Inhibition of anion and glucose permeabilities by anesthetics in erythrocytes. The mechanisms of action of positively and negatively charged drugs

Sodium valproate inhibits glucose transport and exacerbates Glut1-deficiency in vitro

Anticonvulsant sodium valproate interferes with brain glucose metabolism. The mechanism underlying such metabolic disturbance is unclear. We tested the hypothesis that sodium valproate interferes with cellular glucose transport with a focus on Glut1 since glucose transport across the blood-brain barrier relies on this transporter. Cell types enriched with Glut1 expression including human erythrocytes, human skin fibroblasts, and rat astrocytes were used to study the effects of sodium valproate on glucose transport. Sodium valproate significantly inhibited Glut1 activity in normal and Glut1-deficient erythrocytes by 20%-30%, causing a corresponding reduction of Vmax of glucose transport. Similarly, in primary astrocytes as well as in normal and Glut1-deficient fibroblasts, sodium valproate inhibited glucose transport by 20%-40% (P < 0.05), accompanied by an up to 60% downregulation of GLUT1 mRNA expression (P < 0.05). In conclusion, sodium valproate inhibits glucose transport and exacerbates Glut1 deficiency in vitro. Our findings imply the importance of prudent use of sodium valproate for patients with compromised Glut1 function.

The behavior of the human erythrocyte as an imperfect osmometer: A hypothesis

The human erythrocyte does not behave as a perfect osmometer that is its volume does not change as predicted with the change of the tonicity of the medium, as if there was a fraction of the cell water not participating in the osmotic exchange. A mechanism of control of the erythrocyte shape has been previously proposed in which Band 3 (AE1), the protein anion exchanger of Cl(-) and HCO(3)(-), plays a central role. Specifically, decrease and increase of the ratio of its outward-facing conformation and inward-facing conformation (Band 3(o)/Band 3(i)) contract and relax the membrane skeleton, thus favoring echinocytosis and stomatocytosis, respectively. The equilibrium Band 3(o)/Band 3(i) ratio is determined by the Donnan equilibrium ratio of anions and protons, increasing with it (r=Cl(i)(-)/Cl(o)(-)=HCO 3(i)(-)/HCO 3(o)(-)=H(o)(+)/H(i)(+)). The Donnan ratio is influenced by the erythrocyte transport and metabolic activities. The volume change of the human erythrocyte alters the skeleton conformation as it is accompanied by a change of the membrane curvature. Thus, the mechanism could be a hypothesis for explaining the behavior of the human erythrocyte as an imperfect osmometer since the Donnan ratio controls the Band 3(o)/Band 3(i) ratio which controls the volume by a control of the degree of contraction or relaxation of the skeleton. Predictions made by the hypothesis on the Ponder's coefficient R' values in the presence of sucrose or Band 3 substrates slowly transported as well as on the participation of Band 3 in the osmotic hemolysis appear to be corroborated by previous observations. If the hypothesis was valid, it would follow that there is a pressure gradient across the erythrocyte membrane. The equilibrium volume is antagonistically determined by the Donnan ratio per se and Band 3. Band 3, rather than the ratio of surface-to-volume, primarily controls the osmotic hemolysis.

The effects of phenytoin and its metabolite 5-(4-hydroxyphenyl)-5-phenylhydantoin on cellular glucose transport

Glucose is the principal fuel for brain metabolism and its movement across the blood-brain barrier depends on Glut1. Impaired glucose transport to the brain may have deleterious consequences. For example, Glut1 deficiency syndrome (Glut1DS) is the result of heterozygous loss of function Glut1 mutation leading to energy failure of the brain and subsequently, epileptic encephalopathy. To preserve the integrity of the energy supply to the brain in patients with compromised glucose transport function, consumption of compounds with glucose transport inhibiting properties should be avoided. Phenytoin is a widely used anticonvulsant that affects carbohydrate metabolism. In this study, the hypothesis that phenytoin and its metabolite 5-(4-hydroxyphenyl)-5-phenylhydantoin (HPPH) affect cellular glucose transport was tested. With a focus on Glut1, the effects of phenytoin and HPPH on cellular glucose transport were studied. Glucose uptake assay measuring the zero-trans influx of radioactive-labeled glucose analogues showed that phenytoin and HPPH did not exert immediate effects on erythrocyte Glut1 activity or glucose transport in Hs68 control fibroblasts, Glut1DS primary fibroblasts isolated from two patients, or in rat primary astrocytes. Prolonged exposure to the two compounds could stimulate glucose transport by up to 30-60% over the control level (p <0.05) in Hs68 and Glut1DS fibroblasts as well as in rat astrocytes. The stimulation of glucose transport by HPPH was dose-dependent and accompanied by an up-regulation of GLUT1 mRNA expression (p <0.05). In conclusion, phenytoin and HPPH do not compromise cellular glucose transport. Prolonged exposure to these compounds can modify carbohydrate homeostasis by up-regulating glucose transport in both normal and Glut1DS conditions in vitro.

