Denervation and hyperinnervation in the nervous system of diabetic animals: III. Functional alterations of G proteins in diabetic encephalopathy

Therapeutic potential of medicinal marijuana: an educational primer for health care professionals

With the proposed Canadian July 2018 legalization of marijuana through the Cannabis Act, a thorough critical analysis of the current trials on the efficacy of medicinal marijuana (MM) as a treatment option is necessary. This review is particularly important for primary care physicians whose patients may be interested in using MM as an alternative therapy. In response to increased interest in MM, Health Canada released a document in 2013 for general practitioners (GPs) as an educational tool on the efficacy of MM in treating some chronic and acute conditions. Although additional studies have filled in some of the gaps since the release of the Health Canada document, conflicting and inconclusive results continue to pose a challenge for physicians. This review aims to supplement the Health Canada document by providing physicians with a critical yet concise update on the recent advancements made regarding the efficacy of MM as a potential therapeutic option. An update to the literature of 2013 is important given the upcoming changes in legislation on the use of marijuana. Also, we briefly highlight the current recommendations provided by Canadian medical colleges on the parameters that need to be considered prior to authorizing MM use, routes of administration as well as a general overview of the endocannabinoid system as it pertains to cannabis. Lastly, we outline the appropriate medical conditions for which the authorization of MM may present as a practical alternative option in improving patient outcomes as well as individual considerations of which GPs should be mindful. The purpose of this paper is to offer physicians an educational tool that provides a necessary, evidence-based analysis of the therapeutic potential of MM and to ensure physicians are making decisions on the therapeutic use of MM in good faith.

Evidence for D2 receptor mediation of amphetamine-induced normalization of locomotion and dopamine transporter function in hypoinsulinemic rats

Dopamine (DA) D2 receptors regulate DA transporter (DAT) activity, and mediate some behavioral effects of amphetamine. DA clearance and amphetamine-stimulated locomotion are reduced in hypoinsulinemic [streptozotocin (STZ)-treated] rats, and these deficits are normalized by repeated treatment with amphetamine. Here, a role for D2 receptors in mediating amphetamine-induced normalization of these parameters was investigated. One week after a saline or STZ injection (50 mg/kg), rats were treated with amphetamine (1.78 mg/kg), raclopride (0.056 mg/kg), saline, or combinations thereof, every-other-day for 8 days with locomotor activity measured following each treatment. Conditioned place preference (CPP) for amphetamine and in vivo chronoamperometry to measure DA clearance were carried out on days 17 and 18, respectively, after STZ or saline. Baseline locomotion and DA clearance were significantly reduced in STZ-treated rats compared with control rats. In STZ-treated rats, amphetamine treatment normalized DA clearance, and restored the locomotor-stimulating effects of amphetamine. Raclopride prevented normalization of these parameters. Amphetamine produced CPP in both STZ-treated and control rats; raclopride significantly attenuated amphetamine-induced CPP in control and not in STZ-treated rats. These results support a role for D2 receptors in regulating DA transporter activity, and further demonstrate that D2 receptors contribute to changes in sensitivity to amphetamine in hypoinsulinemic rats.

Insulin replacement restores the behavioral effects of quinpirole and raclopride in streptozotocin-treated rats

Streptozotocin (STZ)-induced diabetes can modulate dopamine (DA) neurotransmission and thereby modify the behavioral effects of drugs acting on DA systems. Insulin replacement, and in some conditions repeated treatment with amphetamine, can partially restore sensitivity of STZ-treated rats to dopaminergic drugs. The present study sought to characterize the role of insulin and amphetamine in modulating the behavioral effects of drugs that selectively act on D2/D3 receptors. In control rats, quinpirole and quinelorane produced yawning, whereas raclopride and gamma-hydroxybutyric acid (GHB) produced catalepsy. Raclopride antagonized quinpirole- and quinelorane-induced yawning with similar potency. STZ treatment increased blood glucose concentration, decreased body weight, and markedly reduced sensitivity to quinpirole-induced yawning, quinelorane-induced yawning as well as to raclopride-induced catalepsy, while enhancing sensitivity to GHB-induced catalepsy. Repeated treatment with amphetamine partially restored sensitivity of STZ-treated rats to amphetamine-stimulated locomotion and also produced conditioned place preference, without affecting blood glucose and body weight changes. However, amphetamine treatment did not restore sensitivity to the behavioral effects of quinpirole, raclopride, or GHB, suggesting differential regulation of dopamine transporter activity and sensitivity of D2 receptors in hypoinsulinemic rats. Insulin replacement in STZ-treated rats normalized blood glucose and body weight changes and fully restored sensitivity to quinpirole-induced yawning, as well as to raclopride-induced catalepsy, while reducing sensitivity to GHB-induced catalepsy. Overall, these data indicate that changes in insulin status markedly affect sensitivity to the behavioral effects of dopaminergic drugs. The results underscore the importance of insulin in modulating DA neurotransmission; these effects might be especially relevant to understanding the co-morbidity of eating disorders and substance abuse.

