Simulated microgravity enhances cerebral artery vasoconstriction and vascular resistance through endothelial nitric oxide mechanism

Growing evidence suggests that cardiac enkephalins and their receptors are involved in ischemic preconditioning (IPC). Because there is no evidence for vesicular storage of small bioactive enkephalins in the heart, studies were designed to test the hypothesis that ischemia depletes cardiac enkephalins and that IPC preserves the same enkephalins by accelerating their processing from the larger proenkephalin precursor (PEP) pool. The precursors and two bioactive representatives, Met-enkephalin (ME) and Met-enkephalin-Arg-Phe (MEAP), were separated by size-exclusion chromatography and quantified by radioimmunoassay. Isolated perfused rat hearts were prepared and exposed to global ischemia. After 30 min of global ischemia and 40 min of reflow, the PEP pool was reduced (from 17.99 +/- 1.52 to 14.20 +/- 2.38 pmol/g wet wt), MEAP increased by 53%, and ME declined by 68%. The sum of the two smaller peptides was unchanged (9.78 +/- 0.83 vs. 9.33 +/- 2.81). Thus the total enkephalin peptide content was not altered (27.77 +/- 1.69 vs. 24.10 +/- 4.75). Peptide distribution after ischemia and reflow was also unaltered by pretreatment with peptidase inhibitors. However, when the hearts were preconditioned, the PEP pool remained significantly lower and both of the bioactive peptides, MEAP and ME, were elevated (+49% and +86%, respectively). The decline in the PEP pool was prevented by peptidase inhibition and the rise in MEAP was exaggerated. In separate protocols, synthetic enkephalins (ME, MEAP, and Leu-enkephalin) were added to the coronary inflow before 30 min of global ischemia and throughout the subsequent reflow. The added enkephalins (10(-8) M) had no inotropic effect on baseline function but completely prevented the mechanical dysfunction observed in untreated controls during reflow. Thus IPC appears to increase available bioactive enkephalins (MEAP + ME) within the heart by enhancing synthesis of precursors and their subsequent processing from the PEP pool.     

Tissue Content of Opioid Peptides in the Myenteric Plexus-Longitudinal Muscle of Guinea-Pig Small Intestine

We have developed a method that is based on two HPLC systems and permits the separation of endogenous opioid peptides in tissue extracts. The individual peptides are bioassayed on the mouse isolated vas deferens; naloxone (100 nM) ensures opioid specificity. In the myenteric plexus-longitudinal muscle preparation of the guinea-pig small intestine, the tissue content of prodynorphin-derived peptides is lower than those of proenkephalin-derived peptides. No beta-endorphin was detected. Of the prodynorphin fragments, alpha-neoendorphin, beta-neoendorphin, dynorphin A(1-8), and dynorphin B are present in equimolar concentrations (12-15 pmol/g) whereas the tissue content of dynorphin A is only 0.8 pmol/g. Processing of proenkephalin leads to at least six opioid peptides. The tissue contents of [Leu5]enkephalin, [Met5]enkephalyl-Arg-Gly-Leu, and [Met5]enkephalyl-Arg-Phe are 90-100 pmol/g and the content of [Met5]enkephalin is 405 pmol/g. BAM-18 and [Met5]enkephalyl-Arg-Arg-Val-NH2 are present in much lower concentrations, 24 and 5 pmol/g, respectively. Although present in low amounts, BAM-18 and [Met5]-enkephalyl-Arg-Arg-Val-NH2 have high affinity for the mu-opioid binding site and to a lesser extent for the kappa-site; this binding profile differs from that of the other proenkephalin fragments all of which have high affinities for the mu- and delta-sites.     

Reserpine Increases Chromaffin Cell Enkephalin Stores Without a Concomitant Decrease in Other Proenkephalin-Derived Peptides

Reserpine increases the levels of enkephalins in adrenal medullary chromaffin cells; however, the origin of the newly apparent pentapeptides has been the subject of debate, because no increase in the levels of proenkephalin mRNA has been observed. The present study was performed for determining if the reserpine-induced increase in context of enkephalins was derived from processing of preexisting fragments of proenkephalin. Bovine chromaffin cell enkephalins and larger enkephalin-containing peptides were separated by reversed-phase HPLC and identified by approximate molecular weight, elution with peptide standards, and enkephalin sequences contained. Treatment of the cells with reserpine increased the levels of enkephalins and of enkephalin-containing peptides of up to approximately 3 kilo-daltons without reducing the levels of larger enkephalin-containing peptides. Similar results were obtained with another catecholamine-depleting drug, tetrabenazine. In contrast, treatment of chromaffin cells with theophylline or forskolin increased the levels of both enkephalins and enkephalin-containing peptides of all sizes. The results suggest that new synthesis of proenkephalin is required for the effects of reserpine, although proenkephalin processing is also altered by this drug.     

