The blood–brain barrier as an endocrine tissue

Abstract

The blood–brain barrier (BBB) was first noted for its ability to prevent the unregulated exchange of substances between the blood and the central nervous system (CNS). Over time, its characterization as an interface that enables regulated exchanges between the CNS and substances that are carried in the blood in a hormone-like fashion have emerged. Therefore, communication between the CNS, BBB and peripheral tissues has many endocrine-like properties. In this Review, I examine the various ways in which the BBB exhibits endocrine-related properties. The BBB is a target for hormones, such as leptin and insulin, that affect many of its functions. The BBB is also a secretory body, releasing substances either into the blood or the interstitial fluid of the brain. The BBB selectively allows classical and non-classical hormones entry to and exit from the CNS, thus allowing the CNS to be both an endocrine target and a secretory tissue. The BBB is affected by endocrine diseases such as diabetes mellitus and can cause or participate in endocrine diseases, including those related to thyroid hormones and obesity. The endocrine-like mechanisms of the BBB can extend the definition of endocrine disease to include neurodegenerative conditions, including Alzheimer disease, and of hormones to include cytokines, triglycerides and fatty acids.

Key points

  • The blood–brain barrier (BBB) acts as both a secretory and target endocrine tissue.

  • By regulating the transport of hormones into and out of the brain, the BBB provides a mechanism by which the central nervous system can act as an endocrine secretory and target tissue.

  • The BBB facilitates substances not typically thought of as hormones acting in an endocrine-like fashion, including triglycerides, short-chain fatty acids and lipopolysaccharide.

  • BBB function can be altered in endocrine diseases either because of adaptions to the disease condition or because it is a disease target.

  • BBB dysfunction or impairment can cause or promote the progression of endocrine or metabolic diseases.

  • The BBB is an important factor in the treatment of many diseases, often as a therapeutic target or as a barrier that must be negotiated.

Introduction

The blood–brain barrier (BBB) was originally defined by its ability to prevent the unregulated entry of blood-borne substances into the central nervous system (CNS). At least three barriers are currently recognized: a vascular BBB formed by brain endothelial cells; a blood–cerebrospinal fluid (CSF) barrier formed by epithelial cells; and a barrier formed by tanycytes, which are special ependymal cells within the brain, that interfaces between the circumventricular organs and other adjacent brain tissue1. These barriers all possess tight junctions (Box 1) that block intercellular leakage, have reduced macropinocytosis and contain no fenestrae (small pores in the cell membrane), which together all reduce intracellular leakage (Fig. 1).

Fig. 1: The presence and architecture of tight junctions results in polarization of the brain endothelial cell.
figure1

a | Endocytosis, membrane protein distribution and receptor response in a nonpolarized cell all occur in a single extracellular environment. b | Tight junctions create polarization so that the extracellular environment is divided into luminal (blood) and abluminal (brain) spaces. Transcytosis (blood–brain barrier penetration) and exocytosis can occur. Receptor activation can result in a response on the luminal or abluminal side of the cell. Inset shows that the fence function of tight junctions prevents the exchange of luminal and abluminal outer membrane leaflet contents, allowing for unique populations of lipids and proteins. The fence function extends only to the outer leaflet and therefore does not divide the inner leaflet into distinct luminal and abluminal regions.

These BBBs perform many functions in addition to that of barrier formation. Through the transport of glucose, vitamins, minerals and other substances, the BBBs supply the CNS with its nutritional needs2. Through efflux transporters and enzymatic activities, the BBBs help rid the brain of toxins produced there and prevent those exogenous to the CNS from accumulating or entering the CNS in the first instance3. By regulating the influx and efflux of informational molecules such as peptides and regulatory proteins, and because it is a source of informational molecules itself, the BBBs, and especially the vascular BBB, act as interfaces in a humoral-based communication axis between the CNS and the blood4,5.

These extra-barrier properties create a blood–brain interface6 that is endowed with many endocrine-like properties (Fig. 2). In brief, these are the ability of brain endothelial cells to react to both traditional hormones and other circulating elements (thus, the barrier cells act as classic endocrine target tissues); the ability of barrier cells, especially brain endothelial cells, to secrete substances into the blood that affect other cells (thus, the barrier cells can act as classic endocrine secretory tissues); the ability of the BBBs to transport substances from the blood into the brain and from the brain into the blood (thus, the BBBs selectively overcome their own blockade, allowing the brain to act as an endocrine target and facilitating it as an endocrine secretory tissue); the BBBs can be targets of endocrine diseases, and the resulting alterations in BBB properties can underlie some of the clinical manifestations of those endocrine diseases; and finally, malfunctions of the BBBs can cause endocrine diseases. Together, these five properties provide unique ways in which the BBB can be involved in the treatment of endocrine diseases.

Fig. 2: The vascular blood–brain barrier as an endocrine tissue.
figure2

The vascular blood–brain barrier (BBB) is a target for circulating substances in the blood7,8,11,12,17,18,19. In addition to this, the vascular BBB secretes substances into the blood that act in autocrine, paracrine and haemocrine fashions34,36,37,38,39,40,41,45,46,47,48 and it regulates the exchange of informational substances between the central nervous system and the blood50,51,55,64,65,66,80,81, resulting in humorally mediated brain–body communication.

In this Review, I explore these aspects of the endocrine BBB. Some of these properties have many examples, and I am unable to cover them all. Therefore, this Review focuses on the endocrine-like properties of the vascular BBB and defines the underlying principles that are currently understood.

The BBB as an endocrine target

The many barrier and interface functions of the BBB require its performance to be fine-tuned to the needs of the CNS and to be adaptable to the changing peripheral environment. As such, the BBB responds to hormones, which makes it an endocrine target tissue. For example, oestradiol modulates sodium–potassium–chloride co-transporters, and through this mechanism it can influence post-stroke brain oedema7. 1,25-dihydroxyvitamin D3 regulates the expression of the efflux transporter LDL receptor-related protein 1 (LRP1) and the influx transporter receptor for advanced glycation end products (RAGE)8 (of note, among the many ligands of these transporters are the amyloid-β proteins9,10). Through the AT1 receptor, angiotensin II modifies BBB permeability, including via transcytosis11,12. Thus, angiotensin II could be the hormone responsible for opening the BBB during hypertensive emergencies, resulting in hypertensive encephalopathy13.

Although insulin does not modulate GLUT1, which is the BBB transporter for glucose, insulin has other effects on BBB functions. For example, insulin enhances the transport of tryptophan across the BBB, which in turn could affect brain levels of serotonin and kynurenine, substances that are related to sleep and depression14,15,16. Insulin also regulates the activity or affects the levels of brain endothelial cell alkaline phosphatase, glutathione and P-glycoprotein17,18,19, and insulin increases leptin transport across the BBB20. The BBB transporter for insulin itself is, in turn, regulated by several substances, including levels of iron, cholecystokinin and triglycerides21,22,23.

Many other functions of the BBB are regulated in an endocrine-like fashion. Leptin transport is modulated by blood-borne glucose, triglycerides, adrenaline and probably oestrogens20,24. Amylin alters tyrosine and tryptophan transport25, the BBB efflux pump peptide transport system 1 (PTS1) is altered by branched-chain amino acids26 and the mannose 6-phosphate receptor (which transports some lysosomal enzymes across the neonatal BBB) is modulated by circulating adrenaline and retinoic acid27,28,29. Evidence suggests that the decreases in the expression of GLUT1 associated with a high-fat diet are rescued by vascular endothelial growth factor30.

A number of substances have been proposed that could regulate the induction or integrity of the tight junctions, which, as mentioned earlier, are critical for maintaining the barrier functions of the BBB. Most of these substances are secreted from cells on the brain side of the BBB, particularly pericytes and astrocytes. However, some classic hormones might also have a role in regulating tight junctions, including glucagon-like peptide 1 (GLP1) and transforming growth factor 1 (refs31,32). In addition to substances that could regulate tight junctions, substances that were not traditionally thought of as hormones circulate in the blood and affect the function of the BBB and brain endothelial cells. These include cytokines, triglycerides, free fatty acids and lipopolysaccharide (LPS)32,33,34,35 (Box 2).

The aforementioned examples illustrate several principles: barrier cells constitute an endocrine tissue that responds to circulating factors; alterations in the BBB affect all the known categories of BBB function (that is, effects on barrier properties, transporters, enzymatic activities and secretory processes); and some effects are mediated through classic hormones, such as insulin. Other blood-borne effectors are not typically thought of as hormones, such as triglycerides and LPSs, and some effectors do not cross the intact BBB (for example, adrenaline). Other effectors do cross, including those with effects on other transporters. For example, insulin, which is transported across the BBB, increases leptin transport and triglycerides (which are also transported across the BBB), affecting both insulin and leptin transport. Thus, the scene is set for complex physiological interactions. As a result of these endocrine effects, the BBB is able to adapt to environmental influences that, in turn, affect brain function. Hence, the BBB aids the CNS in detecting and responding to environmental conditions as presented to it through the bloodstream. Signals coming from the CNS to the brain endothelial cells allow the BBB to respond to the changing needs of the brain.

