Insulin gene expression and insulin synthesis in mammalian neuronal cells

Effect of development and hypoxic-ischemia upon rabbit brain glucose transporter expression

We have cloned and sequenced a full length rabbit GLUT 1 and partial rabbit GLUT 3 cDNAs. The derived rabbit GLUT 3 peptide revealed 84% homology to the mouse, 82% to the rat, human, dog, and sheep, and 69% to the chicken GLUT 3 peptides. Using Northern blot analysis, we investigated the tissue and brain cellular distribution of GLUT 1 and GLUT 3 expression. In addition, we examined the effect of development and hypoxic-ischemia upon brain GLUT 1 and GLUT 3 mRNA levels. While GLUT 1 mRNA was observed in most tissues, GLUT 3 was expressed predominantly in the brain, placenta, stomach, and lung with minor amounts in the heart, kidney and skeletal muscle. In the brain, both GLUT 1 and GLUT 3 were noted in neuron- and glial-enriched cultures. Both GLUT 1 and GLUT 3 mRNA levels demonstrated a similar developmental progression (p<0.05) secondary to post-transcriptional mechanisms. Further, while hypoxic-ischemia did not significantly affect brain GLUT 1 mRNA and protein, it altered GLUT 3 mRNA levels in a region-specific manner, with a three-fold increase in the cerebral cortex, a two-fold increase in the hippocampus, and a 50% increase in the caudate nucleus (p<0.05). We conclude, that the rabbit GLUT 3 peptide sequence exhibits 82-84% homology to that of other species in the coding region with a 62-89% sequence identity in the 3'-untranslated region. The tissue-specific expression of rabbit GLUT 3 mimics that of the human closely. Postnatal development and hypoxic-ischemia with reperfusion injury cause an increase in brain GLUT 3 expression, as a response to synaptogenesis and substrate deprivation, respectively.

Brain insulin response to feeding in the rat is both macronutrient and area specific

Using microdialysis, we showed recently that hypothalamic immuno-reactive insulin (IRI) levels increased after a meal of chow and decreased in response to a fat meal. In the present study, we have compared extracellular hypothalamic and extrahypothalamic basal IRI levels and investigated the effect of meals composed exclusively of either carbohydrates (85% starch, 15% sucrose) or casein on both plasma and medial hypothalamic (PVN-VMH) insulin. The response of IRI to a carbohydrate meal was also investigated in the cerebellum. Basal hypothalamic IRI was twofold higher in the hypothalamus as compared to the cerebellum (33 +/- 4 and 15 +/- 2 pg/mL, respectively). Hypothalamic IRI increased twofold in response to the carbohydrate meal (72 +/- 15 pg/mL) but remained unchanged during the casein meal. No IRI change was found in the cerebellum after a meal of carbohydrates (16 +/- 2 pg/mL). Insulinemia was increased by both the carbohydrate and the casein meal. However, the protein-induced increase was less pronounced (maximum + 359% compared to 1650% for carbohydrates). The present data show a dual specificity of brain insulin response to feeding; in addition to the macronutrient specific variations, a regional specificity was also observed. Taken together with previous observations, the present data are in favor of an involvement of PVN-VMH insulin in the control of feeding and macronutrient-specific appetites.

Brain insulin response to feeding in the rat is both macronutrient and area specific

Using microdialysis, we showed recently that hypothalamic immunoreactive insulin (IRI) levels increased after a meal of chow and decreased in response to a fat meal. In the present study, we have compared extracellular hypothalamic and extrahypothalamic basal IRI levels and investigated the effect of meals composed exclusively of either carbohydrates (85% starch, 15% sucrose) or casein on both plasma and medial hypothalamic (PVN-VMH) insulin. The response of IRI to a carbohydrate meal was also investigated in the cerebellum. Basal hypothalamic IRI was twofold higher in the hypothalamus as compared to the cerebellum (33 +/- 4 and 15 +/- 2 pg/mL, respectively). Hypothalamic IRI increased twofold in response to the carbohydrate meal (72 +/- 15 pg/mL) but remained unchanged during the casein meal. No IRI change was found in the cerebellum after a meal of carbohydrates (16 +/- 2 pg/mL). Insulinemia was increased by both the carbohydrate and the casein meal. However, the protein-induced increase was less pronounced (maximum + 359% compared to 1650% for carbohydrates). The present data show a dual specificity of brain insulin response to feeding; in addition to the macronutrient specific variations, a regional specificity was also observed. Taken together with previous observations, the present data are in favor of an involvement of PVN-VMH insulin in the control of feeding and macronutrient-specific appetites.

