Evaluation of the selected barrier properties of retinal pigment epithelial cell line ARPE-19 for an in-vitro blood-brain barrier model.

In-vitro models that maintain complex transport mechanisms and structural properties associated with the blood-brain barrier in vivo would be useful in drug permeability and neurotoxicological studies. To evaluate the suitability of a human retinal pigment epithelial cell line for a blood-brain barrier model, we have compared the barrier properties of the human retinal pigment epithelial cell line ARPE-19, the human colonic adenocarcinoma cell line Caco-2, and primary porcine microvessel endothelial cells. The tight junction proteins occludin and ZO-1 were stained immunocytochemically. The paracellular ionic permeability was evaluated by measuring the trans-epithelial or trans-endothelial electric resistance. To evaluate the active transport mechanisms, the existence and the activity of the efflux transporters, P-glycoprotein and multidrug resistance-associated proteins, were studied. All the cell types in this study stained positively for occludin and ZO-1. However, the trans-endothelial electric resistance of ARPE-19 cells was low compared with that of primary porcine microvessel endothelial cell and Caco-2 cells. In addition, both the P-glycoprotein expression and its activity in ARPE-19 cells were low. In conclusion, the barrier properties of the human ARPE-19 cell line were not satisfactory for a blood-brain barrier model. For future studies, it is important to develop a human brain endothelial cell line with expression of the complex in-vivo properties of the blood-brain barrier.

Evaluation of the effects of aluminium, ethanol and their combination on rat brain synaptosomal integral proteins in vitro and after 90-day oral exposure.

The effects of aluminium lactate (Al-lactate) on the rat cerebral synaptosome integral proteins adenosinetriphosphatase (ATPase) and acetylcholinesterase(AChE) were studied in vitro and in vivo. Coexposure with ethanol (EtOH) was studied in both situations. Isolation of synaptosomes was carried out using isoosmotic Percoll gradients. In in vitro experiments, the synaptosomes were exposed to different concentrations of Al-lactate in the incubation mixture. Al-lactate caused decreases in total ATPase and AChE activities concentration dependently. The decrease in ATP activity started at 0.2 mM concentration, and concentration for the 50% decrease of the enzyme activity (EC50 ) was 1.1 mM. The decrease in AChE activity started at 5-10 mM concentration, and the EC50 value was 15.8 mM. Coexposure with ethanol (2 mM) increased the EC50 values similarly in both cases. After 90-day oral exposure of rats to Al-lactate (91.8 mg/kg/day), the serum aluminium level was 0.9-1.3 ptM/l. Coexposure with EtOH(3.0 g/kg/day) did not significantly increase the blood Al(0.7 2.2 pM/l). Aluminium exposure caused a decrease in the blood EtOH concentration (0.6 mM/1) compared with blood EtOH (12.3 mM/1) in the rats exposed to ethanol only. In the rats studied 2 weeks after the Al exposure, the activities of ATPase and AChE were significantly lower than in the rats studied immediately after the exposure. Correspondingly, a significant decrease in AChE activity was found in Al and EtOH-exposed rats, but in the control rats there were no differences between the study groups. Immediately after the 90-day dosing, the exposed rats did not differ significantly from the control rats. Based on the in vitro results, the neural membrane integral proteins ATPase and AChE may be considered as targets for the effects of aluminium and ethanol. Ninety-day in vivo exposure of rats to aluminium caused decrease in ATPase and AChE activities, detectable 2 weeks after the exposure.

Evaluation of the cytotoxicity of selected systemic and intravitreally dosed drugs in the cultures of human retinal pigment epithelial cell line and of pig primary retinal pigment epithelial cells.

The cytotoxicity of the selected systemic and intravitreally dosed drugs tamoxifen, toremifene, chloroquine, 5-fluorouracil, gentamicin and ganciclovir was studied in retinal pigment epithelium (RPE) in vitro. The cytotoxicity was assayed in the human RPE cell line D407 and the pig RPE cell culture using the WST-1 test, which is an assay of cell proliferation and viability. The effects of experimental conditions on the WST-1 test (cell density, serum content in the culture medium, the exposure time) were evaluated. The EC50 values in tamoxifen-treated D407 cells ranged between 6.7 and 8.9 micromol/l, and in pig RPE cells between 10.1 and 12.2 micromol/l, depending on the cell density used. The corresponding values for toremifene were 7.4 to 11.1 micromol/l in D407 cells and 10.0 to 11.6 micromol/l in pig RPE cells. In chloroquine-treated cells, the EC50 values were 110.0 micromol/l for D407 cells and 58.4 micromol/l for pig RPE cells. Gentamicin and ganciclovir did not show any toxicity in micromolar concentrations. The exposure time was a significant factor, especially when the drug did not induce cell death, but was antiproliferative (5-fluorouracil). Serum protected the cells from the toxic effects of the drugs. Both cell cultures were most sensitive to tamoxifen and toremifene, and next to chloroquine. The drug toxicities obtained in the present study were quite similar in both cell types; that is, the pig RPE cells and the human D 407 cell line, despite the differences in, for example, the growth rate and melanin contents of the cell types. Owing to the homeostatic functions important for the whole neuroretina, RPE is an interesting in vitro model for the evaluation of retinal toxicity, but, in addition to the WST-1 test, more specific tests and markers based on the homeostatic functions of the RPE are needed.