The BBB as an endocrine secretory tissue

All the barrier cells are able to secrete a number of substances, but of the barrier cells, brain endothelial cells are the most widely studied. Brain endothelial cells can release nitric oxide, prostaglandins, cytokines and adrenomedullin36,37,38,39,40,41. The epithelial cells that constitute the blood–CSF barrier and the tanycytes that form a barrier between circumventricular organs and adjacent brain tissue also have secretory functions42,43. Secretions are both constitutive and induced and are influenced by the other cells of the neurovascular unit44. IL-6 and granulocyte–macrophage colony-stimulating factor (GM-CSF; which is classically know as a white blood cell growth factor) secreted from brain endothelial cells act in an autocrine or paracrine fashion to enhance the translocation of the HIV-1 virus across the BBB45. The secretions from barrier cells can also affect other cells within the CNS. For example, brain endothelial cells secrete fever-mediating prostaglandins46,47 and IL-1β and IL-6, which influence the uptake of copper and iron by glial cells34,48.

A mechanism for intercellular crosstalk within the CNS, and the basis for an endocrine-like axis that connects peripheral tissue events with brain functions, is the process by which circulating substances act at the luminal side of brain endothelial cells to stimulate the release of substances from these cells into brain interstitial fluid that then act on other brain cells. In summary, the ability of barrier cells to secrete substances into blood establishes an endocrine-like axis; the ability of barrier cells to secrete substances into the brain compartment in response to blood hormones establishes a mechanism by which those hormones can influence brain function without crossing the BBB.

The aforementioned examples illustrate several principles: all barrier cells (endothelial, epithelial and tanycytic) are secretory; secretions can be constitutive or induced; for brain endothelial cells, secretions can be into the bloodstream or into the CNS; and paracrine and autocrine effects have been demonstrated. Whether substances derived and released by brain endothelial cells into the bloodstream affect distal peripheral tissues in the classic haemocrine fashion has not been demonstrated.

Transfer of hormones across the BBB

Pathways and mechanisms

The barrier function of the BBB would ordinarily deny water-soluble hormones entry to the CNS, thus making it impervious to many circulating signals. However, the BBB is permeable to many hormones, which makes the CNS an endocrine target tissue.

There are many examples of circulating substances that cross the BBB to act on the CNS. Classic examples of substances that cross the BBB are the steroid hormones that cross by the non-saturable process of transcellular diffusion. The rate at which these steroids enter the brain is determined primarily by their lipid solubility but is also influenced by their molecular mass as well as the extent and type of plasma protein binding. Accumulation within the brain is dictated by avidity of the CNS binding receptor, especially when compared with the avidity of binding to plasma proteins. The presence of brain-to-blood efflux systems can also affect the accumulation of a substance, as demonstrated for dehydroepiandrosterone sulfate49.

As shown by the classic studies of Geraldo David and Tapan Anand Kumar, and of Samuel Marynick and colleagues, the degree of uptake of steroid hormones ranges widely, as exemplified by the accumulation of 17α hydroxyprogesterone being ~200 times greater than that of testosterone50,51. Many of the efflux systems that include steroid hormones among their ligands are members of the organic anion transporting polypeptide (OATP) and the ATP-binding cassette (ABC) transporter families49,52,53. Non-sulfated steroids are predominantly transported in the brain-to-blood direction at both the vascular BBB and the choroid plexus. However, for hydrophilic sulfated steroid hormones, which are in far greater abundance than non-sulfated steroids, current evidence strongly suggests that the main direction of transport is blood to brain and is mediated by the OATPs53,54. Brain endothelial cells can also act as a enzymatic barrier to some steroids54.

Other classic hormones, such as thyroid hormones, insulin, leptin and ghrelin, cross the BBB by saturable processes. Many of the major saturable transport systems, such as GLUT1, are non-energy-requiring systems (that is, they function via facilitated diffusion). Other transport systems rely on pores or are transcytotic — the latter always require energy and are variably classified as carrier-mediated transport, receptor-mediated transport, receptor-mediated transcytosis and adsorptive transcytosis. A substance that is the primary ligand for a transport system typically crosses at a rate that is 10–100-fold greater than the rate predicted on the basis of its lipid solubility and other physicochemical characteristics2. A saturable transport system also means that the BBB can negate the effect of binding to those plasma proteins when the BBB transporter has a higher binding affinity than the plasma protein.

Thyroid hormones

Thyroid hormones are transported across cell membranes by various transporters, but only monocarboxylate transporter 8 (MCT8) and OATP1C1 are thought to be physiologically relevant55. Thyroid hormone receptors do not function as BBB transporters, and the large neutral amino acid (LNAA) transporters LAT1 and LAT2, although capable of transporting thyroid hormones, do not seem to have a physiological role.

Evidence suggests that there are species specific differences in the rate of thyroid hormone transfer across the BBB. For example, in humans thyroid hormone transfer relies on MCT8, whereas in murine endothelial cells OATP1C1 is abundantly expressed in addition to MCT8 (refs56,57). On the basis of mutation and knockout studies, MCT8 in mice favours T3 transport, whereas OATP1C1 favours T4 transport; MCT8 might also have some efflux function56. This combination of influx and efflux transport is consistent with an early study in mice finding that T3 crosses mainly in the blood-to-brain direction and T4 is dominated by an efflux system58.

Deiodinases are abundantly expressed in the brain59 and they are thought to have a major role not only in the conversion of T4 to T3 but also in the catabolism of thyroid hormones to inactive forms. Finally, free iodide is rapidly transported out of the brain60,61. MCT8 and OATP1C1 are also abundantly expressed in the choroid plexus. When injected into the lateral ventricle, however, T3 and T4 penetrate only into the periventricular area. Such limited periventricular distribution is the rule for substances injected into the CSF62,63. As such, it is likely that the choroid plexus is a source of thyroid hormones for brain regions that are in contact with the CSF.

Insulin

Insulin is a classic hormone that in mammals is secreted predominantly by a single tissue, the pancreas. Insulin is transported across the BBB by a saturable system64,65,66. In the periphery, insulin acts as a metabolic hormone, inducing the uptake and metabolism of glucose by tissues that express GLUT4, such as muscle, adipose tissues and the liver. Transport of glucose across the BBB, however, depends on the non-insulin-sensitive GLUT1, and few brain cells possess GLUT4. Many of the properties of CNS insulin are more mitogenic (that is, they are involved in growth) than metabolic (such as being used for energy); it seems as if peripheral and CNS insulin have followed separate evolutionary paths67. It should be noted, however, that studies that investigate the actions of substances, not just hormones, on the CNS often administer pharmacologically relevant doses, therefore the elucidated actions on the CNS might not occur physiologically. In addition, any substance that is injected into the CSF will eventually enter the bloodstream as the CSF is reabsorbed. Studies have shown that substances injected into the CSF can be found peripherally in the blood. In some cases, those levels are the same as if the substance had been given as an intravenous infusion68,69. Ironically, then, the effects produced after an intracerebroventricular injection can actually on occasion be mediated at peripheral sites of action.

Research published between 1984 and 1992 reported that CNS insulin influenced feeding and blood glucose levels; however, the effects were paradoxical to those of insulin-induced hypoglycaemia70,71. That is, CNS insulin inhibits feeding, elevates blood glucose and even reduces blood insulin. Therefore, insulin is one of several hormones that have paradoxical CNS effects, raising the possibility that they act as their own counter-regulatory hormones72.

CNS insulin acts in the hypothalamus to control ~50% of hepatic glucose production by stimulating sympathetic outflow to the liver73,74. Although, it is insulin’s mitogenic-like effects on brain cells and its possible use in the treatment of Alzheimer disease that have raised the most excitement of late75. As mentioned previously, the actions of insulin in the CNS are more mitogenic than metabolic and resemble more closely those of the ancestral insulins than those of peripheral insulin76. Insulin administered to the brains of healthy humans, cognitively impaired humans or mouse models of Alzheimer disease results in rapid improvement in cognition and has effects on neuronal survival, plasticity of synapses, dendritic arborization and the formation of neuronal circuits77,78,79.

Ghrelin

Ghrelin is illustrative of other hormones that are predominantly synthesized in the periphery but are capable of crossing the BBB80,81 in quantities sufficient to influence CNS function. Aside from having effects on feeding and energy expenditure, ghrelin also has effects on cognition, synaptic function, neuronal survival and neurogenesis (effects that are shared by insulin and leptin)82,83,84,85.

Fibroblast growth factors

Fibroblast growth factors (FGFs) are a large family of regulatory proteins with many functions, including endocrine functions. FGF2 (basic FGF), FGF19 (human homologue of mouse FGF15, henceforth referred to as FGF15/FGF19) and FGF21 have all been shown to cross the BBB, and there is preliminary evidence for FGF1 (acidic FGF) crossing86,87,88,89. In many cases, the ability of a peripherally administered FGF to cross the BBB underlies its ability in part or in whole to induce CNS effects, including those on analgesia86, circadian behaviour90 and metabolism91. FGF15/FGF19 and FGF21 can circulate as hormones and have acute metabolic effects that are mediated, at least in part, through the liver and adipose tissues91. Both FGF15/FGF19 and FGF21 increase energy expenditure and decrease plasma insulin, glucose and triglyceride concentrations in the liver in obese animals, but in a chronic setting these effects, as well as weight loss, are predominantly mediated through the CNS. Part of the CNS effect of FGF15/FGF19 and FGF21 is mediated through sympathetic outflow from the CNS to brown fat. Peripherally administered FGF1 also acts within the CNS to decrease hyperglycaemia92.