Proenkephalin expression and enkephalin release are widely observed in non-neuronal tissues

Enkephalins are opioid peptides that are found at high levels in the brain and endocrine tissues. Studies have shown that enkephalins play an important role in behavior, pain, cardiac function, cellular growth, immunity, and ischemic tolerance. Our global hypothesis is that enkephalins are released from non-neuronal tissues in response to brief ischemia or exercise, and that this release contributes to cardioprotection. To identify tissues that could serve as potential sources of enkephalins, we used real-time PCR, Western blot analysis, ELISA, immunofluorescence microscopy, and ex vivo models of enkephalin release. We found widespread expression of preproenkephalin (pPENK) mRNA and production of the enkephalin precursor protein proenkephalin (PENK) in rat and mouse tissues, as well as in tissues and cells from humans and pigs. Immunofluorescence microscopy with anti-enkephalin antisera demonstrated immunoreactivity in rat tissues, including heart and skeletal muscle myocytes, intestinal and kidney epithelium, and intestinal smooth muscle cells. Finally, isolated tissue studies showed that heart, skeletal muscle, and intestine released enkephalins ex vivo. Together our studies indicate that multiple non-neuronal tissues produce PENK and release enkephalins. These data support the hypothesis that non-neuronal tissues could play a role in both local and systemic enkephalin-mediated effects.     

Biochemistry of the enkephalins and enkephalin-containing peptides

The enkephalins are present in many tissues not only as the free pentapeptides, but also as internal sequences in larger polypeptides of varying size. Fourteen enkephalincontaining peptides (EC peptides) from beef adrenal medulla were isolated and sequenced, and the presence of a protein that contained several [Met]enkephalin sequences and one of [Leu]enkephalin was demonstrated. Because the latter was assumed to represent the gene product, it was named proenkephalin. Sequence data from the EC peptides made possible the synthesis of a polynucleotide probe with essentially no degeneracy and permitted the cloning of a partial proenkephalin cDNA. The complete structure of proenkephalin was deduced from both peptide and cDNA sequencing data. Proenkephalin is now known to be one of three enkephalin-containing gene products, each of which gives rise to many physiologically active peptides.     

Effects of ischemia on the metabolism of cardiac enkephalins

The effect of ischemia on cardiac Leucine enkephalin (Leu-enk) content, degradation and coronary release was studied in the isolated perfused hearts of male Sprague Dawley rats. Hearts were electrically stimulated at 180 beats/min. Cardiac Leu-enk concentrations were increased when hearts were perfused (635 +/- 41 vs 301 +/- 60 fmol/g in control non-perfused hearts,) or during ischemia-reperfusion (520 +/- 78 vs 277 +/- 42 fmol/g in heart submitted to ischemia alone). The quantity of leucine-enkephalin released by the heart during perfusion was four times higher than the initial content measured in the heart tissue. The rate of this release was the same throughout the experiment (25.9 +/- 2.9 fmol/min/g during perfusion vs. 19.2 +/- 1.6 during ischemia-reperfusion). These findings suggested that cardiac enkephalin metabolism is regulated by cardiac events. In fact, enzymes involved in enkephalin degradation were decreased during perfusion (39%) and increased during ischemia (50%). The decrease in the enzyme activity during coronary perfusion depended on a reduced activity in the membrane fraction only while membrane and soluble fractions were interested in the increased enzyme activity after ischemia. Ischemia-reperfusion induced a larger release of Leu-enk than perfusion without ischemia. In view of the protective actions of enkephalin peptides against oxidative stress, we can infer from our results an implication of Leu-enk in ischemia-reperfusion and thus eventually in preconditioning phenomenon.     