Glutamate and calcium uptake in astrocytes after acute lead exposure.

The neurotoxicity of lead is well known, but the exact mechanisms of its toxicity are not solved yet. Disturbances in glutamate homeostasis of neural tissue and interactions of lead with calcium metabolism have been considered as a potential mechanism. In the present study, we evaluate the acute effect of lead exposure by studying glutamate uptake, as well as calcium influx and intracellular lead in cultured rat cerebellar astrocytes. Glutamate uptake and calcium influx were evaluated with radioactive probes by liquid scintillation counting, and intracellular lead was measured by atomic absorption method. Lead increased calcium influx dose dependently, but in an acute 15-min exposure it had no statistically significant effect on glutamate uptake into astrocytes. Calcium decreased at high concentrations but did not hamper lead uptake. Lead uptake rate into astrocytes was high compared to calcium.

Effects of mercuric chloride exposure on the glutamate uptake by cultured retinal pigment epithelial cells.

The cytotoxicity of mercuric chloride and the effects of mercuric chloride on glutamate and calcium uptake and the factors regulating glutamate uptake were studied in retinal pigment epithelium (RPE) cell cultures. RPE cells isolated from pig eyes and human RPE cell line (D407) cells were cultured to confluency and further subcultured according to the test protocol in question. The cytotoxicity caused by 15 min of exposure to mercuric chloride (0.01--1000 microM) was evaluated by WST-1 assay based on the activity of mitochondrial dehydrogenases. [(3)H]Glutamate uptake was measured after the cells were exposed to 0.1--100 microM mercuric chloride and the selected regulators of protein kinase C (PKC) pathway: PKC activator SC10, PKC inhibitor chelerythrine chloride, phospholipase A(2)/C inhibitor manoalide, tyrosine kinase inhibitor lavendustin A, competitive NMDA receptor antagonist AP7 and IP(3) receptor antagonist heparin. Intracellular calcium was monitored with Fluo-3 probe starting immediately after the exposure to 1--1000 microM mercuric chloride. Mercuric chloride showed concentration-dependent effects on cell viability, on glutamate uptake and on intracellular calcium concentration. The results give some support to the concept that glutamate uptake is affected by PKC. The PKC inhibitor chelerythrine chloride decreased glutamate uptake by 25%, but the PKC activator SC10 could partly prevent the inhibitory effect of mercuric chloride. Lavendustin A, manoalide and heparin had smaller, but statistically significant, effects. All these substances act on mediators which can regulate the activity of PKC. However, PKC is not likely to be the only regulator of glutamate uptake. The rise observed in [Ca(2+)](i) may initiate various cellular events during mercury intoxication.

The phagocytosis of rod outer segments is inhibited by selected drugs in retinal pigment epithelial cell cultures.

The effects of tamoxifen, toremifene and chloroquine on the phagocytosis of rod outer segments by retinal pigment epithelium were evaluated in human retinal pigment epithelial cell line D407 and pig retinal pigment epithelial cell culture. Retinal pigment epithelial cells were exposed to different concentrations of tamoxifen (1-20 microM), toremifene (1-20 microM) and chloroquine (1-1000 microM), and challenged with FITC-labeled rod outer segments for 24 hr. The phagocytized (bound and ingested) rod outer segments were measured fluorometrically, and the effect of the drugs on the phagocytosis was determined. The cytotoxicity of the drugs was evaluated by measuring their effects on mitochondrial enzyme activities (WST-1-test). The results showed that the test compounds inhibited the phagocytosis of rod outer segments in both D407 and pig retinal pigment epithelial cells. The phagocytic activity was more sensitive to tamoxifen (EC(50) 7.2 microM for D407 cells and 3.6 microM for pig retinal pigment epithelial cells) and toremifene (EC(50) 6.2 microM and 3.1 microM respectively) than to chloroquine (EC(50) 77.2 microM for D407 cells). The inhibition of rod outer segment phagocytosis in both cell cultures started at lower dose levels of test compounds than the cytotoxicity indicated by the WST-1-test. The experiments were carried out both in serum-free medium and serum-containing medium. Serum seemed to be a critical factor in the medium and caused difficulties in the interpretation of the results.