Sp1 and Sp3 regulate transcriptional activity of the facilitative glucose transporter isoform-3 gene in mammalian neuroblasts and trophoblasts

The murine facilitative glucose transporter isoform 3 (Glut 3) is developmentally regulated and is predominantly expressed in neurons and trophoblasts. Employing the primer extension and RNase protection assays, the transcription start site (denoted as +1) of the murine Glut 3 gene was localized to 305 base pairs (bp) 5' to the ATG translation start codon. Transient transfection assays in N2A, H19-7 neuroblasts, and HRP.1 trophoblasts using sequential 5'-deletions of the murine Glut 3-luciferase fusion gene indicated that the -203 to +237 bp region with reference to the transcriptional start site contained promoter activity. Repressor function was limited to the -137 to -130 bp region within the transcriptional activation domain. The nuclear factors Sp1 and Sp3 bound this GC-rich region in N2A, H19-7, and HRP.1 cells. Dephosphorylation of Sp1 was essential for Glut 3 DNA binding. The related Sp3 protein also bound this same region of mouse Glut 3 in all three cell lines. Mutations of the Sp1-binding site employed in transient transfection and mobility shift assays confirmed the nature of the DNA-binding proteins, while supershift assays with anti-Sp1 and anti-Sp3 IgGs characterized the differences in the two DNA-binding proteins. Co-transfection of the Glut 3-luciferase fusion gene with or without mutations of the Sp1-binding site along with the Sp1 or Sp3 expression vectors in Drosophila SL2 cells confirmed a reciprocal effect, with Sp1 suppressing and Sp3 activating Glut 3 gene transcription.

Synthesis and differentially regulated processing of proinsulin in developing chick pancreas, liver and neuroretina

Regulated preproinsulin gene expression in nonpancreatic tissues during development has been demonstrated in rodents, Xenopus and chicken. Little is known, however, about the synthesis and processing of the primary protein product, proinsulin, in comparison with these events in pancreas. Using specific antisera and immunocytochemistry, immunoblot and HPLC criteria, we characterize the differential processing of proinsulin in developing neuroretina, liver and pancreas. The chick embryo pancreas expresses the convertase PC2, and largely processes proinsulin to insulin. In contrast, little or no mature PC2 is present in embryonic liver and neuroretina and the (pro)insulin immunoactivity identified is predominantly proinsulin.

Some of my not so favorite things about insulin and insulin-like growth factors in fish

Several topics concerning insulin and IGFs rarely discussed among comparative physiologists/endocrinologists are reviewed. The topics addressed include interpretation of radioimmunoassays for insulin and IGFs, the major sites of synthesis of these peptides, signal transduction via peptide-specific, and possibly via related peptide receptors. Finally, some thoughts are included on the development of the insulin superfamily of peptides in two phyletic lines of evolution, proto- and deuterostomian animals. The author considers all the above topics open to further exploration. If this review initiates more discussion leading to unorthodox input to the field, its goal is achieved.

Intrathymic expression of genes involved in organ specific autoimmune disease

Insulin, thyroglobulin and myelin basic protein (MBP) are implicated as autoantigens in the autoimmune diseases, insulin-dependent diabetes mellitus (IDDM), autoimmune thyroid-disease and multiple sclerosis. Self tolerance to these antigens, until recently only thought to be present extrathymically, is generally considered to be maintained by 'peripheral' mechanisms, such as clonal anergy or clonal ignorance. The techniques of reverse transcription and polymerase chain reaction (RT-PCR) were used to investigate the intrathymic expression of these genes. Expression was examined in mRNA isolated from complete adult rat thymus, various mouse thymic cell-types isolated from fetal thymic-organ cultures and from neonatal-mouse thymocyte subsets. mRNA for insulin, thyroglobulin and MBP were detected in unfractionated adult rat and embryonic mouse thymus. Rat thymus expressed both insulin I and II, while mouse thymus only expressed insulin II. Thyroglobulin and MBP, but not insulin mRNA were detected in mouse MHC class II+ thymic epthelial cells and class II+ dendritic cells and in certain thymocyte subsets. The presence of insulin, thyroglobubin and MBP mRNA in the thymus has important implications for the development of the T-cell repertoire, particularly for the mechanisms of tolerance that prevent autoreactivity to these antigens in healthy individuals.

