain barrier. A handful of small structures clustered around the brain’s fluid-filled ventricles, known collectively as the circumventricular organs, have capillaries that are deliberately leaky. These regions need direct access to the bloodstream because their job is to sense circulating hormones, blood chemistry, and immune signals and relay that information to the rest of the brain.⁹

There are roughly seven circumventricular organs, and they split into two functional groups. Secretory ones, like the pineal gland and the neurohypophysis (the back part of the pituitary), release hormones into the blood. Sensory ones, like the area postrema and the subfornical organ, monitor the blood for signals the brain needs to respond to.¹⁰ The area postrema, for instance, detects toxins in the blood and can trigger vomiting, which is why certain chemotherapy drugs cause nausea so reliably: the drugs reach the area postrema without having to cross a barrier at all. These windows are tightly regulated by surrounding glial cells to prevent leakiness from spreading into adjacent brain tissue.¹¹

Pathogens That Break In

Viruses and bacteria are not supposed to cross the blood-brain barrier, but several have evolved strategies to do so. Viruses can damage the tight junctions directly, either by disrupting the structural scaffolding inside endothelial cells or by triggering enzymes that chew through the protein matrix holding cells together.¹² Once the junctions are loosened, viral particles and inflammatory cells from the blood can flood in.

Bacteria use a broader playbook. Some, like certain strains of E. coli, group B streptococcus, and Listeria, can trick endothelial cells into engulfing them and carrying them across the barrier, essentially hijacking the transcytosis machinery the brain uses for nutrients.¹³ Others pry apart tight junctions and squeeze through the gaps between cells. A third strategy is the Trojan horse approach: a bacterium infects a white blood cell in the bloodstream, rides inside it as it crosses the barrier during normal immune surveillance, and then replicates once inside the brain.¹⁴ Bacterial meningitis often involves a combination of these routes, compounded by the host’s own inflammatory response further damaging the barrier.¹⁵

Toxic Metals

Heavy metals like lead, mercury, manganese, and iron reach the brain more readily than you might expect given how well the barrier excludes other substances. They manage this by exploiting the same carrier-mediated transport systems the brain uses to import essential nutrients, by diffusing passively when they are in lipid-soluble chemical forms, or by slipping through damaged areas of the endothelium.¹⁶ Lead, for example, can substitute for calcium in certain transport processes. Manganese can hitch a ride on iron and calcium transporters.

The chemical form of a metal matters enormously. Organic mercury compounds, like the methylmercury found in contaminated fish, are lipophilic enough to diffuse across the barrier in both directions. Inorganic mercury species behave differently: in laboratory models, inorganic mercury was actually transported preferentially out of the brain-facing compartment, suggesting the barrier may have some protective capacity against inorganic forms.¹⁷ Regardless of route, toxic metals can also damage the barrier itself once they accumulate, making it leakier to other harmful substances.¹⁸

Why Some Drugs Make You Drowsy and Others Do Not

The difference between a first-generation antihistamine that knocks you out and a newer one that does not comes down almost entirely to blood-brain barrier penetration. Older antihistamines like diphenhydramine are small, lipophilic, and flexible molecules that cross the barrier easily and bind histamine receptors in the brain, causing sedation. Newer antihistamines were deliberately redesigned with zwitterionic structures, meaning they carry both positive and negative charges, which dramatically reduces their ability to cross the barrier and limits their effects to the body’s peripheral tissues.¹⁹

The same principle works in reverse for Parkinson’s disease. The brain needs more dopamine, but dopamine itself cannot cross the barrier. Instead, patients take L-dopa, a precursor molecule that is small and polar enough to be recognized by amino acid transporters in the barrier. Once inside the brain, L-dopa is converted into dopamine.²⁰ This workaround, using a prodrug that the barrier’s own transport machinery will accept, remains one of the most successful strategies in neuropharmacology decades after its introduction.

Emerging Strategies to Open the Barrier on Demand

For diseases like brain tumors and neurodegenerative conditions, the barrier that protects the brain also prevents most therapeutic drugs from reaching it. Researchers are developing several methods to temporarily and locally open the barrier when treatment requires it.