Reserpine-induced alterations in the processing of proenkephalin in cultured chromaffin cells. Increased amidation

We have used antisera directed towards eight different portions of the proenkephalin molecule to examine the processing rates and patterns of proenkephalin-derived peptides in chromaffin cell cultures in the presence and absence of reserpine. Reserpine treatment produced profound effects on the molecular weight profile of nearly all enkephalin-containing peptides. Increased production of low molecular weight immunoreactive [Met5]enkephalin, [Leu5]enkephalin, [Met5]enkephalin-Arg6-Gly7-Leu8, and [Met5]enkephalin-Arg6-Phe7 was observed in reserpine-treated cultures; immunoreactivity corresponding to several intermediate sized enkephalin-containing peptides such as Peptide B and the high molecular weight [Met5]enkephalin-Arg6-Gly7-Leu8 immunoreactive peptide was decreased. The production of two amidated opioid peptides, amidorphin and metorphamide, was greatly accelerated in the presence of reserpine. The increased levels of low molecular weight enkephalins could not be accounted for by assuming decreased basal release. These results indicate that reserpine treatment is able to increase the extent of post-translational processing of proenkephalin within chromaffin cells.     

Mitogenic activation of the human lymphocytes induce the release of proenkephalin derived peptides

Several reports indicate that enkephalins participate in lymphocyte proliferation and several events of the immune response. It has been proposed that peptides involved in these processes may originate in the nervous system or endocrine glands. We have found that human peripheral blood lymphocytes (PBL) activated with a mitogenic agent contain and release proenkephalin derived peptides. The kinetics of met-enkephalin and cryptic products of proenkephalin in PBL activated with phytohemaglutinin (PHA) were studied. Peptides were released to the supernatant of stimulated PBL, reaching the highest values after 18 to 24 hours. The material secreted corresponds to high, intermediate and low molecular weight peptides derived from proenkephalin, displaying met-enkephalin and synenkephalin (proenkephalin 1-70) immunoreactivity. Therefore, an intrinsic lymphocytic proenkephalin system is induced by PHA and may play an important role in the regulation of the immune response.     

Downregulation of Na+-K+-ATPase pumps in skeletal muscle with training in normobaric hypoxia.

To investigate the effects of training in normoxia vs. training in normobaric hypoxia (fraction of inspired O2 = 20.9 vs. 13.5%, respectively) on the regulation of Na+-K+-ATPase pump concentration in skeletal muscle (vastus lateralis), 9 untrained men, ranging in age from 19 to 25 yr, underwent 8 wk of cycle training. The training consisted of both prolonged and intermittent single leg exercise for both normoxia (N) and hypoxia (H) during a single session (a similar work output for each leg) and was performed 3 times/wk. Na+-K+-ATPase concentration was 326 +/- 17 (SE) pmol/g wet wt before training (Control), increased by 14% with N (371 +/- 18 pmol/g wet wt; P < 0.05), and decreased by 14% with H (282 +/- 20 pmol/g wet wt; P < 0.05). The maximal activity of citrate synthase, selected as a measure of mitochondrial potential, showed greater increases (P < 0.05) with H (1.22 +/- 0.10 mmol x h-1 x g wet wt-1; 70%; P < 0.05) than with N (0.99 +/- 0.10 mmol x h-1 x g wet wt-1; 51%; P < 0.05) compared with pretraining (0.658 +/- 0.09 mmol x h-1 x g wet wt-1). These results demonstrate that normobaric hypoxia induced during exercise training represents a potent stimulus for the upregulation in mitochondrial potential while at the same time promoting a downregulation in Na+-K+-ATPase pump expression. In contrast, normoxic training stimulates increases in both mitochondrial potential and Na+-K+-ATPase concentration.     

Posttranslational Processing of Proenkephalin in SK-N-MC Cells: Evidence for Phosphorylation