Modulatory Effect of Gymnema montanum Leaf Extract on Brain Antioxidant Status and Lipid Peroxidation in Diabetic Rats

The effects of leaf extract from Gymnema montanum, an endangered and endemic plant, were examined on brain lipid peroxidation in experimental diabetic rats. Ethanolic extract of G. montanum leaves was administered orally (50, 100, and 200 mg/kg of body weight) for 3 weeks, and changes in blood glucose, plasma insulin, and lipid peroxidation markers such as thiobarbituric acid-reactive substances (TBARS), hydroperoxides, and levels of antioxidants, namely, superoxide dismutase, catalase, glutathione peroxidase, reduced glutathione, and glutathione-S-transferase, were examined in the brain of alloxan-induced diabetic rats. Glibenclamide was used as a standard reference drug. A significant increase in the activities of antioxidants was observed in brain on treatment with G. montanum leaf extract and glibenclamide for 3 weeks. Both the treated groups showed significant decreases in formation of TBARS and hydroperoxides in brain, suggesting a role in protective action against lipid peroxidation-mediated membrane damage. Our findings indicate that G. montanum leaf extract possesses antiperoxidative and antioxidant effects in addition to its antidiabetic activity. This report helps to create awareness on the need for conservation of medicinal plants, and G. montanum is one such plant that needs to be conserved through various propagation trials.

Preventive effects of Cassia auriculata L. flowers on brain lipid peroxidation in rats treated with streptozotocin

The effect of aqueous extract of the flowers of Cassia auriculata were examined on antioxidants and lipid peroxidation in the brain of streptozotocin diabetic rats. Significant increase in the activities of superoxide dismutase, catalase, glutathione peroxidase, glutathione-S-transferase and reduced glutathione were observed in brain on treatment with Cassia auriculata flower extract (CFEt) and glibenclamide. Both the treated groups showed significant decrease in thiobarbituric reactive substances (TBARS) and hydroperoxide formation in brain, suggesting its role in protection against lipid peroxidation induced membrane damage. Since the study of induction of the antioxidant enzymes is considered to be a reliable marker for evaluating the antiperoxidative efficacy of medicinal plant, these findings are suggestions of possible antiperoxidative role played by Cassia auriculata flower extract.

Characterization of the signalling pathways involved in ATP and basic fibroblast growth factor-induced astrogliosis

1. A brief challenge of rat astrocytes with either alpha, beta-methyleneATP (alpha, beta-meATP) or basic fibroblast growth factor (bFGF) resulted, three days later, in morphological differentiation of cells, as shown by marked elongation of astrocytic processes. The P2 receptor antagonist suramin prevented alpha, beta-meATP- but not bFGF-induced astrocytic elongation. Similar effects on astrocytic elongation were also observed with ATP and other P2 receptor agonists (beta, gamma meATP, ADP beta S, 2meSATP and, to a lesser extent, UTP). 2. Pertussis toxin completely abolished alpha, beta-meATP- but not bFGF-induced effects. No effects were exerted by alpha, beta-meATP on cyclic AMP production; similarly, neomycin had no effects on elogation of processes induced by the purine analogue, suggesting that adenylyl cyclase and phospholipase C are probably not involved in alpha, beta-meATP-induced effects (see also the accompanying paper by Centemeri et al., 1997). The tyrosine-kinase inhibitor genistein greatly reduced bFGF- but not alpha, beta-meATP-induced astrocytic elongation. 3. Challenge of cultures with alpha, beta-meATP rapidly and concentration-dependently increased [3H]-arachidonic acid (AA) release from cells, suggesting that activation of phospholipase A2 (PLA2) may be involved in the long-term functional effects evoked by purine analogues. Consistently, exogenously added AA markedly elongated astrocytic processes. Moreover, various PLA2 inhibitors (e.g. mepacrine and dexamethasone) prevented both the early alpha, beta-meATP-induced [3H]-AA release and/or the associated long-term morphological changes, without affecting the astrocytic elongation induced by bFGF. Finally, the protein kinase C (PKC) inhibitor H7 fully abolished alpha, beta-meATP- but not bFGF-induced effects. 4. Both alpha, beta-meATP and bFGF rapidly and transiently induced the nuclear accumulation of Fos and Jun. Both c-fos and c-jun induction by the purine analogue could be fully prevented by pretreatment with suramin. In contrast, the effects of bFGF were unaffected by this P2 receptor antagonist. 5. It was concluded that alpha, beta-meATP- and bFGF-morphological differentiation of astrocytes occurs via independent transductional pathways. For the purine analogue, signalling involves a Gi/G(o) protein-coupled P2Y-receptor which may be linked to activation of PLA2 (involvement of an arachidonate-sensitive PKC is speculated); for bFGF, a tyrosine kinase receptor is involved. Both pathways merge on some common intracellular target, as suggested by induction of primary response genes, which in turn may regulate late response genes mediating long-term phenotypic changes of astroglial cells. 6. These findings implicate P2 receptors as novel targets for the pharmacological regulation of reactive astrogliosis, which has intriguing implications in nervous system diseases characterized by degenerative events.

