SEARCH

SEARCH BY CITATION

Keywords:

  • blood–brain barrier;
  • peptide;
  • leptin;
  • insulin;
  • secretions

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Insulin and the BBB
  5. Leptin and the BBB
  6. BBB Secretions
  7. Summary
  8. References

The blood–brain barrier (BBB) plays a critical role in the transduction of signals between the central nervous system and peripheral tissues. It does so through several mechanisms, including the direct transport of peptides and regulatory proteins such as insulin and leptin. Another mechanism that may be important is the secretion by brain endothelial cells of substances that affect feeding, such as proinflammatory cytokines and NO. We have recently shown that the BBB is capable of receiving an input from one side and secreting a substance into the other. Additionally, BBB secretions can be modulated by substances that affect feeding, such as adiponectin and lipopolysaccharide.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Insulin and the BBB
  5. Leptin and the BBB
  6. BBB Secretions
  7. Summary
  8. References

Feeding is one of the most integrated of all behaviors. The decision to feed requires integration of signals external to the animal with internal signals. Many of the internal signals arise from the peripheral tissues. These internal signals must be transmitted to the brain for integration. Broadly, the mechanisms for transmission are either neuronal-based or endocrine-based. The latter involves secretion of a substance into the blood that interacts with the central nervous system (CNS)1 through one of several mechanisms. Most of these endocrine-based mechanisms involve the blood–brain barrier (BBB) in one form or another. The most direct mechanism is for the substance secreted into the blood to cross the BBB and to directly interact with CNS receptors (Figure 1). A less direct and much less studied mechanism is for the circulating substance to induce the BBB to secrete a substance into the brain that affects CNS function.

image

Figure 1. Mechanisms of BBB communication. Two of the mechanisms by which the BBB can transduce signals between the CNS and peripheral tissues are shown here. Peptides and regulatory proteins can be transported across the BBB. Leptin and insulin, the two substances discussed here, have saturable transport only in the blood-to-brain direction, but other substances can be transported in the brain-to-blood direction or bidirectionally. The cells that comprise the BBB can also secrete substances that affect feeding. Such secretions can be stimulated or inhibited as discussed in the text. Shown here is one variation on the theme of secretion: a stimulatory input from the blood inducing a brain endothelial cell to secrete a substance into the brain.

Download figure to PowerPoint

Here, we will review two substances, insulin and leptin, shown to directly cross the BBB. These substances have been extensively studied with regard to the interactions with the BBB and the roles of those interactions with feeding and obesity. We will also review the literature on the abilities of the brain endothelial cells that comprise the BBB to secrete substances that affect feeding.

Insulin and the BBB

  1. Top of page
  2. Abstract
  3. Introduction
  4. Insulin and the BBB
  5. Leptin and the BBB
  6. BBB Secretions
  7. Summary
  8. References

Studies investigating whether insulin can cross the BBB date back to the 1950s (1, 2). These and other studies conducted well into the 1980s concluded that insulin could not cross the BBB. The earlier studies concluded that there was no insulin in the CNS or the cerebrospinal fluid (CSF). The later studies, conducted after it was clear that insulin was present in the CNS and CSF, concluded that insulin was produced by the CNS, and, therefore, CNS insulin was of central, not peripheral, origin. These studies were countered by those from a few other workers, primarily from Woods et al. and Porter et al. (2, 3), which showed a distinct non-linear relation between CSF and plasma levels of insulin. As it became increasingly clear that the CNS produced little or no insulin, it also became clear that the BBB was transporting insulin.

We used radioactively labeled insulin to confirm both that insulin can cross the BBB and that it does so by a saturable transporter. This transporter is not uniformly distributed throughout the BBB. For example, the BBB located within the midbrain and thalamus does not transport measurable amounts of insulin in the ICR mouse, whereas the olfactory bulb transports insulin three to seven times faster than the average for whole brain (4).

Immunoactive studies have confirmed several features of insulin transport (5). In these studies, human insulin was infused peripherally into mice by Alzet pump until steady state was reached. Species specific immunoassays were used to distinguish human from mouse insulin in both brain and serum. These studies showed a non-linear relation between serum and brain levels of human insulin. Thus, these studies reproduced the original immunoactive results used to argue that CNS insulin was derived from the periphery and that such derivation was by a saturable mechanism. Unlike those original studies, however, the CNS insulin could not be endogenously derived from brain, because the mouse brain does not make human insulin. These studies also found an inverse relation within the brain between levels of mouse insulin and human insulin. This shows that peripherally derived CNS insulin has a direct effect on endogenous brain insulin levels. The most parsimonious explanation for this is that the endogenous brain insulin is also peripherally derived, so that as exogenous insulin replaces endogenous insulin in the serum, CNS insulin follows suit. Finally, we found that brain/insulin ratios were decreased at levels that had little or no effect on serum glucose (6). This shows that insulin signaling across the BBB occurs at euglycemic levels and has little or no involvement in the hypoglycemic response.

Insulin plays many roles within the CNS. Several laboratories have shown that some of the CNS effects of insulin are the opposite of those effects mediated through peripheral tissues. In particular, CNS insulin increases glucose and inhibits feeding, whereas serum insulin decreases glucose and increases feeding (2, 7). Thus, to some extent, insulin acts as its own counterregulatory hormone, with CNS insulin producing features of insulin resistance (1).

Many of the BBB transporters for peptides and regulatory proteins are themselves regulated. This also seems to be true for insulin. For example, insulin transport is increased in mice with insulinopenic diabetes and decreased in obese and hibernating animals. Insulin transport across the BBB is greatly increased (8) in mice treated with lipopolysaccharide (LPS). LPS is derived from the bacterial walls of gram-negative bacteria and induces profound immune responses, especially related to the release of proinflammatory cytokines (9). An enhanced transport of insulin across the BBB would intensify the CNS actions of insulin, including the insulin resistance–like effect. We have suggested, therefore, that enhanced transport of insulin as induced by LPS may be a model for the study of insulin in proinflammatory states such as sepsis.