SK-N-MC cells have recently been shown to be a rich source of proenkephalin and/or the proenkephalin-derived peptide, peptide B. We have investigated the synthesis and the posttranslational processing of proenkephalin in these cells. SK-N-MC cells retain very little of the proenkephalin synthesized; greater than 99% of the immunoreactive enkephalin synthesized within a 48-h period is secreted into the medium rather than contained intracellularly. When medium samples were subjected to gel filtration and assayed for the various enkephalins present within proenkephalin, only two major molecular-weight classes of peptides, with molecular weights and immunoreactive profiles consistent with those of proenkephalin and the 3.6-kDa carboxyl-terminal fragment peptide B, were observed. The proenkephalin-like peptide present in medium samples was shown by western blot procedures to consist of a 32-kDa protein with a slight amount of a higher-molecular-weight immunoreactive component above it. Only proenkephalin-sized peptides were present within cell extracts. Radiolabeled proenkephalin added to cell cultures was also cleaved to products similarly sized to those found in medium extracts; radiolabeled proenkephalin incubated in the absence of cells was not cleaved. Cleavage of exogenous proenkephalin thus probably at least partially occurs following secretion. Cell radiolabeling experiments with [32P]orthophosphate demonstrated that SK-N-MC proenkephalin is phosphorylated. Microheterogeneity of proenkephalin was also observed using isoelectric focusing coupled with western blotting. Our results suggest that the SK-N-MC cell line represents a useful model to study the earliest steps of the posttranslational processing of human proenkephalin in a neuronal cell type.     

Effects of enkephalins on perfusion pressure in isolated hindlimb preparations

A series of studies were conducted to determine the effects of leucine-(leu-) enkephalin and methionine-(met-) enkephalin on perfusion pressure. These experiments utilized isolated perfused femoral arterial preparations in pentobarbital-anesthetized cats. The enkephalins were administered intraarterially into the femoral artery and changes in perfusion pressure recorded. Leu-enkephalin in doses of 1 μg to 320 μg produced significant dose-dependent decreases in perfusion pressure (4.0 ± 1.3% with 1 μg to 19.1 ± 2.1% with 320 μg). Similar declines in perfusion pressure (5.2 ± 2.4% with 1 μg to 21.7 ± 4.1% with 320 μg) were observed following the administration of met-enkephalin. Pretreatment with naloxone (3 mg/kg) antagonized the effects of both enkephalins. Diphenhydramine (2 mg/kg) effectively antagonized the leu-enkephalin elicited decline in perfusion pressure but blocked the effects of met-enkephalin only at lower agonist doses. Propranolol treatment (4 mg/kg) did not alter the pressure responses to either enkephalin. The results of the study show that intraarterially administered enkephalins exert a vasodilatory effect on vasculature in skeletal muscle which may be direct, indirect or both. The differential antagonism of the effects of the two enkephalins suggest that the two opioids act through different receptors or multiple receptors.     

Evolution of Proenkephalin and Prodynorphin

SYNOPSIS. Since the discovery of the enkephalins in 1975, threeseparate families of opioid peptides have been identified, proopiomelanocortin,proenkephalin, and prodynorphin. The lattertwo have multiplecopies of the enkephalin sequence in larger enkephalin containingpolypeptides (ECPs). The amino acid and nucleotide sequencesof these two proteins from human as well as other species havebeen determined. A comparison of each of the proteins betweenspecies indicates a number of highly conserved regions in notonly the bioactive peptides but at other locations as well.A number of common features between the two proteins suggesteda common ancestral protein. Using computeraided metric analysis,it is possible to align their DNA segments so thatthe cysteineresidues at the amino end are aligned, as well as the enkephalinsequences at the carboxyl end. The exact alignment of the centralregions is currently being studied. Metric analysis thus providesstrong evidence that proenkephalin and prodynorphin divergedfrom a single ancestral protein.     

Microdialysis and mass spectrometric monitoring of dopamine and enkephalins in the globus pallidus reveal reciprocal interactions that regulate movement

Pallidal dopamine, GABA and the endogenous opioid peptides enkephalins have independently been shown to be important controllers of sensorimotor processes. Using in vivo microdialysis coupled to liquid chromatography-mass spectrometry and a behavioral assay, we explored the interaction between these three neurotransmitters in the rat globus pallidus. Amphetamine (3 mg/kg i.p.) evoked an increase in dopamine, GABA and methionine/leucine enkephalin. Local perfusion of the dopamine D(1) receptor antagonist SCH 23390 (100 μM) fully prevented amphetamine stimulated enkephalin and GABA release in the globus pallidus and greatly suppressed hyperlocomotion. In contrast, the dopamine D(2) receptor antagonist raclopride (100 μM) had only minimal effects suggesting a greater role for pallidal D(1) over D(2) receptors in the regulation of movement. Under basal conditions, opioid receptor blockade by naloxone perfusion (10 μM) in the globus pallidus stimulated GABA and inhibited dopamine release. Amphetamine-stimulated dopamine release and locomotor activation were attenuated by naloxone perfusion with no effect on GABA. These findings demonstrate a functional relationship between pallidal dopamine, GABA and enkephalin systems in the control of locomotor behavior under basal and stimulated conditions. Moreover, these findings demonstrate the usefulness of liquid chromatography-mass spectrometry as an analytical tool when coupled to in vivo microdialysis.     