Endogenous mono-ADP-ribosylation in retina and peripheral nervous system. Effects of diabetes

The extranuclear endogenous mono-ADP-ribosylation of proteins was monitored in cellular preparations of retina, superior cervical ganglion, dorsal root ganglia and peripheral nerve. At least 6 protein fractions are ADP-ribosylated in the crude extract fraction from retina control preparations, while in diabetic rats the number of retina labeled proteins and the extent of labeling are highly reduced. In the superior cervical ganglion labeling was present in 10 proteins, in diabetics it was greatly decreased. Treatment of diabetic rats with silybin, a flavonoid mono-ADP-ribosyltransferase inhibitor, did not affect hyperglycemia, but prevented the alteration of extent of protein ADP-ribosylation. These data suggest that proteins of retina and peripheral ganglia are excessively ADP-ribosylated in vivo. The effects of silybin treatment on excessive mono-ADP-ribosylation of proteins was associated with the prevention of reduction of substance P-like immunoreactivity levels, that is typical of diabetic neuropathy. In the membrane fraction of sciatic nerve Schwann cells, at least 9 proteins were ADP-ribosylated, diabetes caused a marked increase of labeling. A comparable increase involving the same proteins is triggered by chronic nerve injury and by corticosteroid treatment. Silybin treatment of diabetic rats prevented such an increase. We propose that the inhibition of excessive protein mono-ADP-ribosylation by silybin prevented the onset of diabetic neuropathy. While the effects on Schwann cells is likely indirect and secondary to the improvement of diabetic axonopathy.

Alterations of protein mono-ADP-ribosylation and diabetic neuropathy: a novel pharmacological approach

This study monitored the extranuclear endogenous mono ADP-ribosylation of proteins. At least 10 proteins were ADP-ribosylated in a crude extract from control superior cervical ganglia, and 7 were labeled in control dorsal root ganglia; whereas in the diabetic rat the extent of labeling was reduced. These data suggest that proteins of peripheral ganglia are excessively ADP-ribosylated in vivo. Treatment of diabetic animals with silybin, a flavonoid with ADP-ribosyltransferase inhibitory activity, did not affect hyperglycemia, but prevented the alterations in the extent of mono-ADP-ribosylation of proteins. This effect was associated with the prevention of substance P-like immunoreactivity loss in the sciatic nerve. In the membrane fraction of sciatic nerve Schwann cells, at least 9 proteins were ADP-ribosylated, in diabetic rats the extent of labeling was increased. A comparable increase involving the same proteins was triggered by chronic nerve injury and by corticosteroid treatment. Silybin treatment of diabetic rats prevented such an increase. We propose that the inhibition of excessive protein mono-ADP-ribosylation by silybin prevented the onset of diabetic neuropathy, while the silybin effect on mono-ADP-ribosylation of Schwann cells is likely indirect and secondary to the improvement of diabetic axonopathy.

Diabetes and the nervous system

Hyperglycemia and its vascular complications affect the entire nervous system, contributing to increased morbidity and mortality. Chronic hyperglycemia is not only a known and major risk factor for cerebral vascular diseases but also the presence of hyperglycemia at the time of a cerebrovascular event may adversely influence the outcome. It also affects the treatment of some neurodegenerative disorders, and there are suggestions that diabetes may in fact suffer from a "chronic diabetic encephalopathy." Its varied effects on the peripheral nervous system result in several forms of diabetic neuropathies, the exact pathogenesis of which is still obscure. There is, however, some new information that may link metabolic and vascular hypotheses.

