Studies on the neurotoxicity of 6,7-dihydroxy-1-methyl-1,2,3,4-tetrahydroisoquinoline (salsolinol) in SH-SY5Y cells

We have studied the hypothesis that 6,7-dihydroxy-1-methyl-1,2,3,4-tetrahydroisoquinoline (salsolinol) is neurotoxic. Salsolinol induced a significant time and dose related inhibition of 3[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide; thiazoyl blue (MTT) reduction, and increased lactate dehydrogenase release (LDH) release from human SH-SY5Y neuroblastoma cells, at concentrations within the range of 1-methyl-4-phenylpyridinium (MPP+) cytotoxicity, in vitro. Cytotoxicity was not inhibited by the addition of antioxidants, monoamine oxidase inhibitors or imipramine. In confluent monolayers, salsolinol stimulated catecholamine uptake with EC50 values of 17 muM and 11 muM, for noradrenaline and dopamine, respectively. Conversely, at concentrations above 100 muM, salsolinol inhibited the uptake of noradrenaline and dopamine, with IC50 values of 411 muM and 379 muM, respectively. The inhibition of catecholamine uptake corresponded to the increase displacement of [3H]nisoxetine from the uptake 1 site by salsolinol, as the Ki (353 muM) for displacement was similar to the IC50 (411 and 379 muM) for uptake. Salsolinol stimulated catecholamine uptake does not involve the uptake recognition site, or elevation of cAMP, cGMP, or inhibition of protein kinase C. Salsolinol also inhibited both carbachol (1 mM) and K+ (100 mM, Na+ adjusted) evoked released of noradrenaline from SH-SY5Y cells, with IC50 values of 500 muM and 120 muM, respectively. In conclusion, salsolinol appears to be cytotoxic to SH-SY5Y cells, via a mechanism that does not require uptake 1, bioactivation by monoamine oxidase, or membrane based free radical damage. The effects of salsolinol on catecholamine uptake, and the mechanism of toxicity require further investigation.

The neurotoxicity of acrylamide

The experiments described in this thesis compared conventional methods of screening for neurotoxins with potential electrophysiological and pharmacological tests in an attempt to improve the sensitivity of detection of progressive distal neuropathy. Adult male albino mice were dosed orally with the neurotoxicant acylamide and subjected to a test of limb strength and co-ordination and a functional observational battery. These methods established a no observable effect level of 10 mg/kg. A dose of 200 mg/kg resulted in abnormalities of gait and reduced limb strength and/or co-ordination. Analysis of the in vitro 'jitter' of the latency of trains of action potentials evoked at a frequency of 30 Hz in the mouse phrenic nerve/hemidiaphragm preparation showed this technique to be unsuitable for detection of the early phases of acrylamide induced peripheral neuropathy (l00 mg/kg). The evoked and spontaneous twitch responses of the hemidiaphragm preparation following in vitro exposure to the organophosphorous anticholinesterase compound ecothiopate were altered by in vivo pre treatment with acrylamide. Acrylamide caused an increase in the time course of the potentiation of stimulated twitches and a decrease in the maximum potentiation. Spontaneous twitches were reduced in amplitude and frequency. These effects occurred at an acrylamide dose level insufficient to cause clinical signs of neuropathy. Investigations into the mechanisms underlying these observations yielded the following observations. Analysis of miniature endplate potentials at this dose level indicated prolongation of the life of acetylcholine in the synaptic cleft but the implied decrease in cholinesterase activity could not be demonstrated biochemically or histologically. The electrical excitability of the nerve terminal region of phrenic motor nerves was reduced following acrylamide although a possible compromise of antidromic action potential conduction could not be confirmed. There was no histopathological evidence of neuropathy at this dose level. Further exploration of this phenomenon is desirable in order to ascertain whether the effect is specific to acrylamide and/or ecothiopate and to elucidate the mechanisms behind these novel observations.

Development of a post-mitotic, neuronal-astrocytic cell system, for the prediction of acute neurotoxicity in humans

