Interactions of IGF-1 with the blood-brain barrier in vivo and in situ

Insulin-Like Growth Factor Delivery Across the Blood-Brain Barrier

BACKGROUND

Recent data indicate that insulin-like growth factor-1 (IGF-1) can cross the blood-brain barrier when bound to binding proteins. IGF-1 is very important for early brain development, myelination and survival of cerebellar neurons. It is antiapoptotic and is important for synaptic development. The aim of this study was to identify neurological conditions with low CNS IGF-1.

METHODS

Levels of CSF IGF-1 were determined by radio-immunoassay (Mediagnost, Tubingen, Germany).

RESULTS

Levels of CSF IGF-1 were low in some progressive diseases of white matter and the cerebellum. In some other diseases not involving cerebellum or white matter, CSF IGF-1 was normal.

CONCLUSIONS

Our findings might pave the way for rational therapy in patients with neurodegeneration. Patients with low CNS IGF-1 might be candidates for IGF-1 therapy.

Central and opposing effects of IGF-I and IGF-binding protein-3 on systemic insulin action

IGF-I is recognized as an insulin sensitizer at the liver and muscle, while recent evidence suggests that IGF-binding protein (IGFBP)-3 acts as an insulin antagonist. As there is a paucity of IGF-I receptors in the liver and as the IGF-IGFBP system in the central nervous system is emerging as physiologically relevant, we examined whether the effects of IGF-I and IGFBP-3 on insulin action are mediated through central mechanisms. Intracerebroventricular (ICV) infusion of IGF-I during the insulin clamp (3 mU x kg(-1) x min(-1)) resulted in significant improvement in hepatic insulin action (50%, P < 0.05). In contrast, ICV infusion of IGFBP-3 significantly impaired insulin action at the liver (45% increase in hepatic glucose production, P < 0.01). While IGF-I marginally increased peripheral glucose uptake, IGFBP-3 significantly decreased peripheral glucose uptake (approximately 30%, P < 0.01). As the nuclear localization signal mutant IGFBP-3, which has a normal affinity to IGFs but binds other IGFBP-3 partners poorly and fails to normally internalize, has reduced central activity on metabolism, we conclude that the effects of IGFBP-3 on the hypothalamus involve activity mediated by interfacing with other molecules in addition to IGFs. Marked, opposing, and independent physiological effects of IGF-I and IGFBP-3 through central mechanisms may have implications on potential strategies in specific modulation of peripheral insulin action.

Growth hormone and IGF-I modulate local cerebral glucose utilization and ATP levels in a model of adult-onset growth hormone deficiency

Decreases in plasma IGF-I levels that occur with age have been hypothesized to contribute to the genesis of brain aging. However, support for this hypothesis would be strengthened by evidence that growth hormone (GH)/IGF-I deficiency in young animals produces a phenotype similar to that found in aged animals. As a result, we developed a unique model of adult-onset GH/IGF-I deficiency by using dwarf rats specifically deficient in GH and IGF-I. The deficiency in plasma IGF-I is similar to that observed with age (e.g., 50% decrease), and replacement of GH restores levels of IGF-I to that found in young animals with normal GH levels. The present study employs this model to investigate the effects of circulating GH and IGF-I on local cerebral glucose utilization (LCGU). Analysis of LCGU indicated that GH/IGF-I-deficient animals exhibit a 29% decrease in glucose metabolism in many brain regions, especially those involved in hippocampally dependent processes of learning and memory. Similarly, a high correlation between plasma IGF-I levels and glucose metabolism was found in these areas. The deficiency in LCGU was not associated with alterations in GLUT1, GLUT3, or hexokinase activity. A 15% decrease in ATP levels was also found in hippocampus of GH-deficient animals, providing compelling data that circulating GH and IGF-I have significant effects on the regulation of glucose utilization and energy metabolism in the brain. Furthermore, our results provide important data to support the conclusion that deficiencies in circulating GH/IGF-I contribute to the genesis of brain aging.

Reciprocal interactions of insulin and insulin-like growth factor I in receptor-mediated transport across the blood-brain barrier

Although the blood-brain barrier limits free passage of peptides and proteins from the peripheral circulation to the central nervous system, specific transport systems for insulin and IGF-I have been identified. To further determine whether insulin and IGF-I share the same transport system, and if not, whether the two transport systems interact with each other, we performed multiple-time regression analysis in mice after iv injection and in situ brain perfusion of these peptides. Insulin and IGF-I caused reciprocal inhibition of each other's transport, although the effect of insulin was detected only by the in situ brain perfusion system. The interaction took place mainly at the step of cell surface binding as seen in cultured rat brain endothelium 4 brain microvessel endothelial cells. Further studies in 3T3 cells stably overexpressing the insulin receptor showed that the sharing of the transport systems was only partial. We conclude that insulin and IGF-I are mainly transported by their own transport systems, but a small amount can enter the brain by their "noncognate" transporters. The redundancy of their transport systems illustrates the regulatory function of the blood-brain barrier and reflects the importance of blood-borne insulin and IGF-I in the central nervous system.