Insulin responses to a fat meal in hypothalamic microdialysates and plasma

In a recent microdialysis study in freely-behaving rats, we observed changes in immunoreactive insulin (IRI) in hypothalamic dialysates after a meal of standard laboratory chow. These changes did not always parallel plasma insulin variations, suggesting a partial independence from peripheral insulin. In the present study, we have attempted to assess the profile of medial hypothalamus (VMPH-PVN) extracellular insulin and peripheral insulin before and after a fat meal (lard). In contrast to the increase we previously observed with chow meals, hypothalamic extracellular IRI decreased during the fat meal and fell to 60% 30 min after the meal. Plasma insulin levels did not change. The intake of the lard meal, provided in unlimited amounts, was much larger in calories than the intake of a chow meal under the same conditions. However, when rats were offered a meal of chow after they had eaten a meal averaging 6.7 g of fat (61 calories), they immediately began eating the chow. Thus, the meal of fat produced no general satiation. On the contrary, the rats consumed a second chow meal only after a delay of approximately 40 min after the first one. The present data, in conjunction with our previous observations with chow fed rats, suggest that the level of extracellular hypothalamic IRI may decrease independently of plasma insulin levels and may, at least partially, account for the observed lack of satiation.

Effect of streptozotocin-induced maternal diabetes on fetal rat brain glucose transporters

Glucose, an essential substrate for brain oxidative metabolism, is transported across the blood-brain barrier and into neuronal and glial cells via Glut 1 and Glut 3 facilitative glucose transporter isoforms. To examine the effect of excessive circulating glucose on fetal brain glucose transporter expression, we investigated the effect of streptozotocin-induced maternal diabetes (SEVERE-D; n = 29) on the 20-d gestation fetal rat brain Glut 1 and Glut 3. We studied the effect of streptozotocin alone (STZ-ND; n = 12) in a nondiabetic state as well, along with vehicle injected controls (C; n = 24). In the presence of fetal hyperglycemia (12.63 +/- 0.82 nM-SEVERE-D versus 2.35 +/- 0.28-STZ-ND and 2.42 +/- 0.16-C; p < 0.001) and hypoinsulinemia (0.38 +/- 0.03 nM-SEVERE-D versus 0.50 +/- 0.07-STZ-ND and 0.55 +/- 0.06-C; p < 0.02), no detectable change in fetal brain Glut 1 and Glut 3 pretranslational expression (transcription/elongation rates and corresponding steady state mRNA levels) was noted when simultaneously compared with the STZ-ND and C groups. In contrast, a trend toward a decline in Glut 1 (approximately 25 to 30%, p = 0.05) and a substantive decrease in Glut 3 (approximately 35 to 50%, p = 0.0006) protein concentrations was present in both the STZ-ND and SEVERE-D groups when compared with the C group. These observations support a chemical effect of streptozotocin independent of maternal diabetes upon the translation or posttranslational processing of fetal brain glucose transporters. Maternal diabetes with fetal hyperglycemia, however, failed to substantively alter fetal brain glucose transporters independent of the streptozotocin effects upon neuroectodermally derived tissues. We conclude that maternal diabetes with associated overt fetal hyperglycemia does not significantly change fetal brain glucose transporter levels.

Insulin-like growth factors and insulin stimulate erythropoietin production in primary cultured astrocytes

Erythropoietin (EPO) is established as a major regulator of erythropoiesis. However, we and others have shown that neurons express erythropoietin receptor (EPO-R), that astrocytes produce EPO and that EPO may act as a neurotrophic factor in the CNS. We also found that EPO production is activated by insulin and insulin-like growth factors (IGFs) in astrocytes in a dose-dependent manner and that IGF-I was the most potent activator. The concentrations required for half-maximal activation were 3 nM IGF-I, 10 nM IGF-II and 100 nM insulin. The oxygen concentration regulates EPO production; hypoxia stimulates EPO production in astrocytes. The stimulatory effect of IGFs and insulin on EPO production in astrocytes was not affected by the oxygen concentration of astrocyte culture. Insulin and IGFs did not increase the total protein synthesis of astrocytes but increased EPO mRNA levels, indicating that EPO production is stimulated at the mRNA level. It appeared that the growth factor-induced accumulation of EPO mRNA in astrocytes was caused by activation of the tyrosine kinase-signal transduction pathway, because tyrosine phosphorylation of receptors for IGF-I and insulin was activated when astrocytes were stimulated by these growth factors.