The human NT2.D1 cell line was differentiated to form both a 1:2 co-culture of post-mitotic NT2 neuronal and NT2 astrocytic (NT2.N/A) cells and a pure NT2.N culture. The respective sensitivities to several test chemicals of the NT2.N/A, the NT2.N, and the NT2.D1 cells were evaluated and compared with the CCF-STTG1 astrocytoma cell line, using a combination of basal cytotoxicity and biochemical endpoints. Using the MTT assay, the basal cytotoxicity data estimated the comparative toxicities of the test chemicals (chronic neurotoxin 2,5-hexanedione, cytotoxins 2,3- and 3,4-hexanedione and acute neurotoxins tributyltin- and trimethyltin- chloride) and also provided the non-cytotoxic concentration-range for each compound. Biochemical endpoints examined over the non-cytotoxic range included assays for ATP levels, oxidative status (H2O2 and GSH levels) and caspase-3 levels as an indicator of apoptosis. although the endpoints did not demonstrate the known neurotoxicants to be consistently more toxic to the cell systems with the greatest number of neuronal properties, the NT2 astrocytes appeared to contribute positively to NT2 neuronal health following exposure to all the test chemicals. The NT2.N/A co-culture generally maintained superior ATP and GSH levels and reduced H2O2 levels in comparison with the NT2.N mono-culture. In addition, the pure NT2.N culture showed a significantly lower level of caspase-3 activation compared with the co-culture, suggesting NT2 astrocytes may be important in modulating the mode of cell death following toxic insult. Overall, these studies provide evidence that an in vitro integrated population of post-mitotic human neurons and astrocytes may offer significant relevance to the human in vivo heterogeneous nervous system, when initially screening compounds for acute neurotoxic potential.

Evaluating the potential neurotoxicity of hexanedione isomers:an in vitro approach

2,5-hexanedione (2,5HD) is the neurotoxic metabolite of the aliphatic hydrocarbon n-Hexane. The isomers, 2,3-hexanedione (2,3HD) and 3,4-hexanedione (3,4HD) are used as food additives. Although the neurotoxicity of 2,5HD is well established, there are no human data of the possible toxicity of the 2,3- and 3,4- isomers. MTT and flow cytometry were utilised to determine the cytotoxicity of hexanedione isomers in neuroblastoma cells. The neuroblastoma cell lines SK-N-SH and SH-SY5Y are sufficiently neuron-like to provide preliminary assessment of the neurotoxic potential of these isomers, in comparison with toxicity towards human non-neuronal cells. Initial studies showed that 2,5HD was the least toxic in all cell lines at all times (4, 24 and 48h). Although considerably lower than for 2,5HD, in general the IC50s for the α isomers were not significantly different from each other and, besides 4h exposure, the SH-SY5Y cells were significantly more sensitive to 2,3HD and 3,4HD than the SK-N-SH cells. All three isomers caused varying degrees of apoptosis in the neuroblastoma lines, with 3,4HD more potent than 2,3HD. Flow cytometry highlighted cell cycle arrest indicative of DNA damage with 2,3- and 3,4HD. The toxicity of the isomers towards 3 non-neuronal cell lines (MCF7, HepG2 and CaCo-2) was assessed by MTT assay. All 3 hexanedione isomers proved to be cytotoxic in all non-neuronal cell lines at all time points. These data suggest cytotoxicity of 2,3- and 3,4HD (mM range), but it is difficult to define this as specific neurotoxicity in the absence of specific neurotoxic endpoints. However, the neuroblastomas were significantly more susceptible to the cytotoxic effects of the α hexanedione isomers at exposures of 4 and 24 hours, compared to non-neuronal lines. Finally, a mechanism of toxicity is suggested for the α HD isomers whereby inhibition of the oxoglutarate carrier (OGC) releases apoptosis inducing factor (AIF), causing apoptosis-like cell death.

Differentiating human NT2/D1 neurospheres as a versatile in vitro 3D model system for developmental neurotoxicity testing

Developmental neurotoxicity is a major issue in human health and may have lasting neurological implications. In this preliminary study we exposed differentiating Ntera2/clone D1 (NT2/D1) cell neurospheres to known human teratogens classed as non-embryotoxic (acrylamide), weakly embryotoxic (lithium, valproic acid) and strongly embryotoxic (hydroxyurea) as listed by European Centre for the Validation of Alternative Methods (ECVAM) and examined endpoints of cell viability and neuronal protein marker expression specific to the central nervous system, to identify developmental neurotoxins. Following induction of neuronal differentiation, valproic acid had the most significant effect on neurogenesis, in terms of reduced viability and decreased neuronal markers. Lithium had least effect on viability and did not significantly alter the expression of neuronal markers. Hydroxyurea significantly reduced cell viability but did not affect neuronal protein marker expression. Acrylamide reduced neurosphere viability but did not affect neuronal protein marker expression. Overall, this NT2/D1 -based neurosphere model of neurogenesis, may provide the basis for a model of developmental neurotoxicity in vitro.