Focused ultrasound combined with injected microbubbles is the most clinically advanced approach. A beam of ultrasound energy is aimed at a specific brain region using MRI or neuronavigation guidance. Tiny gas-filled microbubbles in the bloodstream oscillate in response to the ultrasound waves, mechanically loosening the tight junctions in that area for a few hours.²¹ The opening is reversible and can be targeted with millimeter precision. Recent clinical work has demonstrated this technique in pediatric brain tumor patients, where focused ultrasound successfully opened the barrier as frequently as every two days without requiring sedation.²²

Nanoparticles are another active area of development. By coating tiny drug-carrying particles with ligands that target receptors on the barrier’s endothelial cells, researchers aim to trick the barrier into pulling drugs across via transcytosis. Transferrin, the iron-carrying protein, is a popular coating choice because it binds the transferrin receptor already present on endothelial cells.²³ Various nanoparticle platforms can also exploit carrier-mediated transport, adsorptive transcytosis, or even paracellular routes when the barrier is partially disrupted by disease.²⁴

A completely different approach bypasses the barrier altogether. The nasal cavity connects to the brain through olfactory and trigeminal nerve pathways, and drugs delivered as nasal sprays can travel along these nerve routes directly into brain tissue without ever entering the general bloodstream.²⁵ This nose-to-brain route is already being explored for treatments targeting Alzheimer’s disease and brain cancers.

What Weakens the Barrier Over Time

The blood-brain barrier is not a static structure. It degrades with age, and the rate of decline varies by brain region and sex. A study using non-invasive MRI in 186 cognitively normal people aged 8 to 92 found that barrier function, measured by the rate at which water exchanges between blood and brain, drops sharply starting in the early sixties. That decline was more pronounced in men than women, particularly in the parietal and temporal regions of the brain.²⁶

Sleep loss also appears to weaken the barrier. Chronic sleep deprivation triggers a low-grade inflammatory state in which immune signaling molecules circulate at elevated levels. These inflammatory molecules can act directly on the endothelial cells and supporting cells of the barrier, loosening tight junctions and causing pericytes to detach from capillary walls.²⁷ Animal models of sleep loss show increased barrier permeability and decreased expression of the tight junction proteins that hold the seal together.²⁸

Barrier breakdown has also been observed as one of the earliest detectable changes in Alzheimer’s disease, appearing before significant neurodegeneration or dementia symptoms.²⁹ Whether this leakiness is a cause of the disease, a consequence of early pathology, or both remains an active question, but the finding has shifted some researchers toward viewing Alzheimer’s partly as a vascular disorder.³⁰

Signals From the Gut

Bacteria in your gut produce small molecules called short-chain fatty acids when they ferment dietary fiber. These metabolites enter the bloodstream from the gut lining and can reach the brain’s circulation, where there is active research into whether and how they cross the blood-brain barrier and influence brain function.³¹ Animal studies suggest short-chain fatty acids may help maintain barrier integrity by supporting tight junction proteins, which would mean your gut microbiome indirectly influences how permeable your blood-brain barrier is. The research is still in relatively early stages, but it represents one of the more surprising ways the barrier turns out to be connected to the rest of the body.

How Researchers Study the Barrier

Much of what we know about barrier permeability comes from two complementary approaches: imaging the living human brain and building miniature barrier models in the lab.

Dynamic contrast-enhanced MRI is the main tool for measuring barrier leakiness in living people. A contrast agent is injected into the blood, and the MRI tracks how quickly it seeps into brain tissue. Validation work has shown that blood volume in a brain region is the strongest predictor of how much contrast leaks through, with grey matter showing higher permeability than white matter.³² Nuclear imaging techniques and intravital fluorescence microscopy offer complementary views, each with different strengths in resolution and the types of transport they can detect.³³

On the lab bench, organ-on-chip models recreate the barrier in miniature by growing human endothelial cells, pericytes, and astrocytes in tiny channels with fluid flowing through them to mimic blood flow. These microfluidic systems produce molecular permeability values close to what is seen in living tissue and allow researchers to test drug candidates and watch barrier responses in real time.³⁴ A persistent challenge is standardizing how different labs measure permeability and flow conditions, which currently makes it hard to compare results across different chip designs.³⁵

The Barrier Across the Animal Kingdom

Virtually all complex animals have some form of blood-brain barrier, but the cellular architecture varies dramatically. In insects and other arthropods, the barrier is built from glial cells sealed by septate junctions rather than endothelial cells with tight junctions. Octopuses and squids have two distinct barriers depending on the type of brain blood vessel: one made by glial cells and another by pericytes. Sharks, skates, and sturgeons also rely on glial-cell barriers. Most other vertebrates, including all mammals, shifted to the endothelial-cell model seen in humans.³⁶ Despite these structural differences, the molecular signaling that builds and maintains the barrier shows striking conservation across species, suggesting the basic blueprint is ancient and the specific cell type performing the job is secondary to the function itself.