Differential BBB interactions of three ingestive peptides: obestatin, ghrelin, and adiponectin

Endogenous compounds, including ingestive peptides, can interact with the blood-brain barrier (BBB) in different ways. Here we used in vivo and in vitro techniques to examine the BBB permeation of the newly described satiety peptide obestatin. The fate of obestatin in blood and at the BBB was contrasted with that of adiponectin. By the sensitive multiple time-regression method, obestatin appeared to have an extremely fast influx rate to the brain whereas adiponectin did not cross the BBB. HPLC analysis, however, showed the obestatin result to be spurious, reflecting rapid degradation. Absence of BBB permeation by obestatin and adiponectin was in contrast to the saturable transport of human ghrelin reported previously. As a positive control, ghrelin showed saturable binding and endocytosis in RBE4 cerebral microvessel endothelial cells. By comparison, obestatin lacked specific binding and endocytosis, and the small amount internalized showed rapid intracellular degradation before the radioactivity was released by exocytosis. The differential interactions of obestatin, adiponectin, and ghrelin with the BBB illustrate their distinctive physiological interactions with the CNS.

Receptor-mediated transport of LIF across blood-spinal cord barrier is upregulated after spinal cord injury

Leukemia inhibitory factor (LIF) crosses the normal blood-brain and blood-spinal cord barrier (BBB) by a saturable transport system [Pan, W., Kastin, A.J., Brennan, J.M., 2000. Saturable entry of leukemia inhibitory factor from blood to the central nervous system. J. Neuroimmunol. 106, 172-180]. Since LIF is a cytokine beneficial to spinal cord regeneration, understanding the regulation of its transport across the injured BBB may help in the design of strategies for the treatment of spinal cord injury (SCI). In this study, we initially showed that transport of LIF is mediated by its specific receptor LIFRalpha (gp190), using both adult mice and monolayers of mouse brain microvessel endothelial cells. Permeation of radioactively labeled LIF was inhibited not only by excess unlabeled LIF, but also by a blocking antibody to the extracellular domain of gp190 LIFRalpha receptor. This showed that the saturable transport of LIF across the BBB involves LIFRalpha. We then tested the hypothesis that this transport system can be upregulated after SCI. SCI was generated by an established compression method at the upper lumbar level. Transport was studied 1 week after SCI, a time of tissue repair following ischemia and inflammation. Spinal cord uptake of 99mTc-albumin 10 min after intravenous injection was used as an indicator of paracellular permeability of the BBB, its small but significant increase at the injury site indicating the level of persistent BBB disruption. The uptake of 125I-LIF by the injured lumbar spinal cord was significantly greater than that in the uninjured controls as well as that of 99mTc-albumin. Both excess unlabeled LIF and the blocking antibody against LIFRalpha significantly suppressed the increased entry of 125I-LIF without affecting that of 99mTc-albumin. Thus, the increased blood-to-spinal cord permeation of LIF was not solely explained by barrier disruption but involved LIFRalpha. This enhanced transport correlated with increased expression of LIFRalpha shown by immunofluorescent staining and Western blot. Therefore, LIFR at the BBB provides an important target for therapeutic intervention.

Age-related effects of IGF-1 on the NMDA-, GH- and IGF-1-receptor mRNA transcripts in the rat hippocampus

Insulin-like growth factor-1 (IGF-1) and growth hormone (GH) have been suggested to promote memory and cognitive capabilities. In a recent publication we observed that GH increase the proportion of the NR2B subunit mRNA transcript of the NMDA receptor in rat hippocampus. NR2B has been suggested to be essential for spatial learning and long-term potentiation (LTP). This effect of GH might be IGF-1-mediated or a result of a co-ordination with IGF-1. To test this hypothesis further, we examined the effects of 10 daily s.c. injections of IGF-1 on NMDA receptor subunits (NR1, NR2A, and NR2B), GH receptor (GHR), GH binding protein (GHBP) and type 1 IGF receptor (IGF-1R) gene transcripts in the hippocampus. The NR2B subunit mRNA increased in young (11 weeks) but not in older (14-16 months) rats and the expression of the NR2A mRNA was decreased in both groups. The ratio of NR2B to NR2A is suggested to mirror the potential for synaptic plasticity. In both age groups, IGF-1 treatment resulted in a significant increase of this ratio at transcription level. The GHR mRNA increased in young rats, mimicking the effect of GH, while the IGF-1R mRNA was decreased in the older group of rats after IGF-1 treatment. These results suggest that IGF-1 in many aspects may mediate the actions earlier shown for GH.