Preproinsulin I and II mRNAs and insulin electron microscopic immunoreaction are present within the rat fetal nervous system

Insulin-like substance has been found within the nervous system. In the rat, preproinsulin II mRNA was shown within the brain and preproinsulin I mRNA within the retina. The present study demonstrates the presence of preproinsulin mRNAs within the 15, 17 and 19 day gestational age fetal rat brain, spinal cord and dorsal root ganglia (DRG), employing RNA template-specific polymerase chain reaction (RS-PCR), semi-nested PCR and RNase protection assay. Preproinsulin I mRNA was present in the 17 and 19 day gestational age brain, spinal cord and DRG, and only in the brain of the 15 day gestational age brain. Preproinsulin II mRNA was present in all the gestational ages studied in the brain, spinal cord and DRG. The RS-PCR and the semi-nested PCR demonstrated products that co-migrated with the pancreatic control. The semi-nested products were characterized as preproinsulin I and II by restriction enzyme digestion and sequence. RNase protection assay using specific cRNA for preproinsulin I and II showed a band that co-migrated with pancreatic preproinsulin I and II mRNAs, and confirmed the PCR results. In addition, insulin receptor mRNA was detected by RS-PCR. Ultrastructural studies showed insulin immunoreaction within the endoplasmic reticulum, Golgi apparatus, cytoplasm, axon, dendrites, and in relation to the synapses. Thus, we demonstrated the presence of preproinsulin I and II mRNA, insulin receptor mRNA and insulin immunoreaction within the rat fetal central and peripheral nervous system.

The early intracellular signaling pathway for the insulin/insulin-like growth factor receptor family in the mammalian central nervous system

Several studies support the idea that the polypeptides belonging to the family of insulin and insulin-like growth factors (IGFs) play an important role in brain development and continue to be produced in discrete areas of the adult brain. In numerous neuronal populations within the olfactory bulb, the cerebral and cerebellar cortex, the hippocampus, some diencephalic and brainstem nuclei, the spinal cord and the retina, specific insulin and IGF receptors, as well as crucial components of the intracellular receptor signaling pathway have been demonstrated. Thus, mature neurons are endowed with the cellular machinery to respond to insulin and IGF stimulation. Studies in vitro and in vivo, using normal and transgenic animals, have led to the hypothesis that, in the adult brain, IGF-I not only acts as a trophic factor, but also as a neuromodulator of some higher brain functions, such as long-term potentiation and depression. Furthermore, a trophic effect on certain neuronal populations becomes clearly evident in the ischemic brain or neurodegenerative disorders. Thus, the analysis of the early intracellular signaling pathway for the insulin/IGF receptor family in the brain is providing us with new intriguing findings on the way the mammalian brain is sculpted and operates.

Oxidative metabolism deficiencies in brains of patients with Alzheimer's disease

Glucose metabolism in the brain has an important influence on many normal cellular processes. It contributes to the synthesis of acetylcholine, glutamate, aspartate, gamma-aminobutyric acid, glycine, and ATP production (the driving force behind almost all cellular and molecular activity). Neuronal glucose metabolism is controlled antagonistically by insulin and cortisol. Desensitization of the neuronal insulin receptor causes abnormalities in oxidative energy metabolism. During normal aging, the cerebral energy pool is slightly diminished, but its level increases after stressful events. In age-related sporadic late-onset dementia of the Alzheimer type (SDAT), glucose metabolism and formation of cellular energy are severely reduced. Desensitization of the neuronal insulin receptor seems to be an early event in the pathogenesis or even etiology of SDAT causing disturbances in oxidative glucose metabolism and energy failure in insulin-sensitive brain structures. These abnormalities appear to induce a cascade of disturbances that leads to abnormal APP processing and amyloid formation, membrane damage, and neuronal death.