Flunitrazepam partitioning into natural membranes increases surface curvature and alters cellular morphology

In recent studies, we showed that flunitrazepam (FNTZ) and other benzodiazepines interact with artificial phospholipid membranes locating at the polar head group region, inducing a membrane expansion, reducing the molecular packing and reorganising molecular dipoles. In the present paper we investigated the possibility that those phenomena could be transduced into changes in the curvature of membranes from natural origin. Hence we studied the effect of FNTZ on cellular morphology using human erythrocyte as a natural assay system. Shape changes of erythrocytes were evaluated by light microscopy and expressed as a morphological index (MI). FNTZ induced echinocytosis in a time-dependent manner with MI values significantly higher than those of control (without drug) or DMSO (vehicle) samples. Lidocaine, a local anesthetic known to induce stomatocytosis by incorporating in the inner monolayer, counterbalanced the concentration-dependent FNTZ crenating effects. FNTZ induced protective effects, compared with control and DMSO, against time-dependent hemolysis. Hypotonic-induced hemolysis, was also lowered by FNTZ in a concentration-dependent manner. Both antihemolytic effects suggested a drug-induced membrane expansion allowing a greater increase in cell volume before lysis. In such a complex system like a cell, curvature changes triggered by drug partitioning towards the plasma membrane, might be an indirect effect exerted through modifications of ionic-gradients or by affecting cytoskeleton-membrane linkage. In spite of that, the curvature changes can be interpreted as a mechanism suitable to relieve the tension generated initially by drug incorporation into the bilayer and may be the resultant of the dynamic interactions of many molecular fluxes leading to satisfy the spontaneous membrane curvature.

Inhibition of glucose transport and direct interactions with type 1 facilitative glucose transporter (GLUT-1) by etomidate, ketamine, and propofol: a comparison with barbiturates

Ketamine, etomidate, propofol, and pentobarbital were compared for effects on and interactions with the type 1 facilitative glucose transporter (GLUT-1). Fluxes of radiolabeled hexoses were used to determine the effects of anesthetics on GLUT-1 function. Hypotonic hemolysis of human erythrocytes was used to assess perturbations of membrane integrity. Quenching of intrinsic protein fluorescence was used to assess the direct interactions of anesthetics with purified GLUT-1. Pentobarbital, ketamine, etomidate, and propofol inhibited glucose transport in murine fibroblasts with IC(50) values of 0.8, 1. 6, 0.1, and 0.4 mM, respectively. Pentobarbital, ketamine, etomidate, and propofol also inhibited sugar transport in human erythrocytes. The IC(50) values for pentobarbital and ketamine exhibited substrate dependence for equilibrium exchange but not unidirectional effluxes. This was not observed for etomidate. Propofol did not inhibit equilibrium exchange but did inhibit unidirectional efflux with little substrate dependence. Pentobarbital protected against hemolysis, whereas etomidate and ketamine promoted hemolysis of erythrocytes. Propofol had no effect on membrane integrity. Pentobarbital, ketamine, and etomidate all interacted directly with GLUT-1, with apparent K(d) values of 2.2, 0.8, and 0.5 mM, respectively. Like barbiturates, ketamine, etomidate, and propofol inhibited GLUT-1 at concentrations near to those used pharmacologically. Inhibition of GLUT-1 by these intravenous general anesthetics was complex, exhibiting differential kinetic effects on equilibrium exchange versus unidirectional fluxes and contrasting substrate dependencies. Like barbiturates, ketamine and etomidate bound to GLUT-1 with affinities that paralleled inhibition of glucose transport. As a class, intravenous general anesthetics, in contrast to inhalation anesthetics, inhibit GLUT-1-mediated glucose transport.