Piperonyl butoxide potentiates the synaptosome ATPase inhibiting effect of pyrethrin.

Pyrethrins are widely used insecticides in both agriculture and households. In many commercial formulations piperonyl butoxide (PBO) is used with pyrethrins. PBO is a well-known synergist of pyrethrins, used to intensify their effects. One of the cellular targets of pyrethrins is the sodium channel in the membrane. In the present study, the activity of the membrane-bound integral protein ATPase was studied as a biomarker for the membrane effects of pyrethrin and PBO. Cerebral synaptosomes of rat brain were used in the study. The isolation of synaptosomes was performed with the Percoll gradient method. Both total ATPase and Mg2+ activated ATPase were studied by determining inorganic phosphate. Exposure to 0.1-1000 microM of pyrethrin and to 0.4-4000 microM of PBO decreased ATPase activity dose-dependently. The most efficient mixture was the one consisting of one part of pyrethrin and four parts of PBO. The activity of total ATPase decreased 15% in concentrations of 0.1-10 microM pyrethrin, and a 50% decrease was found at 100 microM pyrethrin. The mixture of pyrethrin and PBO caused a 15-60% decrease in the total ATPase activity at 0.1-10 microM pyrethrin and 0.4-40 microM PBO. A 85% decrease was found after exposure to the mixture of 100 microM pyrethrin and 400 microM PBO. PBO alone had no effect at 0.4-40 microM concentrations, but a marked effect was seen at over 40 microM concentrations. The results indicate that PBO is an effective synergist of pyrethrin and that it is very toxic in high concentrations. The results also confirm that neuronal sodium homeostasis is one target of the neurotoxic effect of pyrethroid compounds.

Retinal müller cell culture.

A mini-review is presented of the current techniques for maintaining Müller cells in a culture. Within the retina, Müller cells are the predominant glial cells. These highly specialised cells extend over the entire neural retina. One of the most important of the various physiological functions of Müller cells is to regulate the balance of ions and neurotransmitters in the retina. Disturbance of these regulatory functions may lead to toxic effects on receptor and other neural cells in the neuroretina, and may be a common mechanism of clinical retinal neuropathy. The main excitatory neurotransmitter in the retina is glutamate. Müller cells regulate the amount of glutamate in the synaptic regions of the neural network in the retina. Accumulation of extra glutamate seems to be an important mechanism for initiating pathological changes leading to retinal damage. Many previous in vitro studies on the role of Müller cells in retinal toxicology have been based on the use of morphological and histochemical methods. In cell toxicology studies, it is important to develop culture techniques able to provide more cells for biochemical determinations.

Retinal pigment epithelial cell cultures as a tool for evaluating retinal toxicity in vitro.

This article reviews in vitro testing of retinal toxicity in retinal pigment epithelium (RPE) cell cultures. It is based on the literature on RPE cell cultures and on our recent studies on the retinal toxicity of selected amphiphilic drugs. The RPE plays a major role in maintaining the homeostasis and health of the retina. Various pharmacological agents are known to cause adverse effects in RPE cells. For example, long-term treatment with chloroquine in patients with rheumatoid arthritis has induced retinopathy, and tamoxifen, a drug that is commonly used in the treatment of advanced breast cancer and in the prevention of breast cancer among high-risk women, has been reported to cause retinal changes and impaired vision. During our research, we have developed novel in vitro methods for evaluating the retinal toxicity of xenobiotics. We have used a pig RPE primary culture and a human RPE cell line (D407), which retain epithelial cell characteristics. They form a layer of hexagonal cells with intercellular junctions, and possess a keratin-containing cytoskeleton. They are both good models for determining the retinal cell toxicity of test compounds. Further studies on phagocytic activity, lysosomal enzyme activity and glutamate uptake might generate new methods for the toxicological evaluation of the retinal side-effects of drugs in vitro.

Effects of tamoxifen, toremifene and chloroquine on the lysosomal enzymes in cultured retinal pigment epithelial cells.