Insulin and insulin-like growth factor-I receptors in fish brain

We investigated insulin and insulin-like growth factor-I (IGF-I) receptor-binding and receptor intrinsic tyrosine kinase activity in the brain of carp (Cyprinus carpio) and trout (Salmo trutta fario). Glycoprotein fractions of semi-purified receptors were prepared by WGA-agarose affinity chromatography. Insulin receptors were found in the brains of both fish species investigated. Carp and trout brain preparations bound, respectively (per 50 micrograms glycoprotein), with 6.0 +/- 1.5% and 8.0 +/- 2.0% of 125I-labeled insulin added to the assay. Insulin binding was specific: much higher quantity of IGF-I (EC50 165 +/- 11 nM for carp and 88.0 +/- 6 nM for trout receptors) than insulin (EC50 0.26 +/- 0.04 nM for carp and 0.25 +/- 0.02 nM for trout) was necessary to displace bound insulin tracer. In preparations of brain receptors, IGF-I binding (52.8 +/- 6.5% in carp brain and 55.0 +/- 13.0% in trout brain) surpassed insulin binding several fold. IGF-I bound to the brain receptors with high affinity (Kd for carp was 0.13 +/- 0.06 nM and for trout 0.22 +/- 0.11 nM) and specificity. Although IGF-I binding could be displaced with insulin, EC50 were 660 +/- 51 nM for carp and 1557 +/- 194 nM for trout. Both ligands stimulated phosphorylation of exogenous substrates in a dose-dependent manner. Carp brain receptors were not significantly different from trout receptors with respect to basal phosphotransferase activities (250.0 +/- 50.0 fm P/mg glycoprotein in carp and 330.0 +/- 120.0 fm P/mg glycoprotein in trout). In both species IGF-I caused higher maximal stimulation (308.0 +/- 36.0% and 270.0 +/- 39%, for carp and trout, respectively) than insulin (250.0 +/- 13.0% and 209.0 +/- 6.0%, for carp and trout, respectively).

Brain glucose metabolism is controlled by amplification and desensitization of the neuronal insulin receptor

Glucose metabolism is essential for brain function and structure. Glucose contributes to the formation of neurotransmitters and is normally the only source for energy formation. There is increasing evidence that brain glucose metabolism is under control of the neuronal insulin/insulin receptor signal transduction. The present data clearly show that intracerebroventricularly administered insulin exerts anabolic effects on cerebral glucose/energy metabolism (amplification of the neuronal insulin receptor complex) whereas cortisol (corticosterone) acts antagonistically (desensitization of the neuronal insulin receptor complex). It is also shown that short-term cortisol (corticosterone) enhanced energy turnover in temporoparietal cortex and hippocampus. In contrast, long-term cortisol (corticosterone) reduced energy turnover in both brain structures studied. This metabolic pattern is reminiscent of that found in very old age. Therefore, it is assumed that long-term cortisol accelerates the aging process in the brain and thus the risk for age-related disorders such as dementia.

Desensitization of the neuronal insulin receptor: a new approach in the etiopathogenesis of late-onset sporadic dementia of the Alzheimer type (SDAT)?

Even in its incipient stage, late-onset sporadic dementia of the Alzheimer type (SDAT) is characterized by an abnormal reduction in brain glucose consumption and energy formation. Gathering evidence indicates that cerebral glucose metabolism is controlled by brain insulin/insulin receptors. This led us to hypothesize that the abnormal reduction in glucose utilization found in Alzheimer brains is preceded by a desensitization of cerebral insulin receptors which might be due to enhanced levels of stress factors such as cortisol and catecholamines. The hypothesis is supported by clinical findings of an abnormal response to the oral glucose tolerance test in AD patients. Furthermore, experimental desensitization of the cerebral insulin receptor resulted in both cognitive deficits and metabolic abnormalities in cerebral oxidative glucose metabolism resembling those described in incipient late-onset SDAT. Glucose is the major source of energy in the CNS, and any impairment in cerebral glucose oxidation can be expected to result in deficits in both acetylcholine synthesis and ATP formation, which might contribute to altered APP processing and enhanced susceptibility to neurotoxicity.