Why study transport of peptides and proteins at the neurovascular interface

The blood-brain barrier (BBB) is an immense neurovascular interface. In neurodegenerative, ischemic, and traumatic disorders of the central nervous system (CNS), the BBB may hinder the delivery of many therapeutic peptides and proteins to the brain and spinal cord. Fortunately, the mistaken dogma that peptides and proteins do not cross the BBB has been corrected during the past two decades by the accumulating evidence that peptides and proteins in the periphery exert potent effects in the CNS. Not only can peptides and proteins serve as carriers for selective therapeutic agents, but they themselves may directly cross the BBB after delivery into the bloodstream. Their passage may be mediated by simple diffusion or specific transport, both of which can be affected by interactions in the blood compartment (outside the BBB) and within the endothelial cells (at the BBB level). Although the majority of current delivery strategies focuses on modification of the molecule to be delivered, understanding the mechanisms of transport will eventually facilitate regulation of the BBB directly. We review the different aspects of interactions and discuss recent advances in the cell biology of peptide/protein transport across the BBB. Better understanding of the nature and regulation of the transport systems at the BBB will provide a new direction to enhance the interactions of peripheral peptides and proteins with the CNS.

Efficient transfer of receptor-associated protein (RAP) across the blood-brain barrier

We have sought to identify a high-capacity transport system that mediates transcytosis of proteins from the blood to the brain. The 39 kDa receptor-associated protein (RAP) functions as a specialized endoplasmic reticulum chaperone assisting in the folding and trafficking of members of the low-density lipoprotein (LDL) receptor family. RAP efficiently binds to these receptors and antagonizes binding of other ligands. Previous studies have shown that two large members of the LDL receptor family, LDL receptor-related protein 1 (LRP1) and LDL receptor-related protein 2 (LRP2 or megalin), possess the ability to mediate transcytosis of ligands across the brain capillary endothelium. Here, we tested whether blood-borne RAP crosses the blood-brain barrier (BBB) by LRP1- or megalin-mediated transport by studying the pharmacokinetics of [125I]-RAP transport into the brain in intact mice and across cell monolayers in vitro. Our results show that [125I]-RAP is relatively stable in blood for 30 minutes and has a mean influx constant of 0.62±0.08 μl/g-minute from blood to brain. In situ brain perfusion in blood-free buffer shows that transport of [125I]-RAP across the BBB is a saturable process. Capillary depletion of brain homogenates indicates that 70% of [125I]-RAP is localized in the parenchyma rather than in the vasculature of the brain. Results of transport in stably transfected MDCK cells are consistent with the hypothesis that megalin mediates most of the apical-to-basolateral transport across polarized epithelial cells. The inhibition of [125I]-RAP influx by excess RAP and the involvement of megalin indicate the presence of a saturable transport system at the BBB. The higher permeability of RAP compared with that of melanotransferrin and transferrin show that the LRP receptor is a high capacity transport system. These studies suggest that RAP may provide a novel means of protein-based drug delivery to the brain.

Insulin-like growth factor 1 as a predictor of ischemic stroke outcome in the elderly

PURPOSE

To examine whether serum insulin-like growth factor 1 (IGF-1) and IGF binding protein 3 (IGFBP-3) concentrations, determined early after the onset of stroke, are predictive of clinical outcome in elderly patients.

METHODS

The sample comprised 85 patients (mean [+/- SD] age, 83 +/- 7.4 years; range, 67 to 99 years; 34% male) who were admitted with acute stroke to a geriatric ward between January 1998 and June 2000, and 88 control patients who were similar in age and sex. Clinical and laboratory assessments, computed tomographic scan of the head, carotid ultrasonography, and electrocardiography were employed to define the clinical and etiologic stroke subtype. Fasting blood samples were collected within 24 hours of admission for IGF-I and IGFBP-3 measurement. Univariate and multiple logistic regression analyses, with adjustment for other related clinical covariates, were used to assess the relation of IGF-I and IGFBP-3 to poor outcome, defined as severe disability (Barthel index <60/100) or death, at 1 month (or at discharge), 3 months, and 6 months.