The glucose-normalizing effect of FGF1 is not dependent on weight loss, reduced food intake, changes in basal glucose turnover or production rate, glucose uptake by heart or adipose tissues or changes in blood insulin or glucagon levels. Unlike FGF15/FGF19, FGF1 has little or no effect on the hypothalamic–pituitary–adrenal axis and only has acute effects on sympathetic outflow92. Instead, FGF1 administered by intracerebroventricular injection was shown to increase hepatic and skeletal muscle uptake of glucose, a process that probably occurred via an insulin receptor-dependent mechanism92.

Leptin

The examples of BBB–hormone transport pathways that I have provided in this Review so far mostly relate to brain endothelial cells and the vascular BBB. It is worth noting, however, that although the epithelial barrier cells of the choroid plexus and tanycytes found at the circumventricular organs share characteristics with brain endothelial cells and the vascular BBB, each barrier results in unique distribution patterns for the hormones or other substances it transports93,94. The unique regional distribution patterns are a result of the very low diffusion rates within brain tissue. Furthermore, the low diffusion rates from the CSF into brain tissue means that when a substance is introduced into the CSF it is fairly evenly distributed throughout cranial CSF, but diffusion from the CSF into adjacent brain tissue is poor. As a result of this poor diffusion, a substance transported into the CSF by the choroid plexus remains mostly periventricular60,62,63,95.

In theory, the three barriers (the vascular BBB, the blood–CSF barrier and the tanycytic barrier at the circumventricular organs) work in harmony to support brain function, but there are some examples of how these barriers would affect a specific hormone or influence a given brain region (Box 3; Fig. 3). Studies in rodents that investigated how peripheral leptin is distributed in the arcuate nucleus following transport across the BBB in a fed and non-fed state offer some interesting examples of how the three barriers can interact (Fig. 3). In the fed state, leptin is transported across the vascular BBB by brain endothelial cells at the rate of ~5.9 × 10−4 ml/g, or ~20 times faster than albumin80. In addition to the vascular BBB, a second source of leptin is transported across the choroid plexus and into the CSF69,96.

Fig. 3: Blood-to-brain transport of leptin into the arcuate nucleus: integration of signals and changes with feeding state.
figure3

a | In the fed state, blood-derived leptin (blue dots) reaches the arcuate nucleus by transport across local capillaries with barrier function80, transport across the epithelial cells of the choroid plexus into cerebrospinal fluid (CSF) with subsequent leakage into brain tissue across leaky ependymal cells69,96 and transport into CSF and adjacent brain tissue by the tanycytic barrier after leakage into the median eminence93. b | Three changes found at 24 h of fasting are an extension of the tanycytic barrier into the ventricular region, an increase in the blood–brain barrier (BBB) transport rate of leptin and the conversion of some capillaries in the ventromedial region of the arcuate nucleus to a leaky state93,98. c | At 48 h of fasting, leptin blood levels are greatly reduced, blood triglycerides are elevated, blood triglycerides impair leptin transport across the BBB contributing to peripheral leptin resistance and triglycerides cross the BBB to interfere with activation of the leptin receptor, contributing to central leptin resistance33,93,98,99.

As mentioned earlier, owing to low diffusion rates from the CSF to the brain, tissue concentrations of BBB-transported leptin decrease dramatically and logarithmically from the ventricular surface63 so that leptin remains mainly in the periventricular region of the arcuate nucleus69. The study by Berislav Zlokovic and colleagues96, which was conducted in rats, reported a transport rate at the choroid plexus that was ten times higher than the transport rate found by Naoko Nonaka and colleagues97 in the mouse. One explanation could be that the rat, with its larger arcuate nucleus and therefore its larger diffusional distances, requires higher rates of leptin transport into the CSF than the mouse.

A third pathway that leptin can take to the arcuate nucleus tissue is via diffusion across the leaky BBB of the median eminence. Leptin that enters the CNS via this route is subsequently transported by tanycytes into the CSF and arcuate nucleus tissue. Without this transport function, the barrier function of the tanycytes would prevent leptin from diffusing from the median eminence into the CSF or arcuate nucleus93.

Dramatic changes occur in these systems with food restriction. After 24 h of fasting, leptin transport across the vascular BBB increases98, the tanycytic barrier extends further along the ventricle and some of the brain endothelial cells in the arcuate nucleus closest to the median eminence lose barrier function93. Whereas these events probably increase leptin levels in the arcuate nucleus after 24 h of fasting, the changes at 48 h act in unison to decrease leptin action, resulting in leptin deficiency, leptin peripheral resistance and leptin central resistance. By 48 h of fasting, serum leptin levels are decreased, the leptin transport rate across the vascular BBB is greatly decreased and there is resistance at the leptin receptor. The latter two events — inhibition of leptin transport and receptor binding — are mediated at least in part by triglycerides, which are greatly increased in the blood with starvation and are transported across the BBB33,99 (Fig. 3).

The CNS can contribute levels of a substance in the blood. Such substances enter the bloodstream not only through efflux transporters located at the vascular BBB and choroid plexus but also with the reabsorption of CSF into the bloodstream. Indeed, because of this efflux, substances injected into the lateral ventricles of the brain can produce blood levels that resemble that of an intravenous infusion68. The renowned neuroendocrinologist Seymour Reichlin showed that on stimulation, the CNS can be an impressive source of blood cytokines100,101,102. Others have shown that corticotropin-releasing hormone given into the lateral ventricle of the brain can enter the circulation in amounts sufficient to modulate splenic secretion of IL-1β100.

The microbiota

Once in the bloodstream, the secreted products of the gut microbiota might have many effects on BBB functions and the neuroimmune axes in which the BBB participates103,104. In addition, the products of the microbiome, as exemplified by the short-chain fatty acids, cross the BBB to exert effects on the CNS. A 2016 study showed that short-chain fatty acids crossing the BBB increase microglial activation to the extent that the nature of the microbiome could be a risk factor for Parkinson disease105.

The brain as an endocrine tissue

As detailed in this section, a number of fundamental principles govern the transfer of hormones across the BBB. First, every major class of hormone (steroids, peptides, regulatory proteins and thyroid hormones) has representatives that cross the BBB at physiologically relevant levels. Second, a variety of transfer mechanisms across the BBB exist, with transcellular diffusion and blood-to-brain saturable transporters being the best understood mechanisms. Third, for some hormones, brain-to-blood transporters, plasma protein binding, enzymatic barrier function or CNS receptor affinity is a major or the major determinant of CNS uptake or retention. Furthermore, in many cases, the protein acting as the hormone’s receptor is not the same protein acting as its transporter.

This section elucidated several other points. As illustrated by leptin, the various barrier systems (endothelial–vascular, epithelial–CSF and tanycytic–CSF–circumventricular organ) have ‘spheres of influence’; that is, the distribution of a hormone within the CNS differs depending on the barrier it has crossed. Leptin is also an example of how BBB–hormone transporters are not static but are regulated by peripheral events, including by other hormones.

In some cases, the effect of a hormone on the CNS is complementary to its peripheral action (CNS insulin’s effect on hepatic glucose production); in other cases, it can be contradictory so that in the CNS the hormone acts in counter-regulatory fashion to its peripheral effects (CNS insulin’s ability to increase blood sugar and suppress appetite). In addition to these outcomes, the effects induced in the CNS by a hormone can seem unrelated to those of its peripheral actions (CNS insulin’s effect on cognition). As with the other endocrine functions of the BBB, substances acting in a hormone-like fashion include those not typically classified as hormones and have actions that are not typically thought of as endocrine (triglycerides and leptin resistances). Furthermore, a blood-borne hormone might exert an effect in the periphery that is dependent on a CNS site of action (insulin’s effect on CNS-mediated hepatic glucose output). In this case, the blood-to-brain transfer of a hormone and the subsequent CNS action is part of a more complex brain–body interaction.

Research has also shown that brain-to-blood transfer mechanisms (such as transporters and CSF reabsorption) can, under some circumstances, contribute markedly to blood levels of a hormone as illustrated by leptin, cytokines and potassium. In addition, products from the gut microbiome such as short-chain fatty acids can cross the BBB to influence the CNS.

The BBB as a target of endocrine disease

Both the BBB and the blood–retinal barrier are disrupted in diabetes mellitus in both humans and animal models of diabetes106,107. In streptozotocin-induced diabetes, BBB disruption is progressive, involving bigger lesions and more regions of the brain with disease duration107. Blood–retinal barrier disruption, which results in diabetic retinopathy, is the leading cause of blindness in Western countries. BBB disruption also occurs in diabetes mellitus. Disruption in the occipital cortex can contribute to the visual deterioration108 seen in diabetes. BBB disruption is closely linked with cognitive impairments109 and precedes cognitive impairment110.

In both BBB and blood–retinal barrier disruption, the main cause of disruption seems to arise from pericytic glucotoxicity111,112. Pericytes, which are instrumental in instructing brain endothelial cells to form and maintain barrier functions113,114, take up glucose in an insulin-independent manner so that at high levels more glucose enters the tricarboxylic acid (TCA) cycle, producing more ATP but also increasing oxidative stress. This oxidative stress is blamed for pericyte death, which leads to BBB disruption and blood–retinal barrier disruption. Pericyte loss has been proposed to have a role in other types of brain diseases with a BBB component, including Alzheimer disease, cerebral cavernous malformation and methamphetamine toxicity115,116,117.

In addition to pericyte integrity, other aspects of the BBB can be affected by diabetes mellitus or obesity. Leptin, choline and insulin transport, as well as the efflux transporter LRP1, are decreased in diabetes mellitus or with obesity98,118,119,120,121. Furthermore, the expression of numerous proteins by brain endothelial cells, such as cytoskeletal proteins, enzymes and transport-related proteins, is decreased with obesity122. The lipid composition of brain endothelial cells is not altered in the brains of patients with diabetes mellitus or obesity, but brain endothelial cells in the brains of these individuals do have increased oxidative damage123.