Inhibition of chloride/bicarbonate antiports in monkey kidney cells (Vero) by non-steroidal anti-inflammatory drugs

Two chloride/bicarbonate antiport mechanisms are involved in the regulation of cytosolic pH (pHi) in Vero cells, namely Na+-dependent chloride/bicarbonate antiport to normalize pHi after acidification of the cytosol, and Na+-independent Cl-/HCO3- exchange to regulate pHi back to normal after alkalinization of the cytosol. We have tested the effects of the non-steroidal anti-inflammatory drugs acetylsalicylic acid (aspirin), salicylic acid, indomethacin and piroxicam on chloride/bicarbonate exchange and on chloride self exchange in Vero cells. All these drugs were found to inhibit both the Na+-independent and the Na+-linked chloride/bicarbonate antiport in a dose dependent manner. The Na+-independent chloride/bicarbonate antiport was inhibited by lower doses of the drugs than the Na+-linked antiport. The ability of the drugs to inhibit chloride self exchange did not vary much with varying external pH, indicating that the inhibitory effect is due to the anionic form of the drugs. Inhibition occurred immediately upon addition of the drugs, and it was rapidly reversible, indicating that the inhibitory effect is due to direct interaction of the drugs with chloride/bicarbonate antiport, and not to inhibition of prostaglandin synthesis. The relevance of our findings to the clinical effects of the drugs is discussed.

Interaction of amphiphiles with integral membrane proteins. I. Structural destabilization of the anion transport protein of the erythrocyte membrane by fatty acids, fatty alcohols, and fatty amines

The effect of model amphiphiles on the structural stability of the anion exchange protein (band 3) of the human erythrocyte membrane was studied by differential scanning calorimetry. The concentration of membranes, as well as the concentration, head group, alkyl chain length, degree of unsaturation, and double bond configuration of a variety of alkane derivatives were all varied in a systematic way. The depression of the denaturation temperature of band 3 per unit membrane concentration of the amphiphile was then determined in order to quantitate the potency of each drug. Saturated fatty acids of chain length C8 to C24 displayed a monotonic decrease in potency up to C20, followed by a dramatic diminution in potency at C22 and C24. Unsaturation caused only minor increases in the abilities of fatty acids to perturb the anion exchanger, and surprisingly, there was neither a trend for the number of double bonds nor a significant cis-trans distinction. Arachidonic acid, as an exception, was much more effective than any other amphiphile in destabilizing band 3. Fatty acids were about three times more potent than fatty amines and fatty alcohols; however, the enhanced partitioning of the latter into the membrane compensated at certain membrane/buffer ratios for its reduced intrinsic potency. A quantitative model interpretation of the data is presented in an accompanying paper.

The influence of the antiviral drugs amantadine and rimantadine on erythrocyte and platelet membranes and its comparison with that of tetracaine

The influence of the antivirus drugs amantadine and rimantadine and of the anionic analogue 1-adamantane-carboxylic acid on a range of properties of human erythrocyte membrane and of thrombocytes has been compared with the effect of the local anaesthetic tetracaine. At low antiviral drug concentrations the abilities of the drugs to induce erythrocyte shape change and suppress osmotic haemolysis were quantitatively proportional to their clinical potency (rimantadine more effective than amantadine at the same concentration). Rimantadine was also more effective than amantadine in suppressing influenza virus-erythrocyte fusion and viral induced haemolysis. The antiviral drug effects were qualitatively similar to those induced by tetracaine. At the quantitative level, tetracaine was more efficient than the antiviral drugs in inhibiting osmotic haemolysis, virus membrane fusion and platelet aggregation. In the absence of any specificity of the antiviral drug effects we argue for a lysosomotropic mode of drug action, i.e. that the drugs modify virus-membrane interactions by changing the endosomal or lysosomal pH.