Signal transduction alterations in peripheral nerves from streptozotocin-induced diabetic rats

We have previously determined the presence of muscarinic receptors and the expression of several G proteins in homogenates and myelin fractions from rat sciatic nerves. In the present study we investigated whether changes in several signal transduction pathways in peripheral nerves might be responsible for some of the biochemical abnormalities (e.g., phosphoinositide metabolism) present in sciatic nerves from streptozotocin-induced diabetic rats. Sciatic nerves from 5 week diabetic rats that were prelabelled with [3H]-myo-inositol displayed a significant increase in the basal release of inositol mono- and bis-phosphate, while carbamylcholine-stimulated release was significantly smaller. Basal- and forskolin-stimulated adenylyl cyclase activity was significantly decreased in sciatic nerve homogenates from diabetic animals. However, we were unable to detect any significant differences in the levels of cAMP in intact nerves or in nerve segments that were incubated in the presence or absence of forskolin. ADP-ribosylation experiments showed that in sciatic nerves from experimentally diabetic rats there was a significant increase in the ADP-ribosylation catalyzed by cholera and pertussis toxins. Measurements of the levels of alpha-subunits of G proteins revealed that the expression of Gq/11 alpha, Gs alpha, and Gi-3 alpha was increased by 30 to 50%. These results indicate that during the course of experimental diabetes, peripheral nerves exhibit an abnormal production of inositol phosphates and cAMP, together with an abnormal expression and/or function of G proteins. One of the consequences of such alterations is the diminished release of inositol phosphates triggered by muscarinic agonists in diabetic sciatic nerves.

Striatal dopaminergic D1 and D2 receptors after intracerebroventricular application of alloxan and streptozocin in rat

Intracerebroventricular application of low, nondiabetogenic doses (500 micrograms kg-1) of alloxan and streptozocin is followed by alterations of the dopaminergic system in rat striatum. In this brain region the dopamine content significantly increased, while the density of dopaminergic D1 receptors significantly decreased seven days after the intracerebroventricular application of betacytotoxics, as compared with the control group. The density of dopaminergic D2 receptors in striatum remained unchanged. Dopaminergic D1 and D2 receptors operate through signalling mechanism of G proteins, but no changes of Gs and Gi proteins content have been found in rat striatum after the intracerebroventricular application of betacytotoxics. As intracerebroventricular, nondiabetogenic administration of betacytotoxics produces changes of the striatal dopamine content and D1 receptor density similar to that produced by peripheral, diabetogenic administration of these drugs, the effect might be related not solely to pancreatic beta cells damage, but to alterations of the brain insulin system, as well.

Nitric oxide-sensitive protein ADP-ribosylation is altered in rat diabetic neuropathy

Endogenous ADP-ribosylation of proteins was studied in retina crude extract, membrane and cytosolic fractions of control and diabetic rats. ADP-ribosyltransferase activity is present in all cellular fractions, but protein ADP-ribosylation is reduced in diabetic rat retina. At least 6 proteins are labelled in the crude extract fraction and a similar number in the membrane preparation of control animals. In these preparations from diabetic retina, only two bands were labelled, the 85 K and 36 K for the crude extract, and the 97 K and 39 K for membranes. Labelling of 36 K and 39 K proteins was much less than in controls. In the cytosolic preparations of controls, two proteins of 85 K and 39 K are ADP-ribosylated, while in diabetic rat retina cytosol, only the 85 K is labelled. Treatment of diabetic rats with insulin normalized plasma glucose levels and prevented the alterations of the extent of ADP-ribosylation for the 38 K cytosolic, 39 K membrane and 36 K crude extracts proteins, but it failed to affect the other bands. These results suggest a hyperactivity of endogenous ADP-ribosylases in diabetic rat retina, so that the protein sites for ADP-ribosylation are no longer available. Since insulin treatment prevents the onset of neuropathy and of retinal G protein impairment (Abbracchio et al., J Neurosci Res 29:196-220, 1991) in diabetic rats and, in this study, normalizes ADP-ribosylation of 39 K, 38 K and 36 K proteins, we suggest that the abnormal endogenous ADP-ribosylation of these proteins might play a role in the onset of diabetic neuropathy.