Hyperhomocysteinaemia is also associated with BBB disruption124. Evidence suggests that disruption is via the N-methyl-d-aspartate receptor, a finding that is consistent with homocysteine being an agonist for this receptor. As for diabetes mellitus, BBB disruption precedes the cerebral pathology found in this condition125.

In summary, barrier and transporter functions can be altered in the brains of patients with endocrine and metabolic diseases. Furthermore, barrier dysfunction leads to clinical disease, as exemplified by blindness resulting from blood–retinal disruption and cognitive changes resulting from BBB disruption.

The BBB as a cause of endocrine disease

Thyroid-hormone-related conditions

Impaired expression of MCT8, the main BBB transporter of thyroid hormones in humans, results in Allan–Herndon–Dudley syndrome55. Brain lesions are present prenatally. Interestingly, mice with an MCT8 deficiency do not show altered brain structures or evidence for cognitive impairments. This lack of pathology in MCT8-deficient mice is thought to be caused by the higher expression of OATP1C1 in mice56,57. OATP1C1 favours the transport of T4, but high levels of deiodinases in brain convert T4 to T3. Mice lacking both MCT8 and OATP1C1 have delayed cerebellar development, reduced myelination and other pathologies consistent with CNS hypothyroidism56.

Interestingly, in humans, MCT8 deficiency results in a very different clinical presentation from cretinism, another form of thyroid deficiency that has roots in maternal iodine deficiency126. Cretinism and MCT8 deficiency have many differences in disease course; for example, cretinism results in maternal and fetal whole-body thyroid hormone deficiency, whereas MCT8 deficiency has a paradoxical stage during the perinatal period of cerebral hyperthyroidism. Cerebral hyperthyroidism does not occur in MCT8-deficient mice that also have a knockout of LAT2 (ref.127), suggesting that LAT2 might be an important transporter of T3 in neonates and might even be upregulated during this period in the absence of MCT8.

Glucose and insulin

Impaired transport of glucose and insulin across the BBB have each been proposed as important factors in Alzheimer disease128. Impaired transport of glucose across the BBB certainly results in cognitive impairments, seizures and ketosis as exemplified by the GLUT1 deficiency syndrome129. Despite having an elevated serum insulin level, patients with Alzheimer disease have a decreased level of insulin in the CSF as well as a decreased CSF:serum ratio for insulin, suggestive of a defect in the BBB transport of insulin130. Insulin resistance in the CNS, so-called type 3 diabetes mellitus, can occur independently of peripheral insulin resistance131; indeed, the relationship and interactions between these two spheres of insulin resistance are largely unexplored (Box 4). Tine Sartorius and colleagues132 have noted that in comparison with young mice, aged mice have a delayed increase in cortical brain activity after the peripheral, but not after the central, administration of insulin, suggesting that an impaired transport of insulin across the BBB could be a contributor to CNS insulin resistance132.

Leptin and leptin resistance

The BBB is directly involved in peripheral leptin resistance and is indirectly involved in central leptin resistance, thus contributing to obesity, the metabolic syndrome and diabetes mellitus. Leptin is secreted primarily by adipose cells and is transported across the BBB by a saturable system80,133. Leptin’s rate of transport into the arcuate nucleus area of the hypothalamus is particularly high but is also robust into the hippocampus134. A loss of the anoretic effects of leptin results in profound obesity in both rodents and humans — leptin-deficient mice at 10 months of age weigh more than three times their leptin-replete littermates, and leptin-deficient humans have BMIs in the low 50s135,136.

Leptin-deficient humans and animals are exquisitely sensitive to treatment with leptin. For example, 18 months of leptin replacement therapy resulted in BMI levels decreasing from an average of 51.2 to 26.9 (ref.135). Leptin deficiency is rare in humans, with cohorts of individuals in Turkey135 and Paraguay137. Elevated blood leptin levels in the face of obesity are common in humans and can result from attenuated transport at the BBB, resistance at the leptin receptor or deficits in signalling downstream of the leptin receptor138,139,140. Resistance at the BBB occurs early with the onset of obesity and likely participates in a positive feedback-loop-like fashion to promote obesity.

The mechanisms underlying the development, progression and maintenance of obesity and leptin resistance have been influenced by evolutionary pressures to acquire excess calories and to retain them as fat. Wild-living adult primates have fat masses of ~6% of their body weight141, which is about the same as seen in persons with anorexia nervosa142. Hunter-gatherers typically have BMIs of ~19 (ref.143). In other words, the evolutionary struggle has been to acquire a caloric reserve to see the individual through times when caloric acquisition falls short of immediate requirements and not to avoid obesity and its metabolic consequences.

These evolutionary pressures can be seen in the BBB pharmacokinetic curves for leptin80,134. Leptin transport across the BBB is saturable, which means that the relationship between serum and CNS leptin levels resembles a hyperbolic curve. Where this hyperbolic curve is linear, the blood concentrations of leptin are such that they produce the greatest proportional increases in CNS leptin levels. As blood leptin levels increase, the curve flattens (loses linearity), which results in a smaller proportion of blood leptin entering the CNS134,144. This loss of linearity is the pharmacokinetic proximal cause and an immediate cause of peripheral resistance to leptin. The loss of linearity for both rodent brain levels of leptin and human CSF levels of leptin occurs between 5 and 10 ng/ml (refs134,145), a level typically associated with ideal body weight146. As such, leptin resistance is already occurring in humans at levels of BMI that Western medicine has traditionally labelled ideal body weight, but not at levels seen in wild-living primates or hunter-gatherers.

Thus, saturable transport at the BBB of a hormone adds an additional level of potential resistance to that of a classic resistance syndrome. In non-barrier tissues, receptor resistance can be addressed by increased production of the hormone until either the response returns to normal or hormone production reaches its maximum. However, when the hormone crosses the BBB by a saturable mechanism, the BBB can act as a choke point. Saturation of the transporter at the BBB limits how much hormone can enter the brain, and that limit can be insufficient to raise brain tissue levels to those needed to overcome receptor resistance. In the case of leptin, peripheral resistance is the term used to denote the inability of the BBB to deliver sufficient amounts of hormone to the brain to overcome brain receptor (central) resistance.

Triglycerides are a cause of peripheral leptin resistance, and elevation of triglycerides is the classic dyslipidaemia of the metabolic syndrome. Triglycerides (Fig. 3) act at the brain endothelial cell barrier to decrease the rate of leptin transport across the BBB, thus they are capable of contributing to peripheral leptin resistance33. Triglycerides are also transported across the BBB and can act at the CNS leptin receptor to induce central leptin resistance99. Thus, the BBB as either a target or a conduit is involved in both peripheral and central leptin resistance.

BBB transporter dysfunction and endocrine disease

In summary, BBB transporter dysfunction occurs for a wide range of substances including thyroid hormones, glucose, insulin and leptin. Dysfunction of BBB transporters results in changes in a wide range of symptoms, including cerebral hypothyroidism, cognitive impairment, metabolic dysfunctions and obesity. To date, the clearest connections between BBB transporters and disease exist for substances the brain does not synthesize or can synthesize only in limited amounts. The involvement of the BBB in an endocrine disease might be due to the primary loss of a transporter, as exemplified by Allan–Herndon–Dudley syndrome, or as a downstream event of the disease or condition that modulates the transporter, as exemplified by hypertriglyceridaemia and the metabolic syndrome.

Endocrine BBB and treatment of diseases

The multiple facets of the endocrine BBB that make it vulnerable to being a target or a cause of endocrine disease also suggest roles for it in the treatment of endocrine diseases. Clearly, the role of the BBB as a transporter of hormones into the CNS, which allows the CNS to act as an endocrine target tissue, presents many possibilities.

Phenylketonuria

Phenylketonuria results from a defective metabolism of the LNAA phenylalanine. The other LNAAs share with phenylalanine a transporter across the BBB, therefore an excess of circulating phenylalanine results in both an increase in brain phenylalanine and a decrease in brain levels of other LNAAs caused by competitive inhibition147. CNS symptoms can arise both as a result of toxic levels of phenylalanine and because of a deficiency in neurotransmitter substrates. Although the disease is typically treated with diets low in phenylalanine, several studies have advocated treating with a mixture of LNAAs, which would both block phenylalanine from crossing the BBB in high amounts and increase the uptake of LNAAs that act as precursors to neurotransmitters148,149. A similar strategy has also been advocated for maple syrup urine disease, in which branched-chain LNAAs accumulate in the brain148.

Diabetes mellitus and insulin resistance

The potential merit of the delivery of FGFs to the brain for the treatment of diabetes mellitus is clear from the work discussed above. Indeed, the discovery92 that when doses of FGFs one-tenth of an effective peripheral dose are delivered into the brain they reverse (apparently permanently) mild to moderate levels of hyperglycaemia in some models of moderate obesity shows the potential of combining the endocrine brain and the endocrine BBB. Modifications to transporters, theoretically exemplified by the ability of α-adrenergics to overcome peripheral leptin resistance or triglycerides to enhance transport of insulin and ghrelin, could enable increased access of endogenous hormones to the CNS23,33,150.