The modulatory effect of membrane viscosity on structural and functional properties of the anion exchange protein of human erythrocytes

The sterol content of human erythrocyte membranes was modified by polyvinylpyrrolidone (PVP)-mediated enrichment or depletion of cholesterol (CHL) or incorporation of cholesteryl hemisuccinate (CHS). The effects of these modifications on osmotic fragility and anion exchange protein (AEP) disposition and function were evaluated. CHS enrichment was fast (1 hr, 37 degrees C) and led to a concentration-dependent crenation as well as a decrease in osmotic cell fragility, in parallel with increased membrane microviscosity. CHL caused similar but considerably less marked effects due to slower incorporation rates into membranes. CHS enrichment of cells induced susceptibility of AEP to trypsin, a protease which otherwise does not affect AEP in intact cells. Although transport rates of monosaccharides, nucleosides, and anions were markedly slowed down by CHS enrichment of cells in parallel with increased membrane viscosity, anion transport was the most affected. The temperature profile of anion transport in CHS-enriched cells revealed a 10-kcal/mol increase in the enthalpy of activation relative to normal cells. Anion transport measured in heteroexchange conditions (Cl in--pyruvate out) and (Cl in-sulfate out) was relatively more susceptible to CHS modification than when it was measured in homoexchange conditions (Cl in-Cl out). The results of these measurements indicate that CHS-mediated increase in membrane viscosity affects AEP translocation capacity and transmembrane disposition via changes in lipid compressibility. Specific effects of CHS on AEP function, however, could not be ruled out.

n-Alkanols and halothane inhibit red cell anion transport and increase band 3 conformational change rate

The effects of halothane and n-alkanols on band 3, the anion-exchange protein of the red cell membrane, have been characterized by radioactive sulfate exchange and equilibrium and kinetic binding of a fluorescent anion transport inhibitor, 4,4'-dibenzamido-2,2'-stilbenedisulfonic acid (DBDS), with fluorescence and stopped-flow techniques. Ethanol, butanol, hexanol, heptanol, octanol, and decanol inhibit radioactive sulfate efflux from red blood cells in a dose-dependent manner with an average Hill coefficient of 1.3 +/- 0.1. Over a 10(4)-fold range of buffer concentrations, the calculated membrane alkanol concentrations at which anion transport rates are reduced by 50% are 100-200 mM. At 100-300 mM membrane concentrations, halothane and the n-alkanols increase the apparent rate of DBDS binding to band 3 2-3-fold. Analysis of kinetic and equilibrium DBDS binding data shows that these drugs increase the rate of the DBDS-induced conformational change in the DBDS-band 3 complex. Equilibrium DBDS binding studies reveal differences between the actions of short-chain alkanols (ethanol and butanol) and those of long-chain alkanols (hexanol and longer). Short-chain alkanols reduce the equilibrium affinity of DBDS for band 3, while long-chain alkanols have no effect on equilibrium DBDS binding. The results for halothane and long-chain alkanols suggest a nonspecific, lipid-mediated mechanism of anesthetic action, which may be coupled to protein inactivation by an increase in the rate of protein conformational changes resulting in nonfunctional states. The results for short-chain alkanols indicate that they have the same nonspecific actions as the long-chain alkanols but also have specific effects on the stilbene binding site of band 3.

Kinetic mechanism of chlorpromazine inhibition of erythrocyte 3-O-methylglucose transport

The kinetic mechanism of chlorpromazine inhibition of erythrocyte hexose transport was investigated using the non-metabolizable glucose analog 3-O-methylglucose. It was found that chlorpromazine added to the external medium is a non-competitive inhibitor of both equilibrium exchange and net 3-O-methylglucose transport at pH 7.8, 15 degrees C. The Ki for equilibrium exchange is 76 +/- 21 microM. When net efflux and equilibrium exchange were measured on the same population of cells the equilibrium exchange was 2.5-times the maximum net efflux. The percent reduction of 3-O-methylglucose flux by chlorpromazine is dependent upon chlorpromazine concentration and not 3-O-methylglucose concentration as expected for a non-competitive inhibitor. Equilibrium exchange and net efflux show the same extent of inhibition at each concentration of chlorpromazine evaluated. These results suggest that exchange and net efflux of 3-O-methylglucose in the human erythrocyte may share a common transport system.

Osmotic error in erythrocyte volume determinations

Because of osmotic effects erythrocytes suspended in their native plasma do not have the same volume as the same erythrocytes suspended in Isoton. The discrepancy varies depending upon the osmolality and composition of the native plasma and the length of time the cells have been suspended in Isoton. consequently, the MCV recorded in an electronic particle counter (Coulter in this case) may differ markedly from the true in vivo MCV. A similar error affects the Coulter hematocrit, which is calculated from the MCV and the erythrocyte count. This matrix effect should be taken into account in any laboratory quality assurance program.