Enkephalin peptides in fetal sheep, changes with gestation, origin and production by the placenta.

Enkephalin-containing peptides have been followed in the circulation of fetal sheep between 118-143 days gestation. Using a combination of radioimmunoassay and hplc met5-enkephalin was found in the concentration range 60-500 pg/ml and proenkephalins containing met5-enkephalin had a concentration of 150-4000 pg/ml. The concentration of both increased towards term. The sources of the enkephalin peptides was investigated by measurement of differences across the umbilical circulation and by studying the effects of fetal adrenal demedullation and chemical sympathectomy. The placenta showed a continuous net output of enkephalin peptides which increased close to term. This placental output was increased sharply by reduction of uterine blood flow either using compression of the uterine artery or through infusion of adrenaline at 35 micrograms/min into the maternal circulation. Maternal hypoxia caused by breathing 9% O2 plus 3% CO2 also increased fetal plasma enkephalin levels, although not output from the placenta. Adrenal demedullation, particularly if accompanied by chemical sympathectomy depressed fetal plasma enkephalin concentrations and sharply suppressed the fetal peptide responses to maternal hypoxia. It is concluded that the placenta and the fetal adrenal are important sources of met5-enkephalin-containing peptides in the fetal circulation. The placental production appears to be closely tied to changes in uterine perfusion and adrenal output changes in response to fetal oxygenation.     

Haloperidol-Induced Increase in Striatal Concentration of the Tripeptide, Tyr-Gly-Gly, Provides an Index of Increased Enkephalin Release In Vivo

A sensitive and specific radioimmunoassay has been developed for the tripeptide, Tyr-Gly-Gly, which has been shown previously to be an extraneuronal metabolite of opioid peptides derived from proenkephalin A. Using this assay, we found a regional variation in Tyr-Gly-Gly immunoreactivity in rat brain, with highest levels in striatum and lowest in cerebral cortex. Intracerebroventricular administration of the aminopeptidase inhibitor, bestatin; produced a threefold increase in Tyr-Gly-Gly immunoreactivity in rat striatum, whereas thiorphan, an enkephalinase inhibitor, produced a 45% reduction in striatal Tyr-Gly-Gly immunoreactivity. These data suggest that the tripeptide, Tyr-Gly-Gly, is in a dynamic state in the brain, and provide further support for the hypothesis that its concentration in specific brain areas may reflect the release of endogenous enkephalins in these brain areas. Further confirmation of the validity of measurements of brain Tyr-Gly-Gly as indices of enkephalin release under conditions of altered neuronal activity was provided by our demonstration that chronic dopamine receptor blockade with haloperidol increased striatal concentrations of both Met-enkephalin and Tyr-Gly-Gly.     

Types of opioid receptors: relation to antinociception

The endogenous opioid peptides are derived from three large precursors. Pro-opiocortin and proenkephalin yield [Met]enkephalin, carboxy-extended [Met]enkephalins and [Leu]enkephalin. The fragments of prodynorphin are all carboxy-extended [Leu]enkephalins. Three approaches are of importance for an analysis of the physiological functions of the different endogenous opioid peptides. First, since these peptides interact with more than one of the mu-, delta- and kappa-binding sites and thus with their receptors, it is necessary to synthesize peptides or non-peptides, which bind to only one of the sites. As far as narcotic analgesics are concerned, morphine fulfils these conditions since it interacts almost exclusively with the mu-receptor. Secondly, antagonists are required that are selective for only one of the opioid receptors, even when used in high concentrations. Finally, it is important to find circumscribed areas in the nervous system that possess only one type of opioid receptor. It is now known that in the rabbit cerebellum the opioid receptors are almost exclusively of the mu-type whereas in the guinea-pig cerebellum they are almost exclusively of the kappa-type.     