Peripheral insulin resistance, impaired transport of glucose and insulin across the BBB and central insulin resistance are all implicated in Alzheimer disease. Dual incretin receptor agonists (twincretins) directed at GLP1 and glucose-dependent insulinotropic polypeptide that can cross the BBB are able to improve peripheral glucose control and improve the BBB transport of glucose151,152,153. Whether they can affect CNS insulin resistance or affect insulin transport across the BBB is yet to be published. GLP1 has also been proposed to prevent diabetic disruption of the BBB by limiting the oxidative stress that arises from the hyperglycaemic conditions31.

Conclusions

The BBB performs as both an endocrine target and an endocrine secretory tissue. By selectively regulating the exchange between the CNS and the blood of traditional hormones as well as other informational molecules, the BBB also allows the CNS to act as both an endocrine target and an endocrine secretory tissue. The BBB is a target for some endocrine diseases and is an active participant in the onset and promulgation of endocrine disease; as a result, the BBB can also be key to the treatment of some endocrine diseases. Finally, the view of the BBB as an endocrine tissue expands the list of hormones to include cytokines, lipids and even LPSs.

References

  1. 1.

    Neuwelt, E. et al. Strategies to advance translational research into brain barriers. Lancet Neurol. 7, 84–96 (2008).

  2. 2.

    Davson, H. & Segal, M. B. (eds) in Physiology of the CSF and Blood–Brain Barriers 303–485 (CRC Press, Boca Raton, 1996)

  3. 3.

    Banks, W. A. in Efflux Transporters and the Blood-Brain Barrier (ed. Taylor, E. M.) 21–53 (Nova Science Publishers Inc.,2005).

  4. 4.

    Pan, W. & Kastin, A. J. Cytokine transport across the injured blood-spinal cord barrier. Curr. Pharm. Des. 14, 1620–1624 (2008).

  5. 5.

    Kastin, A. J. & Pan, W. Blood-brain barrier and feeding: Regulatory roles of saturable transport systems for ingestive peptides. Curr. Pharm. Des. 14, 1615–1619 (2008).

  6. 6.

    Banks, W. A. From blood-brain barrier to blood-brain interface: new opportunities for CNS drug delivery. Nat. Rev. Drug Discov. 15, 275–292 (2016).

  7. 7.

    O’Donnell, M. E., Lam, T. I., Tran, L. Q., Foroutan, S. & Anderson, S. E. Estradiol reduces activity of the blood-brain barrier Na-K-Cl cotransporter and decreases edema formation in permenent middle cerebral artery occlusion. J. Cereb. Blood Flow Metab. 26, 1234–1249 (2006).

  8. 8.

    Guo, U. X. et al. 1,25-Dihydroxyvitamin D3 regulates expression of LRP1 and RAGE in vitro and in vivo, enhancing Abeta1 = 40 brain-to-blood efflux and peripheral uptake transport. Neruoscience 322, 28–38 (2016).

  9. 9.

    Deane, R., Wu, Z. & Zlokovic, B. V. RAGE (yin) versus LRP (yang) balance regulates Alzheimer amyloid beta-peptide clearance through transport across the blood-brain barrier. Stroke 35, 2628–2631 (2004).

  10. 10.

    Deane, R., Sagare, A. & Zlokovic, B. The role of the cell surface LRP and soluble LRP in blood-brain barrier A beta clearance in Alzheimer’s disease. Curr. Pharm. Des. 14, 1601–1605 (2008).

  11. 11.

    Fleegal-DeMotta, M. A., Dohgu, S. & Banks, W. A. Angiotensin II modulates BBB permeability via activation of the AT1 receptor in brain endothelial cells. J. Cereb. Blood Flow Metab. 29, 640–647 (2009).

  12. 12.

    Guillot, F. L. & Audus, K. L. Angiotensin peptide regulation of fluid-phase endocytosis in brain microvessel endothelial cell monolayers. J. Cereb. Blood Flow Metab. 10, 827–834 (1990).

  13. 13.

    Vaughan, C. J. & Delanty, N. Hypertensive emergencies. Lancet 356, 411–417 (2000).

  14. 14.

    Cangiano, C. et al. On the stimulation by insulin of tryptophan transport across the blood-brain barrier. Biochem. Int. 7, 617–627 (1983).

  15. 15.

    Daniel, P. M., Love, E. R., Moorhouse, S. R. & Pratt, O. E. The effect of insulin upon the influx of tryptophan into the brain of the rabbit. J. Physiol. 312, 551–562 (1981).

  16. 16.

    Walker, A. K., Wing, E. E., Banks, W. A. & Dantzer, R. Leucine competes with kynurenine for blood-to-brain transport and prevents lipopolysaccharide-induced depression-like behavior in mice. Mol. Psychiatry. https://doi.org/10.1038/s41380-018-0076-7 (2018).

  17. 17.

    Langston, J. W., Li, W., Harrison, L. & Aw, T. Y. Activation of promoter activity of the catalytic subunit of gamma-glutamylcysteine ligase (GCL) in brain endothelial cells by insulin requires antioxidant response element 4 and altered glycemic status: implication for GCL expression and GSH synthesis. Free Radic. Biol. Med. 51, 1749–1757 (2011).

  18. 18.

    Liu, H. et al. Insulin regulates P-glycoprotein in rat brain microvessel endothelial cells via an insulin receptor-mediated PKC/NF-kappaB pathway but not a PI3K/Akt pathway. Eur. J. Pharmacol. 602, 277–282 (2009).

  19. 19.

    Catalan, R. E., Martinez, A. M., Aragones, M. D., Miguel, B. G. & Robles, A. Insulin action on brain microvessels; effect on alkaline phosphatase. Biochem. Biophys. Res. Commun. 150, 583–590 (1988).

  20. 20.

    Kastin, A. J. & Akerstrom, V. Glucose and insulin increase the transport of leptin through the blood-brain barrier in normal mice but not in streptozotocin-diabetic mice. Neuroendocrinology 73, 237–242 (2001).

  21. 21.

    Ben-Shachar, D., Yehuda, S., Finberg, J. P., Spanier, I. & Youdim, M. B. Selective alteration in blood-brain barrier and insulin transport in iron-deficient rats. J. Neurochem. 50, 1434–1437 (1988).

  22. 22.

    May, A. A., Liu, M., Woods, S. C. & Begg, D. P. CCK increases the transport of insulin into the brain. Physiol. Behav. 165, 392–397 (2016).

  23. 23.

    Urayama, A. & Banks, W. A. Starvation and triglycerides reverse the obesity-induced impairment of insulin transport at the blood-brain barrier. Endocrinology 149, 3592–3597 (2008).

  24. 24.

    Kastin, A. J., Akerstrom, V. & Maness, L. M. Chronic loss of ovarian function decreases transport of leptin into mouse brain. Neurosci. Lett. 310, 69–71 (2001).

  25. 25.

    Chance, W. T., Balasubramaniam, A., Thomas, I. & Fischer, J. E. Amylin increases transport of tyrosine and tryptophan into the brain. Brain Res. 593, 20–24 (1992).

  26. 26.

    Banks, W. A. & Kastin, A. J. Modulation of the carrier-mediated transport of the Tyr-MIF-1 across the blood-brain barrier by essential amino acids. J. Pharmacol. Exp. Ther. 239, 668–672 (1986).

  27. 27.

    Urayama, A., Grubb, J. H., Banks, W. A. & Sly, W. S. Epinephrine enhances lysosomal enzyme delivery across the blood-brain barrier by up-regulation of the mannose 6-phosphate receptor. Proc. Natl Acad. Sci. USA 31, 12873–12878 (2007).

  28. 28.

    Urayama, A. et al. Alpha adrenergic induction of transport of lysosomal enzyme across the blood-brain barrier. PLOS ONE 10, e0142347 (2015).

  29. 29.

    Urayama, A., Grubb, J. H., Sly, W. S. & Banks, W. A. Pharmacologic manipulation of lysosomal enzyme transport across the blood-brain barrier. J. Cereb. Blood Flow Metab. 36, 476–476 (2016).

  30. 30.

    Jais, A. et al. Myeloid-cell-derived VEGF maintains brain glucose uptake and limits cognitive impairment in obesity. Cell 165, 882–895 (2016).

  31. 31.

    Fukuda, S. et al. Glucagon-like peptide-1 strengthens the barrier integrity in primary cultures of rat brain endothelial cells under basal and hyperglycemic conditions. J. Mol. Neurosci. 29, 211–219 (2016).

  32. 32.

    Dohgu, S. et al. Transforming growth factor-·1 upregulates the tight junction and p-glycoprotein of brain microvascular endothelial cells. Cell. Mol. Neurobiol. 24, 491–497 (2004).

  33. 33.

    Banks, W. A. et al. Triglycerides induce leptin resistance at the blood-brain barrier. Diabetes 53, 1253–1260 (2004).

  34. 34.

    McCarthy, R. C. & Kosman, D. J. Activation of C6 glioblastoma cell ceruloplasmin expression by neighboring human brain endothelia-derived interleukins in an in vitro blood-brain barrier model system. Cell Commun. Signal 12, 65 (2014).

  35. 35.

    Yu, C., Kastin, A. J., Tu, H., Waters, S. & Pan, W. TNF activates P-glycoprotein in cerebral microvascular endothelial cells. Cell Physiol. Biochem. 20, 853–858 (2007).

  36. 36.

    Reyes, T. M., Fabry, Z. & Coe, C. L. Brain endothelial cell production of a neuroprotective cytokine, interleukin-6, in response to noxious stimuli. Brain Res. 851, 215–220 (1999).