The developing CNS: a scenario for the action of proinsulin, insulin and insulin-like growth factors

The multifunctional cytokines of the family of insulin and insulin-like growth factors (IGFs) have not yet gained general recognition as essential cell signals for the development of the vertebrate nervous system. This is, in part, a consequence of previous constraints in our thinking, focused for many years on the endocrine roles of these factors in late mammalian development and postnatal stages. The cellular distribution of the components of the insulin and IGFs signalling system in the developing mammalian and avian CNS is remarkably conserved. While receptors are widespread, the much less abundant factors and modulatory proteins are highly regulated in time and space. Progression of neural development through the steps of cell proliferation, differentiation, maturation and survival is stimulated, at least in culture, by proinsulin and insulin and the IGFs. Thus, these factors might be important autocrine and paracrine signals during development of the CNS.

An immunohistochemical and in situ hybridization study of insulin-like growth factor I within fetal neuron cell cultures

Fetal neuron cell cultures (NCC) from 22 day gestation and 18 day gestation fetal rabbit brain were studied for the presence of insulin-like growth factor I (IGF I). The 22 day gestation NCC were incubated in an IGF I free/insulin free/serum free medium. The 18 day gestation NCC were incubated in: (1) IGF I free/insulin free/serum free medium, (2) IGF I containing medium (100 ng)/serum free medium, and (3) serum containing medium. The 22 day gestation NCC survived in the IGF I free/insulin free/serum free medium. Furthermore, IGF I was detected in the medium by RIA from day one to day ten of incubation. In contrast, the 18 day gestation NCC did not survive in the IGF I free/insulin free/serum medium, but survived in the serum medium. When the 18 day gestation NCC were incubated in the serum free medium containing 100 ng IGF I the cells survived for a period of 2-3 days. Immunoreactive IGF I was found within the 22 day gestation NCC incubated in the IGF I free/insulin free/serum free medium and 18 day gestation NCC in serum medium. Likewise, IGF I mRNA was found only within the 22 day gestation NCC. Internalization studies of IGF I have shown that the peptide was internalized from the medium by the two different gestational age NCC's studied. IGF I receptors were found in both 22 day gestation and 18 day gestation NCC. In conclusion IGF I may promote cell survival in early stages of brain development, and may be of exogenous origin. In contrast the 22 day gestation NCC are capable of producing and secreting IGF I, and indeed appear to respond to this growth factor in an autocrine fashion.

Intracerebroventricular injection of streptozotocin induces discrete local changes in cerebral glucose utilization in rats

The purpose of the present study was to investigate whether or not cerebral glucose utilization is changed locally after damage of the neuronal insulin receptor by means of intracerebroventricular (icv) streptozotocin (STZ) administered in a subdiabetogenic dosage (1.5 mg/kg bw.). STZ was administered at the start of the study, and 2 and 21 days later bilaterally into the cerebral ventricles in rats of a mean age of 18 months. The local distribution of cerebral glucose utilization was analyzed in conscious rats on the 42nd day after the first STZ injection using the quantitative (14C)-2-deoxyglucose method. Of the 35 brain structures investigated from autoradiograms of brain sections, 17 showed a reduction in glucose utilization. Decreases in glucose utilization were observed in the frontal, parietal, sensory motor, auditory and entorhinal cortex and in all hippocampal subfields. In contrast, glucose utilization was increased in two white matter structures. The decrease in cerebral glucose utilization observed in cortical and hippocampal areas in the present study may correspond to changes in morphobiological parameters which have been found in patients with Alzheimer's disease. The present data are in accordance with the hypothesis that an impairment in the control of neuronal glucose metabolism at the insulin receptor site may exist in sporadic dementia of Alzheimer type (DAT), and can be studied by the icv STZ animal model.

Short-term or long-term intracerebroventricular (i.c.v.) infusion of insulin exhibits a discrete anabolic effect on cerebral energy metabolism in the rat

Evidence for an involvement of insulin in the regulation of cerebral glycolytic glucose catabolism is becoming more and more convincing. To investigate whether short-term or long-term administration of insulin to the brain influences brain energy metabolism, we determined tissue concentrations of energy-rich phosphates in cerebral parietotemporal cortex and in hippocampus of 1- (adult) and 2-year-old (aged) male Wistar rats treated with intracerebroventricular (i.c.v.) infusion of the hormone for 1, 7, and 21 days. In cerebral parietotemporal cortex, tissue concentrations of ATP, PCr, ADP, ATP/ADP ratio and energy charge potential were all unaltered. In hippocampus, however, the concentration of PCr was significantly increased. Thus, short-term and long-term i.c.v. administration of insulin are assumed to have a discrete anabolic effect on the cerebral energy pool.