Cerebral function in diabetes mellitus

Diabetes mellitus is a common metabolic disorder associated with chronic complications such as nephropathy, angiopathy, retinopathy and peripheral neuropathy. Diabetes is not often considered to have deleterious effects on the brain. However, long-term diabetes results in a variety of subtle cerebral disorders, which occur more frequently than is commonly believed. Diabetic cerebral disorders have been demonstrated at a neurochemical, electrophysiological, structural and cognitive level; however, the pathogenesis is still not clear. Probably alterations in cerebral blood supply and metabolic derangements play a role, as they do in the pathogenesis of diabetic neuropathy. Furthermore, the brain is also affected by recurrent episodes of hypoglycaemia and poor metabolic control. We describe herein the cerebral manifestations of diabetes and discuss the putative pathogenetic mechanisms.

Guanosine-5'-(3-O-thio)triphosphate-mediated stimulation of phosphoinositidase C in solubilized rat peripheral nerve myelin and its alteration in streptozotocin-induced diabetes

The regulation of phosphoinositidase C (PIC) activity by guanosine-5'-(3-O-thio)triphosphate (GTP gamma S) was characterized in a cholate-solubilized peripheral myelin-enriched fraction from rat sciatic nerve. The GTP analog maximally enhanced PIC-catalyzed hydrolysis of exogenous phosphatidylinositol-4,5-bisphosphate (PIP2) in a dose-dependent manner only within a narrow range of cholate concentrations. Maximal stimulation was attained at 0.6 microM GTP gamma S and could be completely prevented by 1 microM guanosine-5'-(2-O-thio)diphosphate. Neither adenylyl-imidodiphosphate nor adenosine triphosphate (ATP) enhanced PIC activity. Carbamoylcholine (1 mM) added together with GTP gamma S increased the extent of PIP2 hydrolysis over that elicited by GTP gamma S alone and this stimulation was blocked by the muscarinic receptor antagonist, atropine (50 microM). In detergent-solubilized myelin preparations from streptozotocin-induced diabetic rats, a higher concentration of the guanine nucleotide analog was required to achieve stimulation comparable to that obtained with corresponding preparations from normal animals. These results suggest that sciatic nerve myelin possesses muscarinic receptors coupled via a GTP-binding protein to PIC and that this system can be reconstituted in detergent-solubilized extracts. It is possible that the function of G proteins in cell signaling is impaired in experimental diabetic neuropathy.

Effect of diabetes on levels of lipid peroxides and glycolipids in rat brain

The effects of diabetes on levels of lipid peroxides and glycolipids in brain were studied in alloxan (18 mg/100 g body weight) diabetic rats. Free fatty acid (FFA) and malondialdehyde (MDA) levels were increased in the brains of diabetic animals. On the other hand, activities of the antioxidative enzymes catalase and superoxide dismutase (SOD) were decreased. The study also showed elevated levels of most of the glycolipid fractions except gangliosides, which were found to decrease in diabetic brain. Administration of insulin to diabetic animals results in the restoration of these parameters to normal levels. These changes observed in diabetic brain may be responsible for the increased frequency of stroke in diabetes.

Diabetes-induced alterations of central nervous system G proteins. ADP-ribosylation, immunoreactivity, and gene-expression studies in rat striatum

Previous studies from our laboratory have suggested that diabetes-associated central nervous system abnormalities are characterized by progressive alterations of neurotransmitters and of transductional Gi/Go proteins. In this study, we have further characterized these abnormalities in the striatum of alloxan-diabetic rats by means of adenosine 5'-diphosphate (ADP)-ribosylation, and Western and Northern blotting techniques. Fourteen weeks after diabetes induction, pertussis-toxin (PTX) catalyzed ADP-ribosylation of Gi/Go proteins was markedly reduced in diabetic animals, as shown by a clear decrease of 32P-ADPribose incorporation into G protein alpha subunits. In agreement with our previous pharmacological studies that showed a reduction of Gi-mediated modulation of adenylate cyclase activity only at this stage of diabetes, no changes in PTX-mediated ADP-ribosylation were observed earlier (5-wk diabetes). Immunoblotting studies performed by using antibodies selectively raised against Gi-2, Go, and Gs proteins did not reveal any differences between control and diabetic animals at any stage of diabetes. Similarly, the mRNAs corresponding to the alpha subunits of Gi-2, Go, and Gs proteins did not show any marked changes in chronic diabetic rats with respect to control animals. It is therefore concluded that diabetes is associated with development of a time-related alteration of cerebral Gi/Go proteins and that this defect is not owing to gross changes in either content of G proteins or mRNA level, but probably reflects modifications of G protein's structure or physiological status affecting the coupling with membrane effector systems and the sensitivity to PTX.