Leptin and the BBB

  1. Top of page
  2. Abstract
  3. Introduction
  4. Insulin and the BBB
  5. Leptin and the BBB
  6. BBB Secretions
  7. Summary
  8. References

Leptin is transported widely throughout the CNS, but the most intensive region of uptake is at the arcuate nucleus (10). A barrier comprised of tanycytes and ependymal cells separates the median eminence from the arcuate nucleus and from the CSF of the fourth ventricle, respectively (11, 12). Thus, leptin depends on saturable transport across the vascular barrier and, probably to a lesser extent, across the choroid plexus to reach the arcuate nucleus. This saturable transporter often fails in diet-induced obesity before resistance occurs at the receptor level, thus producing so-called peripheral resistance (13).

The leptin transporter is also not static but modulated in several conditions and by a selected number of substances. For example, epinephrine increases the transport of leptin across the BBB by 2- to 3-fold (14). This is complementary to other known interactions between epinephrine and leptin. In the periphery, leptin stimulates epinephrine release, whereas epinephrine inhibits leptin release and suppresses appetite. Within the brain, leptin inhibits epinephrine release, whereas epinephrine stimulates appetite. Thus, by stimulating leptin transport across the BBB, peripheral epinephrine would facilitate leptin-induced suppression of CNS epinephrine release and so intensify its anorectic action.

The rate at which leptin is transported across the BBB is decreased in obesity and with starvation but increased with short-term fasting. All of these effects can be explained, at least partially, by the ability of triglycerides to inhibit leptin transport across the BBB.

The ability of triglycerides to inhibit leptin transport may explain the mystery of why resistance and decreased transport across the BBB (15) would develop in obesity to an anorectic like leptin. It seems paradoxical that a mechanism would evolve that would block the anorectic signal to brain during obesity, whereas it is logical that such a mechanism would evolve in starvation. There can be little doubt that, in the course of evolution, animals are more often faced with a threat to their survival because of starvation than because of an excess of calories. Therefore, it may be that triglyceride-induced inhibition of leptin transport evolved because hypertriglyceridemia is a signal to the brain of starvation and not a signal of obesity.

BBB Secretions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Insulin and the BBB
  5. Leptin and the BBB
  6. BBB Secretions
  7. Summary
  8. References

A few papers have clearly documented that the cells that comprise the BBB are themselves capable of secreting substances either spontaneously or with stimulation (16, 17). Many of these substances, including NO and the proinflammatory cytokines, have effects on feeding. The possibility exits, therefore, that substances originating from brain endothelial cells may be contributing to either the blood or CNS pools of these substances. The unique position of the BBB, which has one side of the cell facing into the blood and the other side facing into the CNS, means that it can secrete substances into the circulation, into the CNS, or both. Furthermore, it could be that the BBB could receive stimulation from one compartment (e.g., from the blood), but respond by secreting a substance into the other compartment (e.g., the CNS). Thus, secretions from the BBB could not only influence peripheral and CNS events, but could act to coordinate them.

We have recently provided evidence for such functions of the BBB (18). An in vitro monolayer model, which has been used to study the vascular BBB for >20 years, is ideal to address this area. Brain endothelial cells grown in this model polarize so that the brain (or abluminal) side of the endothelial cell orients against a collagen matrix and the blood (luminal) side orients away from the matrix. Luminal and abluminal chambers facilitate sampling of brain endothelial cell secretions from its respective sides. We used protein microarrays to inventory the secretions of 10 cytokines. The brain endothelial cells spontaneously released large amounts of interleukin-6 (IL-6) and, in most experiments, lower amounts of IL-1, IL-10, tumor necrosis factor, and granulocyte-macrophage colony-stimulating factor. Secretion into the luminal chamber was favored over secretion into the abluminal chamber, typically by ∼10-fold. LPS greatly increased the release of IL-6, IL-10, granulocyte-macrophage colony-stimulating factor, and tumor necrosis factor from brain endothelial cells. For IL-6, secretion into the luminal chamber was much more robust when the LPS was applied to the abluminal rather than the luminal chamber. This latter finding supports the general hypothesis that the BBB can receive input from one of its sides that influences its secretions from its other side.

Work with adiponectin represents another example of secretions from the BBB and how they may affect responses to feeding (19). Adiponectin applied either peripherally or centrally has effects on feeding. However, adiponectin seems to cross the BBB very poorly (20). In contrast, adiponectin produced a statistically significant decrease in the release of IL-6 from an immortalized cell line of rat brain endothelial cells.

Summary

  1. Top of page
  2. Abstract
  3. Introduction
  4. Insulin and the BBB
  5. Leptin and the BBB
  6. BBB Secretions
  7. Summary
  8. References

The BBB plays a critical role in the transduction of signals between the CNS and peripheral tissues. It does so through several mechanisms, including the direct transport of peptides and regulatory proteins such as insulin and leptin. Another mechanism that may be important is the secretion by brain endothelial cells of substances that affect feeding, such as proinflammatory cytokines and NO. We have recently shown that the BBB is capable of receiving an input from one side and secreting a substance into the other. Additionally, BBB secretions can be modulated by substances that affect feeding, such as adiponectin and LPS.

Footnotes
  • 1

    Nonstandard abbreviations: CNS, central nervous system; BBB, blood–brain barrier; CSF, cerebrospinal fluid; LPS, lipopolysaccharide; IL, interleukin.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Insulin and the BBB
  5. Leptin and the BBB
  6. BBB Secretions
  7. Summary
  8. References