RESULTS

Mean (+/- SD) IGF-1 levels were lower in patients with stroke than in controls (69 +/- 45 ng/mL vs. 102 +/- 67 ng/mL, P adjusted for age = 0.001). The mean IGF-1/IGFBP-3 molar ratio was also lower in stroke patients (0.12 +/- 0.07 vs. 0.19 +/- 0.09, P adjusted for age <0.0001). However, there was no relation of hormone levels to either the clinical subtype of stroke or the extent of neurologic impairment. IGF-1 levels were inversely related to poor outcome (mainly death) at 3 and 6 months, independent of other clinical covariates that were highly predictive of outcome, such as age and stroke scale score on admission (hazard ratio for death at 6 months for each 20-ng/mL increase = 0.7; 95% confidence interval: 0.5 to 0.9). An independent association of the molar ratio with death at 3 and 6 months was also found.

CONCLUSION

Low levels of circulating IGF-1 may predict the clinical outcome of stroke in elderly patients.

Neuregulin-1-β1 enters brain and spinal cord by receptor-mediated transport

Proteins of the neuregulin (NRG) family play important regulatory roles in neuronal survival and synaptic activity. NRG-1-beta1 has particular potential as a therapeutic agent because it enhances myelination of neurites in spinal cord explants. In this study, we determined the permeation of NRG-1-beta1 across the blood-brain and blood-spinal cord barriers (BBB and BSCB respectively). Intact radioactively labeled NRG-1-beta1 had a saturable and relatively rapid influx rate from blood to the CNS in mice. Capillary depletion studies showed that NRG-1-beta1 entered the parenchyma of the brain and spinal cord rather than being trapped in the capillaries that compose the BBB. The possible mechanism of receptor-mediated transport was shown by the ability of antibodies to erbB3 and erbB4 receptors to inhibit the influx. Lipophilicity, less important for such saturable transport mechanisms, was measured by the octanol : buffer partition coefficient and found to be low. The results indicate that NRG-1-beta1 enters spinal cord and brain by a saturable receptor-mediated mechanism, which provides the opportunity for possible therapeutic manipulation at the BBB level.

GDNF and IGF-I trophic factors delay hereditary Purkinje cell degeneration and the progression of gait ataxia

Neurotrophic factors GDNF and/or IGF-I were chronically infused into shaker mutant rats to rescue cerebellar Purkinje neurons from adult-onset heredodegeneration. The natural expression of the shaker mutation is characterized by spatially restricted degeneration of Purkinje cells that occurs earlier and faster in an anterior vermal compartment and slightly later and more slowly in a posterior vermal compartment. Gait ataxia and whole body tremor develop concomitant with the degeneration of Purkinje neurons. The number and spatial distribution of surviving Purkinje neurons, identified by cell-specific calbindin immunoreactivity, were quantitatively analyzed in mid-sagittal sections and correlated with quantitative movement analysis of hindlimb gait patterns. Compared to the number of surviving Purkinje cells in age-matched, non-infused, or saline-infused control mutants, 4 weeks of infusion of GDNF or IGF-I rescued many anterior compartment Purkinje cells from early degeneration. However, 2 and 4 weeks after cessation of GDNF or IGF-I infusion, respectively, the number and spatial distribution of surviving Purkinje cells was comparable to that observed in age-matched controls. Eight weeks of infusion of trophic factors did not support the continued survival of most anterior compartment Purkinje cells and was partially, and probably only transiently, neuroprotective for some posterior compartment Purkinje cells. When GDNF and IGF-I were infused together for 4 weeks the number of surviving Purkinje cells was additively greater than with either factor alone. Behaviorally, 4 weeks of infusion of trophic factors delayed the development of gait ataxia. Infused GDNF appeared to preserve hip stability, whereas IGF-I stabilized step length. Tremor was attenuated with 8 weeks of infusion of GDNF or IGF-I. GDNF-infused animals showed low power tremor frequencies, whereas IGF-I infusion resulted in a single large power peak with decreased numbers of low-amplitude frequencies. Collectively these findings indicate that exogenous trophic factors can delay the onset of hereditary Purkinje cell degeneration and gait ataxia. Quite surprisingly, GDNF and IGF-I appeared to act on disparate populations of mutant Purkinje cells, whose differential survival affected different aspects of locomotion.