Retinal pigment epithelial cells carry out phagocytosis and digestion of material shed from the photoreceptor outer segments. In this process, the integrity of lysosomal enzymes is of major importance. In the present study the effects of tamoxifen, toremifene and chloroquine on the activity of two lysosomal enzymes (cathepsin D and N-acetyl-beta-D-glucosaminidase) in the retinal pigment epithelial cells were studied. Retinal pigment epithelial cells from pig eyes were cultured for two weeks in Dulbecco's Modified Eagle Medium, after which the cells were exposed to 1-40 microM concentrations of tamoxifen citrate, toremifene citrate and chloroquine diphosphate. To eliminate possible medium-borne oestrogenic mechanisms, the test was repeated using phenol red-free medium with charcoal-stripped fetal calf serum. The exposure time was one week, after which the lysosomal enzymes cathepsin D and N-acetyl-beta-glucosaminidase were determined. Cellular injuries were assessed by quantifying the leakage of lactate dehydrogenase into the culture medium. Cathepsin D and N-acetyl-beta-D-glucosaminidase showed different sensitivities to tamoxifen, toremifene and chloroquine. The main lysosomal protease cathepsin D was more sensitive than N-acetyl-beta-D-glucosaminidase to the effects of tamoxifen and toremifene, possibly due to their antioestrogenic properties. The phenol red-free medium with charcoal-stripped serum seemed to make the drugs more effective than the reference medium. Chloroquine had only a minor effect on the lysosomal protease cathepsin D, but a clearer effect could be seen on N-acetyl-beta-glucosaminidase.

Perturbation of artificial and biological membranes by organic compounds of aliphatic, alicyclic and aromatic structure.

Aliphatic, alicyclic and aromatic hydrocarbons interact with biological membranes. Until now little has been known about their mode of interaction with the membrane bilayer and membrane integral proteins with toxic effects to cells. The lipid theory hypothesis explains the toxic effects by the organic solvent-induced disorder in the lipid bilayer, which indirectly affects the function of membrane-embedded proteins. The extent of bilayer perturbations is ascribed to the solvent accumulation in the bilayer, which is related only to the lipophilicity of the molecule, independent of the chemical structure. In this study the fluidizing effects of aliphatic, alicyclic and aromatic hydrocarbons were compared. Membrane fluidity changes were estimated from the pyrene excimer formation, using pyrene and pyrene derivatives to label specifically the localization of solvent molecules in the transverse plane of the bilayer. Liposomal, microsomal and synaptosomal membrane preparations were evaluated because proteins and cholesterol, as natural membrane components, increase the bilayer order and reduce the organic solvent membrane/buffer partition. In the concentration range investigated, only the aromatic solvents disorder the lipid bilayer, with the greatest perturbation in the centre of the bilayer. These results are related to structural properties of the organic solvents investigated.

Effects of selected organic solvents on the astrocyte membrane ATPase in vitro.

1. The present study deals with astrocyte cultures as a model for studying the membrane-mediated central nervous system-depressing effect of organic solvents. 2. The primary astrocyte cultures were prepared from neonatal rat cerebella. The cells were cultured in modified essential medium. The astrocyte membranes isolated from the cultures were exposed to solvents in incubation mixture at different dose levels (3, 6 and 9 mmol/L) for 1 h. The physiologically important integral proteins Na+, K(+)-ATPase and Mg(2+)-ATPase were studied. 3. The aromatic hydrocarbons (benzene, toluene, styrene, xylene and ethylbenzene) inhibited the ATPase activities according to their lipid solubilities. n-Hexane and cyclohexane clearly had less effect than aromatic hydrocarbons, despite their greater lipid solubilities. 4. Astrocytes were shown to be sensitive targets to the effects of organic solvents, measured as the inhibition of the integral enzymes Na+, K(+)-ATPase and Mg(2+)-ATPase.

Astrocyte membrane Na+, K(+)-ATPase and Mg(2+)-ATPase as targets of organic solvent impact.

Astrocytes serve to maintain the proper homeostatic environment for neurons. In these regulations, the astrocyte membrane ATPase plays an important role. In the present study, the astrocyte were cultured in the modified Minimum Essential Medium (MEM), and the isolated cell membranes were exposed to solvents at different concentrations (3, 6 and 9 mM) for one hour. The activities of the membrane-bound Na+, K(+)-ATPase and Mg(2+)-ATPase were determined after exposure to aromatic, alicyclic and aliphatic hydrocarbons. Aromatic hydrocarbons decreased the enzyme activities dose-dependently according to their lipid solubilities. The slight enzyme-inhibiting effect of aliphatic and alicyclic hydrocarbons was not dependent on the lipophilicity of these solvents. Their molecular structure may have determined the solvent impact. In the present study, the activity of ouabain sensitive Na+, K(+)-ATPase in astrocytes was only 20-15% of the total ATPase activity, and in neurons 45-55%. The enzyme inhibition caused by organic solvents may disturb the homeostatic regulatory functions of astrocytes and thus have a toxic effect on the CNS.

