No abstract

Insulin concentrations in acid/ethanol extracts of the whole rat brain were on the average 25 times higher than plasma insulin levels. Brain insulin was indistinguishable from authentic pancreatic insulin, based on its behavior in radioimmunoassay, radioreceptor assay, and bioassay and its chromatographic pattern on Sephadex G-50 column chromatography. Insulin was found in all regions of the brain examined, but distribution was uneven. Some regions had insulin concentrations as much as 100 times higher than in plasma; levels at least 10 times higher were found in other regions. Recently, we have found that insulin receptors are widely distributed in the central nervous system of the rat (9). Therefore, we were interested to see if insulin itself was present in the central nervous system. In the present study, we have found that insulin concentrations in brain far exceeded insulin levels in blood. Insulin concentrations in extracts of whole brain were, on the average, 25 times higher than in plasma. Different discrete regions of the central nervous system varied in their insulin content, and concentrations between 10 and 100 times higher than plasma were observed. Insulin in brain tissue was determined by a radioimmunoassay and further characterized by radioreceptor assay, bioassay, and chromatography. Concordance of all these methods demonstrated that genuine insulin or an extremely similar peptide was present in the central nervous system. MATERIALS AND METHODS Extraction Procedure. A standard type of extraction procedure was used (10). Male Sprague-Dawley rats (250-300 g) fed ab lib were killed by decapitation at 1600; the brain (including the pituitary) was removed and kept on ice during the whole dissection procedure. Whole brain and macroscopically dissected discrete regions were homogenized (Polytron) in 10 vol of ice-cold 0.2 M HCI/75% ethanol (acid-ethanol).The suspension was stirred or shaken overnight at 40C. After centrifugation at 3500 rpm for 20 min at 40C, the supernatantsThe publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. 125I-insulin (130-180 ,gCi/ug), and porcine insulin standard (Eli Lilly). The insulin content reported refers to the pork standard. No immunoreactivity could be detected in a buffer containing the same amount of bovine serum albumin as in the reconstituted ex-

A B S T R A C T In view of the potent influences of the central nervous system on glucose metabolism and on its hormonal regulators, and our recent finding of insulin and insulin receptors throughout the central nervous systsem, we have examined extreme conditions ofhyperinsulinemia (obese mice) and hypoinsulinemia (streptozotocin-treated rats) with respect to changes in brain insulin and receptor content. Sprague-Dawley rats given streptozotocin (100 mg/kg body wt) developed severe diabetes and by 48 h showed no change in brain insulin. Rats given 65 mg/kg streptozotocin also had severe diabetes, but survived longer. Both at 7 d and at 30 d after streptozotocin treatment there was no significant change in brain insulin or in brain content of insulin receptors, despite the fact that peripheral hepatic receptors were elevated and pancreatic insulin was markedly depleted.The obese mice were studied at 8-10 wk when peripheral plasma insulin concentrations were 50-fold elevated and receptors on peripheral target cells were reduced to -40-50% of normal; brain insulin concentrations and receptor content were indistinguishable from those of thin littermates. Thus, brain insulin, which is typically 10 times higher than plasma insulin concentrations, and brain receptor content, which is equivalent to receptor content on peripheral tissues, appears to be regulated entirely independently of hormone and receptor in the periphery. These findings are consistent with the hypothesis that insulin in the central nervous system is synthesized by the neural elements, and plays a role in the central nervous system which is unrelated to peripheral glucose metabolism.