  37. 37.

    Fabry, Z. et al. Production of the cytokines interleukin 1 and 6 by murine brain microvessel endothelium and smooth muscle pericytes. J. Neuroimmunol. 47, 23–34 (1993).

  38. 38.

    Kis, B. et al. Cerebral endothelial cells are a major source of adrenomedullin. J. Neuroendocrinol. 14, 283–293 (2002).

  39. 39.

    Mandi, Y. et al. Nitric oxide production and MDR expression by human brain endothelial cells. Anticancer Res. 18, 3049–3052 (1998).

  40. 40.

    Verma, S., Nakaoke, R., Dohgu, S. & Banks, W. A. Release of cytokines by brain endothelial cells: a polarized response to lipopolysaccharide. Brain Behav. Immun. 20, 449–455 (2006).

  41. 41.

    McGuire, T. R. et al. Release of prostaglandin E-2 in bovine brain endothelial cells after exposure to three unique forms of the antifungal drug amphotericin-B: role of COX-2 in amphotericin-B induced fever. Life Sci. 72, 2581–2590 (2003).

  42. 42.

    Ebling, F. J. P. & Lewis, J. E. Tanycytes and hypothalamic control of energy metabolism. Glia 66, 1176–1184 (2018).

  43. 43.

    Ghersi-Egea, J. F. et al. Molecular anatomy and functions of the choroidal blood-cerebrospinal fluid barrier in health and disease. Acta Neuropathol. 35, 337–361 (2018).

  44. 44.

    Banks, W. A., Kovac, A. & Morofuji, Y. Neurovascular unit crosstalk: pericytes and astrocytes modify cytokine secretion patterns of brain endothelial cells. J. Cereb. Blood Flow Metab. 38, 1104–1118 (2018).

  45. 45.

    Dohgu, S., Fleegal-DeMotta, M. A. & Banks, W. A. Lipopolysaccharide-enhanced transcellular transport of HIV-1 across the blood-brain barrier is mediated by luminal microvessel IL-6 and GM-CSF. J. Neuroinflamm. 8, 167 (2011).

  46. 46.

    Engstrom, L. et al. Lipopolysaccharide-induced fever depends on prostaglandin E2 production specifically in brain endothelial cells. Endocrinology 153, 4849–4861 (2012).

  47. 47.

    Inoue, W. et al. Brain-specific endothelial induction of prostaglandin E(2) synthesis enzymes and its temporal relation to fever. Neurosci. Res. 44, 51–61 (2002).

  48. 48.

    McCarthy, R. C. & Kosman, D. J. Mechanisms and regulation of iron trafficking across the capillary endothelial cells of the blood-brain barrier. Front. Mol. Neurosci. 8, 31 (2015).

  49. 49.

    Miyajima, M. et al. Organic anion transporter 3 mediates the efflux transport of an amphipathic organic anion, dehydroepiandrosterone sulfate, across the blood-brain barrier in mice. Drug Metab. Dispos. 39, 814–819 (2011).

  50. 50.

    David, G. F. X. & Kumar, T. C. A. Transfer of steroidal hormones from blood to the cerebrospinal fluid in the rhesus monkey. Neuroendocrinology 14, 114–120 (1974).

  51. 51.

    Marynick, S. P., Haven, W. W., Ebert, M. H. & Loriaux, D. L. Studies on the transfer of steroid hormones across the blood-cerebrospinal fluid barrier in Rhesus monkey. Endocrinology 99, 400–405 (1976).

  52. 52.

    Ohtsuki, S. et al. Dominant expression of androgen receptors and their functional regulation of organic anion transporter 3 in rat brain capillary endothelial cells: comparison of gene expression between the blood-brain and -retinal barriers. J. Cell. Physiol. 204, 896–900 (2005).

  53. 53.

    Grube, M., Hagen, P. & Jedlitschky, G. Neurosteroid transport in the brain: role of ABC and SLC transporters. Front. Pharmacol. 9, 354 (2018).

  54. 54.

    Qaiser, M. Z. et al. Uptake and metabolism of sulphated steroids by the blood-brain barrier in the adult male rat. J. Neurochem. 142, 672–685 (2017).

  55. 55.

    Bernal, J., Guadano-Ferraz, A. & Morte, B. Thyroid hormone transporters — functions and clinical implications. Nat. Rev. Endocrinol. 11, 405–416 (2015).

  56. 56.

    Mayerl, S. et al. Transporters MCT8 and OATP1C1 maintain murine brain thyroid hormone homeostasis. J. Clin. Invest. 124, 1987–1999 (2014).

  57. 57.

    Roberts, L. M. et al. Expression of the thyroid hormone transporters monocarboxylate transporter-8 (SLC16A2) and organic ion transporter-14 (SLCO1C1) at the blood-brain barrier. Endocrinology 149, 6251–6261 (2008).

  58. 58.

    Banks, W. A., Kastin, A. J. & Michals, E. A. Transport of thyroxine across the blood-brain barrier is directed primarily from brain to blood in the mouse. Life Sci. 37, 2407–2414 (1985).

  59. 59.

    Kodding, R., Fuhrmann, H. & van zur Muhlen, A. Investigations on iodothyronine deiodinase activity in the maturing brain. Endocrinology 118, 1347–1352 (1986).

  60. 60.

    Cserr, H. F. & Berman, B. J. Iodide and thiocyanate efflux from brain following injection into rat caudate nucleus. Am. J. Physiol. 4, F331–F337 (1978).

  61. 61.

    Davson, H. & Hollingsworth, J. R. Active transport of 131 I across the blood-brain barrier. J. Physiol. 233, 327–347 (1973).

  62. 62.

    Blasberg, R. G. Methotrexate, cytosine arabinoside, and BCNU concentration in brain after ventriculocisternal perfusion. Cancer Treat Rep. 61, 625–631 (1977).

  63. 63.

    Maness, L. M., Banks, W. A., Zadina, J. E. & Kastin, A. J. Periventricular penetration and disappearance of icv Tyr-MIF-1, DAMGO, tyrosine, and albumin. Peptides 17, 247–250 (1996).

  64. 64.

    Woods, S. C. & Porte, D. Jr. Relationship between plasma and cerebrospinal fluid insulin levels of dogs. Am. J. Physiol. 233, E331–E334 (1977).

  65. 65.

    Margolis, R. U. & Altszuler, N. Insulin in the cerebrospinal fluid. Nature 215, 1375–1376 (1967).

  66. 66.

    Greco, A. V., Ghirlanda, G., Fedeli, G. & Gambassi, G. Insulin in the cerebro spinal fluid of man. Eur. Neurol. 3, 303–307 (1970).

  67. 67.

    McRory, J. E. & Sherwood, N. M. Ancient divergence of insulin and insulin-like growth factor. DNA Cell Biol. 16, 939–949 (1997).

  68. 68.

    Bradbury, M. W. B., Segal, M. B. & Wilson, J. Transport of potassium at the blood-brain barrier. J. Physiol. 221, 617–632 (1972).

  69. 69.

    Maness, L. M., Kastin, A. J., Farrell, C. L. & Banks, W. A. Fate of leptin after intracerebroventricular injection into the mouse brain. Endocrinology 139, 4556–4562 (1998).

  70. 70.

    Brief, D. J. & Davis, J. D. Reduction of food intake and body weight by chronic intraventricular insulin infusion. Brain Res. Bull. 12, 571–575 (1984).

  71. 71.

    McGowan, M. K., Andrews, K. M. & Grossman, S. P. Chronic intrahypothalamic infusions of insulin or insulin antibodies alter body weight and food intake in the rat. Physiol. Behav. 51, 753–766 (1992).

  72. 72.

    Banks, W. A. & Kastin, A. J. Physiological consequences of the passage of peptides across the blood-brain barrier. Rev. Neurosci. 4, 365–372 (1993).

  73. 73.

    Sandoval, D. A., Obici, S. & Seeley, R. J. Targeting the CNS to treat type 2 diabetes. Nat. Rev. Drug Discov. 8, 386–398 (2009).

  74. 74.

    Pocai, A. et al. Hypothalmic K(ATP) channels control hepatic glucose production. Nature 434, 1026–1031 (2005).

  75. 75.

    Biessels, G. J. & Reagan, L. P. Hippocampal insulin resistance and cognitive dysfunction. Nat. Rev. Neurosci. 16, 660–671 (2015).

  76. 76.

    Banks, W. A. Insulin in the brain: There and back again. Pharmacol. Ther. 136, 82–93 (2012).

  77. 77.

    Nelson, T. J., Sun, M. K., Hongpaisan, J. & Alkon, D. L. Insulin, PKC signaling pathways and synaptic remodeling during memory storage and neuronal repair. Eur. J. Pharmacol. 585, 76–87 (2008).

  78. 78.

    Biessels, G. J., Bravenboer, B. & Gispen, H. W. Glucose, insulin and the brain: modulation of cognition and synaptic plasticity in health and disease: a preface. Eur. J. Pharmacol. 490, 1–4 (2004).

  79. 79.

    Ferrario, C. R. & Reagan, L. P. Insulin-mediated synaptic plasticity in the CNS: anatomical, functionaland temporal contexts. Neuropharmacology 136, 182–191 (2017).

  80. 80.

    Banks, W. A., Kastin, A. J., Huang, W., Jaspan, J. B. & Maness, L. M. Leptin enters the brain by a saturable system independent of insulin. Peptides 17, 305–311 (1996).