Alterations in G-protein expression, Gi function and stimulatory receptor-mediated regulation of adipocyte adenylyl cyclase in a model of insulin-resistant diabetes with obesity

The stimulatory effect of Mn2+ (1.5-fold), forskolin (1.6-fold) and low (1 microM) concentrations of GTP (1.9-fold) on the adenylyl cyclase of adipocyte membranes from obese, diabetic CBA/Ca mice was markedly enhanced compared to that seen using membranes prepared from their lean littermates. In contrast, receptor-mediated stimulation, achieved with either isoprenaline or secretin was reduced and that by glucagon abolished in membranes from diabetic animals. The levels of expression of alpha-subunits of Gi-1, Gi-2 and Gi-3 were reduced to some 49, 76 and 54%, respectively, in membranes from diabetic animals compared with those from normal animals. Levels of G-protein beta-subunits and Gs alpha-subunits were similar. Receptor-mediated inhibition of adenylate activity elicited by either nicotinic acid or prostaglandin E1 (PGE1) was of a similar magnitude in membranes from normal and diabetic animals but the inhibitory action of N6-(L-2-phenylisopropyl)adenosine (PIA) was greater in membranes from diabetic animals by about 30%. Gi function was similarly evident in membranes from both lean and diabetic animals, as assessed using low concentrations of guanylyl 5'-imidodiphosphate to inhibit forskolin-stimulated adenylyl cyclase activity. However, assessing Gi function using GTP showed marked dissimilarities in that the elevated GTP concentrations expected to occur physiologically were incapable of reversing the stimulation achieved at low concentrations of GTP in membranes from diabetic but not normal animals. The adipocytes of CBA/Ca mice, as do other animal models of insulin resistance, show lesions in adenylyl cyclase regulation, Gi function and G-protein expression.

Early alterations of Gi/Go protein-dependent transductional processes in the retina of diabetic animals

The early alterations of G-protein-dependent transductional mechanisms have been characterized in the retina of alloxan-treated diabetic rats. Five weeks after alloxan injection, pertussis toxin radiolabeling of Gi/Go proteins was markedly reduced in the retina of diabetic animals, suggesting either a reduced expression and/or the presence of some structural modification of these G-protein subtypes. The functional activity of Gs proteins, measured as stimulation of membrane adenylate cyclase by dopamine, did not seem to be impaired at this stage of the pathology; basal adenylate cyclase activity was indeed increased in diabetic rats, consistent with the observed reduction of Gi/Go inhibitory proteins. Such functional alterations of the cAMP producing system were causally related to diabetes induction, since they were reversed by treatment of diabetic animals with insulin. These results suggest that G-protein dependent transduction mechanisms are altered in the retina of diabetic animals, and that a defect of Gi/Go proteins could represent an early transductional damage in the development of diabetic retinopathy.

Attenuation of central alpha 2 adrenergic action in diabetic rats

Molecular components in transmembrane signaling may be dysfunctional in insulin-deficient states. To investigate whether the alpha 2 adrenergic receptor-effector mechanism is functionally altered by insulin deficiency, we determined the hypnotic response to dexmedetomidine [(+)4-[1-(2,3-dimethylphenyl)ethyl]-1H-imidazole, a highly-selective alpha 2 agonist, in streptozotocin-induced diabetic rats. The duration of the loss of righting reflex (sleep time) in response to dexmedetomidine, 0.25 mg.kg-1 IP, was measured in rats pretreated with streptozotocin, 50 mg.kg-1 IP. Dexmedetomidine sleep time was significantly shortened when tested 10 days (-25%), 3 (-29%), 6 (-35%) and 8 (-47%) weeks into the diabetic state. Supplementation of the diabetic rats with insulin normalized alpha 2 responsiveness. Acute hyperglycemia did not reduce dexmedetomidine-induced sleep time. Sleep time was also reduced when dexmedetomidine was administered via the intracerebroventricular (ICV) route at 4 (-21%) and 8 (-29%) weeks after streptozotocin. Thus the central nervous system response to the alpha 2 adrenergic agonist has become attenuated. The mechanism may involve a perturbation of an insulin-sensitive molecular component of the signal transduction pathway responsible for alpha 2 adrenoceptor-mediated anesthetic action.