The effects of general anaesthetics on glucose and phosphate transport across the membrane of the human erythrocyte

A study has been carried out into the effects of clinically important general anaesthetics, althesin, thiopentone and propanidid, on the transport of glucose and phosphate across the membrane of the human erythrocyte. In general these three substances all inhibit both transport processes but with characteristic inhibition profiles and varying degrees of efficacy. Glucose transport was more sensitive to the hydrophobic steroids and phosphate transport to propanidid. Some hydrophobic agents, e.g., iodobenzene and its azide, were not inhibitory. Removal of cholesterol to some extent augmented the inhibitory effects of most of these compounds (not propanidid). It is argued that these effects are due to the penetration of the anaesthetics into the lipid bilayer and either subsequent disruption of the lipid annuli surrounding the integral membrane proteins and/or direct anaesthetic-protein interaction.

Effects of inhibition of RBC HCO3-/Cl- exchange on CO2 excretion and downstream pH disequilibrium in isolated rat lungs

To determine the effects of the speed of the erythrocyte membrane chloride shift on pulmonary gas transfer, CO2 exchange and the kinetics of pH equilibration were measured with isolated rat lungs perfused with 20% suspensions of human erythrocytes. The lungs were ventilated with room air, and the inflowing perfusate blood gases were similar to those in mixed venous blood in vivo. All experiments were performed at 37 degrees C. Rates of CO2 excretion were determined by measuring the fraction of CO2 in mixed expired gas in a steady state. The time-course of extracellular pH equilibration in the effluent perfusate was measured in a downstream stopflow pH electrode apparatus. CO2 excretion was reduced by approximately 16% when the lungs were perfused with suspensions containing erythrocytes whose HCO-3/Cl- exchange rates was inhibited, compared with CO2 excretion when the lungs were perfused with normal erythrocyte suspensions. A fall of 0.06 in effluent perfusate extracellular pH was noted during perfusion with inhibited erythrocyte suspensions, in contrast to no observable downstream pH change during perfusion with normal erythrocyte suspensions. These results are in close agreement with the predictions of a theoretical model. Our observations suggest that CO2 transfer in capillary beds will be adversely affected in vivo when the rate of the erythrocyte HCO-3/Cl- exchange is abnormally low. Since a number of commonly used drugs are known to inhibit the chloride shift in human erythrocytes, these findings may have important clinical implications, especially in patients with impaired lung function.

Chloride permeability in human red cells: influence of membrane protein rearrangement resulting from ATP depletion and calcium accumulation

A 15% of band 3 protein, the assumed chloride channel, is associated with spectrin, the major peripheral protein of a lattice located at the red cell membrane-cytosol interface, the present study was undertaken to evaluate whether a rearrangement of the lattice modifies the functional property of band 3 protein. Such a rearrangement was modulated by depletion of cell ATP and/or by accumulation of Ca2+ ions within the cell. ATP depletion induces an inhibition of the electroneutral one-for-one chloride exchanges. Neither the modification of red cell morphology due to ATP depletion (discocyte-echinocyte transformation) nor a direct effect of the decrease in internal ATP level can account for this inhibition. On the other hand, it seems reasonable to consider that inhibition is related to the changes in membrane protein organization (formation of heteropolymers) induced by the decrease in ATP level. But it does not appear that the degree of inhibition is modified when this altered assembly of membrane protein is stabilized by disulfide linkages. Accumulation of Ca2+ ions in the cell at a relatively low concentration (10 micro M range) inhibits chloride exchange without apparent modification of the assembly of membrane proteins. This effect of calcium on chloride exchanges is speculatively denoted as a "direct" effect of calcium. Calcium loading of fresh red cells at higher concentrations (500 to 1000 micro M) obtained by use of the ionophore A23187 induces a very strong inhibition of chloride exchanges. In this case, inhibition can be reasonably accounted for by two simultaneous effects of calcium: a "direct" effect which explains half of the inhibition and an "indirect effect due to the formation of membrane protein complexes stabilized by covalent crosslinkages (activation by Ca2+ ions of a transglutaminase). It is interesting to note that intracellular calcium, whatever the level, inhibits electroneutral exchanges of chloride but increases net chloride movements.