  81. 81.

    Banks, W. A., Tschop, M., Robinson, S. M. & Heiman, M. L. Extent and direction of ghrelin transport across the blood-brain barrier is determined by its unique primary structure. J. Pharmacol. Exp. Ther. 302, 822–827 (2002).

  82. 82.

    Diano, S. et al. Ghrelin controls hippocampal spine synapse density and memory performance. Nat. Neurosci. 9, 381–388 (2006).

  83. 83.

    O’Malley, D. et al. Leptin promotes rapid dynamic changes in hippocampal dendritic morphology. Mol. Cell Neurosci. 35, 559–572 (2007).

  84. 84.

    Garza, J. C., Guo, M., Zhang, W. & Lu, X. Y. Leptin increases adult hippocampal neurogenesis in vivo and in vitro. J. Biol. Chem. 283, 18238–18247 (2008).

  85. 85.

    Greco, S. J. et al. Leptin reduces pathology and improves memory in a transgenic mouse model of Alzheimer’s disease. J. Alzheimers Dis. 19, 1155–1167 (2010).

  86. 86.

    Cuevas, P., Carceller, F., Munoz-Willery, I. & Gimenez-Gallego, G. Intravenous fibroblast growth factor penetrates the blood-brain barrier and protects hippocampal neurons against ischemia-reperfusion injury. Surg. Neurol. 49, 77–83 (1998).

  87. 87.

    Cuevas, P. et al. Central nervous system distribution of fibroblast growth factor injected into the blood stream. Neurol. Res. 18, 267–272 (1996).

  88. 88.

    Hsuchou, H., Pan, W. & Kastin, A. J. Fibroblast growth factor 19 entry into brain. Fluids Barriers CNS 10, 32 (2013).

  89. 89.

    Wagner, J. P., Black, I. B. & DiCicco-Bloom, E. Stimulation of neonatal and adult brain neurogenesis by subcutaneous injection of basic fibroblast growth factor. J. Neurosci. 19, 6006–6016 (1999).

  90. 90.

    Bookout, A. L. et al. FGF21 regulates metabolism and circadian behavior by acting on the nervous system. Nat. Med. 19, (1147–1152 (2013).

  91. 91.

    Lan, T. et al. FGF19, FGF21, and an FGFR1/β-klotho-activating antibody act on the nervous system to regulate body weight and glycemia. Cell Metab. 26, 709–718 (2017).

  92. 92.

    Scarlett, J. M. et al. Central injection of fibroblast growth factor 1 induces sustained remission of diabetic hyperglycemia in rodents. Nat. Med. 22, 800–806 (2016).

  93. 93.

    Prevot, V. et al. The versatile tanycyte: a hypothalamic integrator of reproduction and energy metabolism. Endocr. Rev. 39, 333–368 (2018).

  94. 94.

    Johanson, C. E., Duncan, J. A., Stopa, E. G. & Baird, A. Enhanced prospects for drug delivery and brain targeting by the choroid plexus-CSF route. Pharm. Res. 22, 1011–1037 (2005).

  95. 95.

    de Lange, E. C. et al. Application of intracerebral microdialysis to study regional distribution kinetics of drug in rat brain. Br. J. Pharmacol. 116, 2538–2544 (1995).

  96. 96.

    Zlokovic, B. V. et al. Differential regulation of leptin transport by the choroid plexus and blood-brain barrier and high affinity transport systems for entry into hypothalamus and across the blood-cerebrospinal fluid barrier. Endocrinology 141, 1434–1441 (2000).

  97. 97.

    Nonaka, N., Hileman, S. M., Shioda, S., Vo, P. & Banks, W. A. Effects of lipopolysaccharide on leptin transport across the blood-brain barrier. Brain Res. 1016, 58–65 (2004).

  98. 98.

    Banks, W. A. & Farrell, C. L. Impaired transport of leptin across the blood-brain barrier in obesity is acquired and reversible. Am. J. Physiol. 285, E10–E15 (2003).

  99. 99.

    Banks, W. A. et al. Triglycerides cross the blood-brain barrier and induce central leptin and insulin receptor resistance. Int. J. Obes. (Lond.) 42, 391–397 (2018).

  100. 100.

    Romero, L. I., Kakucska, I., Lechan, R. M. & Reichlin, S. Interleukin-6 (IL-6) is secreted from the brain after intracerebroventricular injection of IL-1β in rats. Am. J. Physiol. 270, R518–R524 (1996).

  101. 101.

    Chen, G. & Reichlin, S. Clearance of [125 I]-tumor necrosis factor-alpha from the brain into the blood after intracerebroventricular injection into rats. Neuroimmunomodulation 5, 261–269 (1998).

  102. 102.

    Chen, G., McCuskey, R. S. & Reichlin, S. Blood interleukin-6 and tumor necrosis factor-alpha elevation after intracerebroventricular injection of Escherichia coli endotoxin in the rat is determined by two opposing factors: peripheral induction by LPS transferred from brain to blood and inhibition of peripheral response by a brain-mediated mechanism. Neuroimmunomodulation 8, 59–69 (2000).

  103. 103.

    Logsdon, A. F., Erickson, M. A., Rhea, E. M., Salameh, T. S. & Banks, W. A. Gut reactions: how the blood-brain barrier connects the microbiome and the brain. Exp. Biol. Med. (Maywood) 243, 159–165 (2018).

  104. 104.

    Erickson, M. A. & Banks, W. A. Neuroimmune axes of the blood-brain barriers and blood-brain interfaces: bases for physiological regulation, disease states, and pharmacological interventions. Pharmacol. Rev. 70, 278–314 (2018).

  105. 105.

    Sampson, T. R. et al. Gut microbioata regulate motor deficits and neuroinflammation in a model of Parkinson’s disease. Cell 167, 1469–1480 (2016).

  106. 106.

    Starr, J. M. et al. Increased blood-brain barrier permeability in type II diabetes demonstrated by gadolinium magnetic resonance imaging. J. Neurol. Neurosurg. Psychiatry 74, 70–76 (2003).

  107. 107.

    Huber, J. D., VanGilder, R. L. & Houser, K. A. Streptozotocin-induced diabetes progressively increases blood-brain barrier permeability in specific brain regions in rats. Am. J. Physiol. 291, H2660–H2668 (2006).

  108. 108.

    Abuhaiba, S. et al. Occipital blood-brain barrier permeability is an independent predictor of visual outcome in type 2 diabetes, irrespective of the retinal barrier: a logitudinal study. J. Neuroendocrinol. 30, e12566 (2018).

  109. 109.

    Huber, J. D. Diabetes, cognitive function, and the blood-brain barrier. Curr. Pharm. Des. 14, 1594–1600 (2008).

  110. 110.

    Takechi, R. et al. Blood-brain barrier dysfunction precedes cognitive decline and neurodegeneration in diabetic insulin resistant mouse model: an implication for causal link. Front. Aging Neurosci. 9, 399 (2017).

  111. 111.

    Hammes, H. P. et al. Pericytes and the pathogenesis of diabetic retinopathy. Diabetes 51, 3107–3112 (2002).

  112. 112.

    Price, T. O., Eranki, V., Banks, W. A., Ercal, N. & Shah, G. N. Topiramate treatment protects blood-brain barrier pericytes from hyperglycemia-induced oxidative damage in diabetic mice. Endocrinology 153, 362–372 (2012).

  113. 113.

    Dore-Duffy, P. Pericytes: pluripotent cells of the blood brain barrier. Curr. Pharm. Des. 14, 1581–1593 (2008).

  114. 114.

    Daneman, R., Zhou, L., Kebede, A. A. & Barres, B. A. Pericytes are required for blood-brain barrier integrity during embryogenesis. Nature 468, 562–566 (2010).

  115. 115.

    Zhang, Y. et al. Involvement of PUMA in pericyte migration induced by methamphetamine. Exp. Cell Res. 356, 28–39 (2017).

  116. 116.

    Sagare, A. P. et al. Pericyte loss influences Alzheimer-like neurodegeneration in mice. Nat. Commun. 4, 2932 (2013).

  117. 117.

    Schulz, G. B. et al. Cerebral cavernous malformation-1 protein controls DLL4-Notch3 signalling between the endothelium and pericytes. Stroke 46, 1337–1343 (2015).

  118. 118.

    Kaiyala, K. J., Prigeon, R. L., Kahn, S. E., Woods, S. C. & Schwartz, M. W. Obesity induced by a high-fat diet is associated with reduced brain insulin transport in dogs. Diabetes 49, 1525–1533 (2000).

  119. 119.

    Mooradian, A. D. Blood-brain barrier choline transport is reduced in diabetic rats. Diabetes 36, 1094–1097 (1987).

  120. 120.

    Hong, H. et al. Downregulation of LRP1 at the blood-brain barrier in streptozotocin-induced diabetic mice. Neuropharmacology 56, 1054–1059 (2009).

  121. 121.

    Banks, W. A., DiPalma, C. R. & Farrell, C. L. Impaired transport of leptin across the blood-brain barrier in obesity. Peptides 20, 1341–1345 (1999).

  122. 122.

    Ouyang, S. et al. Diet-induced obesity suppresses expression of many proteins at the blood-brain barrier. J. Cereb. Blood Flow Metab. 34, 43–51 (2014).

  123. 123.

    Mooradian, A. D. & Smith, T. L. The effect of experimentally induced diabetes mellitus on the lipid order and composition of rat cerebral microvessels. Neurosci. Lett. 145, 145–148 (1992).