Distribution and localization of neurokinin A-like immunoreactivity and neurokinin B-like immunoreactivity in rat peripheral tissue

Using specific radioimmunoassays and immunocytochemistry for neurokinin A (NKA) and neurokinin B (NKB), distribution and localization of these peptides in rat peripheral tissues were studied. NKA-like immunoreactivity (NKA-LI) was present in highest levels of 15.7–23.9 pmol/g wet wt. and NKB-like immunoreactivity (NKB-LI) was in levels of 0.33–0.67 pmol/g wet wt., throughout the gastrointestinal tract involving stomach, duodenum, jejunum, ileum and colon. Immunocytochemical analysis of gastrointestinal tract revealed that NKA-LI and NKB-LI localized in ganglia of both the submucosal and myenteric plexuses as well as varicose neurons in the mucosa and the muscle layer of the small and large intestine. On the other hand, high levels of NKB-LI were observed in oesophagus (0.83 ± 0.08 pmol/g wet wt.), adrenal (1.02 ± 0.21), head of pancreas (0.73 ± 0.06) and kidney (0.98 ± 0.05).

The present study shows the difference of localization of NKA-LI and NKB-LI in peripheral tissues and suggests that NKB may have some physiological role differing from that of NKA in peripheral tissues.     

Quantitation of ryanodine receptor of rabbit skeletal muscle, heart and brain

The total number of high-affinity ryanodine receptor (RyR) binding sites present in skeletal and cardiac muscle and in brain tissue of the rabbit was determined by [3H]ryanodine binding to subfractions obtained by differential centrifugation of homogenates prepared in a low-ionic strength medium, containing 0.5% Chaps. In all three tissues at least 80% of [3H]ryanodine binding was recovered in the total membrane (TM) fraction obtained by centrifuging between 650 g for 10 min and 120,000 x g for 90 min. Skeletal muscle displayed higher contents of high-affinity RyR sites (about 49 pmol/g wet wt) than heart and brain (about 12 pmol and 3.5 pmol/g wet wt, respectively). The affinity for ryanodine, as well as the affinity for Ca2+, in the absence or presence of Ca2(+)-releasing drugs (caffeine and doxorubicin) of TM from skeletal muscle, were found to be identical to those of purified terminal cisternae. As low as 1 g of tissue was sufficient to perform several experiments.     

Aging is associated with myocardial insulin resistance and mitochondrial dysfunction

Aging is associated with insulin resistance, often attributable to obesity and inactivity. Recent evidence suggests that skeletal muscle insulin resistance in aging is associated with mitochondrial alterations. Whether this is true of the senescent myocardium is unknown. Twelve young (Y, 4 years old) and 12 old (O, 11 years old) dogs, matched for body mass, were instrumented with left-ventricular pressure gauges, aortic and coronary sinus catheters, and flow probes on left circumflex artery. Before surgery, all dogs participated in a 6-wk exercise program. Dogs underwent measurements of hemodynamics and plasma substrates before and during a 2-h hyperinsulinemic-euglycemic clamp to measure whole body and myocardial glucose and nonesterified fatty acid uptake. Following the protocol, myocardial and skeletal samples were obtained to measure components of the insulin-signaling cascade and mitochondrial structure. There was no difference in plasma glucose (Y, 90 +/- 4 mg/dl; O, 87 +/- 4 mg/dl), but old dogs had higher (P < 0.02) nonesterified fatty acids (Y, 384 +/- 48 micromol/l; O, 952 +/- 97 micromol/l) and plasma insulin (Y, 39 +/- 11 pmol/l; O, 108 +/- 18 pmol/l). Old dogs had impaired total body glucose disposition (Y, 11.5 +/- 1 mg x kg(-1) x min(-1); O, 8.0 +/- 0.5 mg x kg(-1) x min(-1); P < 0.05) and insulin-stimulated myocardial glucose uptake (Y, 3.5 +/- 0.3 mg x min(-1) x g(-1); O, 1.8 +/- 0.3 mg x min(-1) x g(-1); P < 0.05). The impaired insulin action was associated with altered insulin signaling and glucose transporter (GLUT4) translocation. There were myocardial mitochondrial structural changes observed in association with decreased expression of uncoupling protein-3. Aging is associated with both whole body and myocardial insulin resistance, independent of obesity and inactivity, but involving altered mitochondrial structure and impaired cellular insulin action.