Insulin has been detected, at levels higher than those in plasma, in a broad range of extrapancreatic tissues in both rats and humans. Rat liver insulin was shown to be indistinguishable from genuine insulin by radioimmunoassay, Sephadex chromatography, bioassay, and antibody neutralization. Liver insulin (like brain insulin) was unchanged in ob/ob mice, in rats treated with streptozotocin, or in fasted rats, despite marked alterations in pancreatic secretion of insulin and in liver content of insulin receptors. Insulin was found in cultured human IM-9 lymphocytes and cultured fibroblasts at concentrations greater than 100 times the levels in the media. IM-9 lymphocyte insulin also was shown to be indistinguishable from genuine insulin, by the same criteria used for liver insulin. The insulin concentration in cultured human cells was unaffected by depletion of insulin from the culture medium or by addition of beef insulin to the medium. The data suggest that a part, if not all, of the extrapancreatic tissue insulin is independent of plasma insulin and may be synthesized by the tissues themselves.Recently, we demonstrated the presence of insulint in the brain in concentrations higher than in plasma (1). This brain insulin was indistinguishable from genuine insulin by multiple criteria, and its tissue concentration did not vary at all with extreme changes in plasma insulin (2). In the present study we show that insulin immunoreactivity is present in essentially all tissues of humans and rats as well as in cultured lymphocytes and fibroblasts at concentrations that are 2-100 times those present in the plasma or culture medium. The material is similar to (or identical with) genuine insulin by radioimmunoassay, gel filtration, bioassay, and antibody neutralization. The concentration of cellular insulin in vio changes little or not at all in response to extremes of hyperinsulinemia and hypoinsulinemia. MATERIALS AND METHODSTissue Preparations. Male Sprague-Dawley rats (250-300 g) fed ad lib were decapitated between 1400 and 1600 hr; "fasted rats" had been denied food for 50-72 hr before they were killed. Rats injected with streptozotocin (65 mg/kg intravenously) and control rats (injected with isotonic saline) were sacrificed 1 month after treatment. Obese (ob/ob) mice and their thin littermates (mixtures of ob/+ and +/+) of the C57BL65 strain were killed at 8-10 weeks of age. Organs and plasma were collected and stored as reported (1, 2).Human tissue samples, obtained either during operations or at autopsy within 3 hr after death, were frozen on dry ice prior to extraction. Whole blood (500 ml) from a fasting (overnight) normal human volunteer was separated into mononuclear and granulocytic fractions by a modification (3) of the method of The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact. 572Boyum (4). The cells were washed in phosphate-buffered salin...

Summary. While insulin effects on the central nervous system (CNS) mediated through hypoglycaemia are well known, direct insulin effects on the CNS remain controversial. Recently, we found insulin receptors in all areas of the rat brain, with highest concentrations in the olfactory bulb, cerebral cortex and hypothalamus; all areas involved in feeding. Insulin receptors in brain were, by multiple criteria, similar to insulin receptors on classical target tissues for insulin, such as liver and fat. Insulin itself has been identified in the rat brain at concentrations on average ten times higher than in plasma. Highest concentrations were found in the olfactory bulb and hypothalamus. Brain insulin was indistinguishable from purified insulin by its behaviour in the radioimmunoassay, radioreceptor assay, bioassay and gel chromatography. In two experimental models representing extremes of plasma insulin concentrations (obese hyperinsulinaemic mice and diabetic insulinopenic rats) there were no significant changes in the concentration of insulin receptors in brain while liver receptors were modified in the expected way. This may reflect the protective influence of the blood-brain barrier or some special quality of brain insulin receptors. Insulin concentrations in brain were also unchanged in both models, which is probably indicative of the local synthesis of insulin. The role of insulin in the CNS is unknown. Besides well known metabolic actions of insulin, new roles can be 9 postulated such as neurotransmission, neuromodulation and paracrine signalling.Key words: Central nervous system, CNS peptides, insulin receptors, insulin, experimental diabetesInsulin is a potent regulator of many aspects of metabolism in almost all mammalian tissues [1]. The central nervous system (CNS) is currently considered by most to be independent of insulin action, at least as far as glucose metabolism is concerned. However, significant effects are observed when the CNS is exposed to insulin. After the administration of insulin into the carotid artery [2], the cisterna magna [3] or into the ventromedial hypothalamus [4], peripheral hypoglycaemia is observed within minutes. The effect is probably mediated through the parasympathetic nervous system, since vagotomy greatly diminishes the hypoglycaemic response to the centrally administered insulin [5]. Whether similar mechanisms are operative under physiological conditions is unclear, but there is no doubt that insulin can act on the CNS under some circumstances. Since it is believed that most, if not all, insulin effects are mediated through its receptor localized on the outer surface of the cell, one would expect to find insulin receptors in the CNS. Insulin can reach the cerebrospinal fluid (CSF) and then, presumably, the brain tissue after its administration into the peripheral circulation [6,7]. The equilibrium between the periphery and the CNS is reached only after several hours and just a small fraction of the injected insulin can be found in the CSF. However, so called non-barrier re...