  124. 124.

    Beard, R. S. Jr, Reynolds, J. J. & Bearden, S. E. Hyperhomocysteinemia increases permeability of the blood-brain barrier by NMDA receptor-dependent regulation of adherens and tight junctions. Blood 118, 2007–2014 (2011).

  125. 125.

    Rhodehouse, B. C., Mayo, J. N., Beard, R. S. Jr, Chen, C. H. & Bearden, S. E. Opening of the blood-brain barrier before cerebral pathology in mild hyperhomocysteinemia. PLOS ONE 8, e63951 (2013).

  126. 126.

    Delange, F. The disorders induced by iodine deficiency. Thyroid 4, 107–128 (1994).

  127. 127.

    Nunez, B. et al. Cerebral cortex hyperthyroidism of newborn mct8-deficient mice transiently suppressed by lat2 inactivation. PLOS ONE 9, e96915 (2014).

  128. 128.

    Heni, M., Kullmann, S., Preissl, H., Fritsche, A. & Haring, H. U. Impaired insulin action in the human brain: causes and metabolic consequences. Nat. Rev. Endocrinol. 11, 701–711 (2015).

  129. 129.

    De Vivo, D. C. et al. Defective glucose transport across the blood-brain barrier as a cause of persistent hypoglycorrhachia, seizures, and developmental delay. N. Engl. J. Med. 325, 703–709 (1991).

  130. 130.

    Craft, S. et al. Cerebrosinal fluid and plasma insulin levels in Alzheimer’s disease: relationship to severity of dementia and apolipoprotein E genotype. Neurology 50, 164–168 (1998).

  131. 131.

    Talbot, K. et al. Demonstrated brain insulin resistance in Alzheimer’s disease patients is associated with IGF-1 resistance, IRS-1 dyregulation, and cognitive decline. J. Clin. Invest. 122, 1316–1338 (2012).

  132. 132.

    Sartorius, T. et al. The brain response to peripheral insulin declines with age: a contribution of the blood-brain barrier? PLOS ONE 10, e0126804 (2015).

  133. 133.

    Pelleymounter, M. A. et al. Effects of the obese gene product on body weight regulation in ob/ob mice. Science 269, 540–543 (1995).

  134. 134.

    Banks, W. A., Clever, C. M. & Farrell, C. L. Partial saturation and regional variation in the blood to brain transport of leptin in normal weight mice. Am. J. Physiol. 278, E1158–E1165 (2000).

  135. 135.

    Licinio, J. et al. Phenotypic effects of leptin replacement on morbid obesity, diabetes mellitus, hypogonadism, and behavior in leptin-deficient adults. Proc. Natl Acad. Sci. USA 101, 4531–4536 (2004).

  136. 136.

    Ingalls, A. M., Dickie, M. M. & Snell, G. D. Obese, a new mutation in the house mouse. J. Hered. 41, 317–318 (1950).

  137. 137.

    Bribiescas, R. G. Serum leptin levels and anthropometric correlates in Ache Amerindians of Eastern Paraguay. Am. J. Phys. Anthropol. 115, 297–303 (2001).

  138. 138.

    Halaas, J. L. et al. Physiological response to long-term peripheral and central leptin infusion in lean and obese mice. Proc. Natl Acad. Sci. USA 94, 8878–8883 (1997).

  139. 139.

    van Heek, M. et al. Diet-induced obese mice develop peripheral, but not central, resistance to leptin. J. Clin. Invest. 99, 385–390 (1997).

  140. 140.

    El Haschimi, K., Pierroz, D. D., Hileman, S. M., Bjorbaek, C. & Flier, J. S. Two defects contribute to hypothalamic leptin resistance in mice with diet-induced obestiy. J. Clin. Invest. 105, 1827–1832 (2000).

  141. 141.

    Rutenberg, G. W. et al. Body composition in baboons: evaluating a morphometric method. Am. J. Primatol. 12, 275–285 (1987).

  142. 142.

    Bribiescas, R. G. Serum leptin levels in Ache Amerindian females with normal adiposity are not significantly different from American anorexia nervosa patients. Am. J. Hum. Biol. 17, 207–210 (2005).

  143. 143.

    Kirchengast, S. Weight status of adult !Kung San and Kavango people from northern Namibia. Ann. Hum. Biol. 25, 541–551 (1996).

  144. 144.

    Schwartz, M. W. et al. Cerebrospinal fluid leptin levels: relationship to plasma levels and adiposity in humans. Nat. Med. 2, 589–593 (1996).

  145. 145.

    Caro, J. F. et al. Decreased cerebrospinal-fluid/serum leptin ratio in obesity: a possible mechanism for leptin resistance. Lancet 348, 159–161 (1996).

  146. 146.

    Mantzoros, C., Flier, J. S., Lesem, M. D., Brewerton, T. D. & Jimerson, D. C. Cerebrospinal fluid leptin in anorexia nervosa: Correlation with nutritional status and potential role in resistance to weight gain. J. Clin. Endocrinol. Metab. 82, 1845–1851 (1997).

  147. 147.

    Smith, Q. R., Momma, S., Aoyagi, M. & Rapoport, S. I. Kinetics of neutral amino acid transport across the blood-brain barrier. J. Neurochem. 49, 1651–1658 (1987).

  148. 148.

    Fernstrom, J. D. Branched-chain amino acids and brain function. J. Nutr. 135, 1439S–1546S (2005).

  149. 149.

    van Spronsen, F. J., de Groot, M. J., Hoeksma, M., Reijngoud, D.-J. & van Rijn, M. Large neutral amino acids in the treatment of PKU: from theory to practice. J. Inherit. Metab. Dis. 33, 671–676 (2010).

  150. 150.

    Banks, W. A., Burney, B. O. & Robinson, S. M. Effects of triglycerides, obesity, and starvation on ghrelin transport across the blood-brain barrier. Peptides 29, 2061–2065 (2008).

  151. 151.

    Holscher, C. Novel dual GLP-1/GIP receptor agonists show neuroprotective effects in Alzheimer’s and Parkinson’s disease models. Neuropharmacology 136, 251–259 (2018).

  152. 152.

    Gejl, M. et al. Blood-brain glucose transfer in Alzheimer’s disease: effect of GLP-1 analog treatment. Sci. Rep. 7, 17490 (2017).

  153. 153.

    Tamargo, I. A. et al. Novel GLP-1R/GIPR co-agonist “twincretin” is neuroprotective in cell and rodent models of mild traumatic brain injury. Exp. Neurol. 288, 176–186 (2017).

  154. 154.

    Deli, M. A., Abraham, C. R., Kataoka, Y. & Niwa, M. Permeability studies on in vitro blood-brain barrier models: physiology, pathology, and pharmacology. Cell. Mol. Neurobiol. 25, 59–127 (2005).

  155. 155.

    Wolburg, H., Noell, S., Mack, A., Wolburg-Buchholz, K. & Dallier-Becker, P. Brain endothelial cells and the glio-vascular complex. Cell Tissue Res. 335, 75–96 (2009).

  156. 156.

    Saunders, N. R., Dziegielewska, K. M., Mollgard, K. & Habgood, M. D. Physiology and molecular biology of barrier mechanisms in the fetal and neonatal brain. J. Physiol. 596, 5723–5756 (2018).

  157. 157.

    Cornford, E. M., Braun, L. D., Oldendorf, W. H. & Hill, M. A. Comparison of lipid-mediated blood-brain-barrier penetrability in neonates and adults. Am. J. Physiol. 243, C161–C168 (1982).

  158. 158.

    Dickson, P. W., Aldred, A. R., Marley, P. D., Bannister, D. & Schreiber, G. Rat choroid plexus specializes in the synthesis and the secretion of transthyretin (prealbumin). Regulation of tranthyretin synthesis in choroid plexus is indpendent from that in liver. J. Biol. Chem. 15, 3475–3478 (1986).

  159. 159.

    Zhao, F. et al. Effects of passage and cryopreservation on neurotophic factor secretion from choroid plexus epithelial cells. Biomed. Rep. 8, 535–539 (2018).

  160. 160.

    Broadwell, R. D. & Sofroniew, M. V. Serum proteins bypass the blood-brain barrier for extracellular entry to the central nervous system. Exp. Neurol. 120, 245–263 (1993).

Download references

Acknowledgements

Reviewer information

Nature Reviews Endocrinology thanks B. Levin, D. Begley and V. Prevot, and the other anonymous reviewers, for their contribution to the peer review of this work.

Author information

Correspondence to William A. Banks.

Ethics declarations

Competing interests

The author declares no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Glossary

Neurovascular unit

A multicellular and multicomponent network that is composed of neurons, glial cells, brain endothelial cells and extracellular matrix components. The neurovascular unit is key to neurovascular coupling and to the delivery of key nutrients and oxygen from the circulatory system into the brain.

Plasma protein binding

The degree to which a substance binds to proteins within the blood.

Choroid plexus

The choroid plexus, which consists of modified ependymal cells, produces the cerebrospinal fluid in the ventricles of the brain.

Allan–Herndon–Dudley syndrome

A condition resulting from deficient thyroid hormone transport across the blood–brain barrier characterized by cognitive impairment, lack of speech and hypotonia.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Banks, W.A. The blood–brain barrier as an endocrine tissue. Nat Rev Endocrinol 15, 444–455 (2019). https://doi.org/10.1038/s41574-019-0213-7

Download citation

Further reading