Retinal pigment epithelium cell culture as a model for evaluation of the toxicity of tamoxifen and chloroquine.

The retinal pigment epithelium (RPE) removes the outer segments of photoreceptor cells by phagocytosis. We studied the effects of tamoxifen and chloroquine on the activity of the lysosomal enzymes N-acetyl-beta-glucosaminidase and cathepsin D in RPE in vitro to evaluate the possible eye toxicity caused by these drugs. The results show decreases in the activities of lysosomal enzymes after drug exposure. The enzymes tested seemed to be more sensitive to tamoxifen than to chloroquine. A profound decrease in the activities of the lysosomal enzymes only started at concentrations above therapeutic dose levels.

Effects of mercury, methylmercury and aluminium on glial fibrillary acidic protein expression in rat cerebellar astrocyte cultures.

The effects of mercuric chloride, methylmercury chloride and aluminium chloride on glial fibrillary acidic protein (GFAP) expression in primary cerebellar astrocyte cultures were studied. GFAP has been found to be a quantitative marker of neuronal injuries on the central nervous system in vivo. The GFAP content of the astrocytes was examined by sandwich ELISA after exposure for 1 wk to various metal concentrations in the culture medium. Cellular injuries were assessed by quantifying the lactate dehydrogenase (LDH) leakage into the culture medium. The expression of GFAP was found at lower dose levels than the leakage of LDH, indicating that the expression of GFAP could be a more sensitive marker of neurotoxicity than LDH leakage. In morphological examination, staining with monoclonal antibody showed GFAP induction after mercury exposure.

Astrocytes as targets for CNS effects of organic solvents in vitro.

The aim of the study was to evaluate astrocytes in vitro as a model for studies of solvent neurotoxicity. Primary astrocyte cultures were established from newborn rat cerebella. The cells were cultured in the modified Minimum Essential Medium (MEM), and the neural membranes isolated from cultures were exposed to solvents in incubation mixtures containing different solvent concentrations (3, 6, and 9 mM) for one hour. The activity of membrane-bound total ATPase was determined after exposure to aromatic hydrocarbons (benzene, toluene, xylene, styrene and ethylbenzene), and to n-hexane and cyclohexane. The enzyme activities were decreased by aromatic hydrocarbons linearly according to the log dose and in order to the log lipid/water partition coefficients, benzene having the smallest and ethylbenzene the greatest effect in all concentrations studied. Cyclohexane caused much smaller enzyme inhibition (18% of control activity in 9 mM concentration) than ethylbenzene (67% in 9 mM), in spite of very similar partition coefficients. N-hexane had clearly slighter enzyme inhibiting effect than aromatic hydrocarbons, in spite of its markedly greater lipophilicity. In addition to lipophilicity, the structure of solvent molecule seems to be important when considering the CNS toxicity. These results suggest that organic solvents exert their toxic effects on CNS, at least in part, by disturbing ATPase-dependent astrocytic regulatory functions.

Effects of toluene and n-hexane on rat synaptosomal membrane fluidity and integral enzyme activities.

The effects of toluene and n-hexane on rat synaptosomal membrane fluidity and the integral enzymes acetylcholinesterase (AChE) and ATPase were studied in vitro. The synaptosome membranes were isolated in Percoll and sucrose gradients. After adding toluene and n-hexane to the incubation mixture (37 degrees) in 2,4,6 and 8 mM concentrations, the fluidity changes were measured by the lateral pyrene diffusion method from Percoll-isolated membranes, and the ATPase and acetylcholinesterase activities were determined from both synaptosome isolations. Addition of toluene caused a linearly correlated increase of the synaptosomal membrane fluidity and a linear decrease of the AChE activity. The ATPase activity did not decrease linearly but dose-dependently. In contrast to the effects of toluene in vitro, addition of n-hexane in the same concentration range had no comparable influence on membrane fluidity nor on the activities of both integral enzymes despite its even higher lipid/water partition coefficient. Toluene increases synaptosomal membrane fluidity and at the same time inhibits the integral enzymes, probably by disturbing the lipid/protein interaction.