Activation of hypoxia-inducible factor-1 in the rat cerebral cortex after transient global ischemia: potential role of insulin-like growth factor-1

Hypoxia-inducible factor-1 (HIF-1) is a transcription factor that regulates the adaptive response to hypoxia in mammalian cells. It consists of a regulatory subunit HIF-1alpha, which accumulates under hypoxic conditions, and a constitutively expressed subunit HIF-1beta. In this study we analyzed HIF-1alpha expression in the rat cerebral cortex after transient global ischemia induced by cardiac arrest and resuscitation. Our results showed that HIF-1alpha accumulates as early as 1 hr of recovery and persists for at least 7 d. In addition, the expression of HIF-1 target genes, erythropoietin and Glut-1, were induced at 12 hr to 7d of recovery. A logical explanation for HIF-1alpha accumulation might be that the brain remained hypoxic for prolonged periods after resuscitation. By using the hypoxic marker 2-(2-nitroimidazole-1[H]-y1)-N-(2,2,3,3,3-pentafluoropropyl)-acetamide (EF5), we showed that the brain is hypoxic during the first hours of recovery from cardiac arrest, but the tissue is no longer hypoxic at 2 d. Thus, the initial ischemic episode must have activated other nonhypoxic mechanisms that maintain prolonged HIF-1alpha accumulation. One such mechanism might be initiated by insulin-like growth factor-1 (IGF-1). Our results showed that IGF-1 expression was upregulated after cardiac arrest and resuscitation. In addition, we showed that IGF-1 was able to induce HIF-1alpha in pheochromocytoma cells and cultured neurons as well as in the brain of rats that received intracerebroventricular and systemic IGF-1 infusion. Moreover, infusion of a selective IGF-1 receptor antagonist abrogates HIF-1alpha accumulation after cardiac arrest and resuscitation. Our study suggest that activation of HIF-1 might be part of the mechanism by which IGF-1 promotes cell survival after cerebral ischemia.

Caloric restricted male rats demonstrate fewer synapses in layer 2 of sensorimotor cortex

Previous studies have demonstrated an age-related decline in the density of presumptive inhibitory synapses in layer 2 of rat sensorimotor cortex [J. Comp. Neurol. 439(1) (2001) 65]. Caloric restriction has been shown to ameliorate age-related deterioration in a variety of systems and to extend life span. The present study tested the hypothesis that caloric restriction would prevent the previously reported age-related synaptic decline. Accordingly, synaptic density in layer 2 of sensorimotor cortex was compared between 29-month-old male rats fed ad libitum and 29-month-old male rats that were caloric restricted (60% of ad libitum calories) from 4 months of age. In serial electron micrographs, the physical disector was used to determine the numerical density of presumptive excitatory and inhibitory synapses (those containing round or nonround vesicles, respectively) as well as that of neurons. Not only was the previously reported age-related decline in numerical density of presumptive inhibitory synapses not ameliorated by caloric restriction, the numerical density was significantly lower in caloric restricted than in ad libitum fed rats for total as well as for presumptive excitatory and inhibitory synapses. Analysis further revealed no difference in the numerical density of neurons in this region. Relating synapse density to neuron density as the ratio of synapses to neuron also demonstrated significantly fewer synapses per neuron in caloric restricted than in ad libitum fed old rats. Finally, synapse length was significantly less in caloric restricted rats. These results suggest that not only does caloric restriction fail to prevent the age-related decline in presumptive inhibitory synapses, it results in fewer presumptive excitatory synapses as well.

Intracerebroventricular infusion of insulin-like growth factor-I ameliorates the age-related decline in hippocampal neurogenesis

The dentate gyrus of the hippocampus is one of few regions in the adult mammalian brain characterized by ongoing neurogenesis. Significantly, recent studies indicate that the rate of neurogenesis in the hippocampus declines with age, perhaps contributing to age-related cognitive changes. Although a variety of factors may influence the addition of new neurons in the adult dentate gyrus, the mechanisms responsible for the age-related reduction remain to be established. Insulin-like growth factor-I (IGF-I) is one promising candidate to regulate neurogenesis in the adult and aging brain since it influences neuronal production during development and since, like the rate of neurogenesis, it decreases with age. In the current study, we used bromodeoxyuridine labeling and multilabel immunofluorescence to assess age-related changes in neuronal production in the dentate gyrus of adult Brown Norway x Fischer 344 rats. In addition, we investigated the relationship between changes in neurogenesis and the age-dependent reduction in IGF-I by evaluating the effect of i.c.v. infusion of IGF-I on neurogenesis in the senescent dentate gyrus. The analyses revealed an age-dependent reduction in the number of newly generated cells in the adult dentate subgranular proliferative zone and, in addition, a 60% reduction in the differentiation of newborn cells into neurons. Restoration of IGF-I levels in senescent rats significantly restored neurogenesis through an approximately three-fold increase in neuronal production. The results of this study suggest that IGF-I may be an important regulator of neurogenesis in the adult and aging hippocampus and that an age-related decline in IGF-I-dependent neurogenesis could contribute to age-related cognitive changes.