Development of a neurotoxicity test-system, using human post-mitotic, astrocytic and neuronal cell lines in co-culture

Astrocytes are essential for neuronal function and survival, so both cell types were included in a human neurotoxicity test-system to assess the protective effects of astrocytes on neurons, compared with a culture of neurons alone. The human NT2.D1 cell line was differentiated to form either a co-culture of post-mitotic NT2.N neuronal (TUJ1, NF68 and NSE positive) and NT2.A astrocytic (GFAP positive) cells (∼2:1 NT2.A:NT2.N), or an NT2.N mono-culture. Cultures were exposed to human toxins, for 4 h at sub-cytotoxic concentrations, in order to compare levels of compromised cell function and thus evidence of an astrocytic protective effect. Functional endpoints examined included assays for cellular energy (ATP) and glutathione (GSH) levels, generation of hydrogen peroxide (H2O2) and caspase-3 activation. Generally, the NT2.N/A co-culture was more resistant to toxicity, maintaining superior ATP and GSH levels and sustaining smaller significant increases in H2O2 levels compared with neurons alone. However, the pure neuronal culture showed a significantly lower level of caspase activation. These data suggest that besides their support for neurons through maintenance of ATP and GSH and control of H2O2 levels, following exposure to some substances, astrocytes may promote an apoptotic mode of cell death. Thus, it appears the use of astrocytes in an in vitro predictive neurotoxicity test-system may be more relevant to human CNS structure and function than neuronal cells alone. © 2007 Elsevier Ltd. All rights reserved.

Brain-derived neurotrophic factor as an indicator of chemical neurotoxicity:an animal-free CNS cell culture model

Recent changes to the legislation on chemicals and cosmetics testing call for a change in the paradigm regarding the current 'whole animal' approach for identifying chemical hazards, including the assessment of potential neurotoxins. Accordingly, since 2004, we have worked on the development of the integrated co-culture of post-mitotic, human-derived neurons and astrocytes (NT2.N/A), for use as an in vitro functional central nervous system (CNS) model. We have used it successfully to investigate indicators of neurotoxicity. For this purpose, we used NT2.N/A cells to examine the effects of acute exposure to a range of test chemicals on the cellular release of brain-derived neurotrophic factor (BDNF). It was demonstrated that the release of this protective neurotrophin into the culture medium (above that of control levels) occurred consistently in response to sub-cytotoxic levels of known neurotoxic, but not non-neurotoxic, chemicals. These increases in BDNF release were quantifiable, statistically significant, and occurred at concentrations below those at which cell death was measureable, which potentially indicates specific neurotoxicity, as opposed to general cytotoxicity. The fact that the BDNF immunoassay is non-invasive, and that NT2.N/A cells retain their functionality for a period of months, may make this system useful for repeated-dose toxicity testing, which is of particular relevance to cosmetics testing without the use of laboratory animals. In addition, the production of NT2.N/A cells without the use of animal products, such as fetal bovine serum, is being explored, to produce a fully-humanised cellular model.

Is glutathione an important neuroprotective effector molecule against amyloid beta toxicity?

Cellular thiols are critical moieties in signal transduction, regulation of gene expression, and ultimately are determinants of specific protein activity. Whilst protein bound thiols are the critical effector molecules, low molecular weight thiols, such as glutathione, play a central role in cytoprotection through (1) direct consumption of oxidants, (2) regeneration of protein thiols and (3) export of glutathione containing mixed disulphides. The brain is particularly vulnerable to oxidative stress, as it consumes 20% of oxygen load, contains high concentrations of polyunsaturated fatty acids and iron in certain regions, and expresses low concentrations of enzymic antioxidants. There is substantial evidence for a role for oxidative stress in neurodegenerative disease, where excitotoxic, redox cycling and mitochondrial dysfunction have been postulated to contribute to the enhanced oxidative load. Others have suggested that loss of important trophic factors may underlie neurodegeneration. However, the two are not mutually exclusive; using cell based model systems, low molecular weight antioxidants have been shown to play an important neuroprotective role in vitro, where neurotrophic factors have been suggested to modulate glutathione levels. Glutathione levels are regulated by substrate availability, synthetic enzyme and metabolic enzyme activity, and by the presence of other antioxidants, which according to the redox potential, consume or regenerate GSH from its oxidised partner. Therefore we have investigated the hypothesis that amyloid beta neurotoxicity is mediated by reactive oxygen species, where trophic factor cytoprotection against oxidative stress is achieved through regulation of glutathione levels. Using PC12 cells as a model system, amyloid beta 25-35 caused a shift in DCF fluorescence after four hours in culture. This fluorescence shift was attenuated by both desferioxamine and NGF. After four hours, cellular glutathione levels were depleted by as much as 75%, however, 24 hours following oxidant exposure, glutathione concentration was restored to twice the concentration seen in controls. NGF prevented both the loss of viability seen after 24 hours amyloid beta treatment and also protected glutathione levels. NGF decreased the total cellular glutathione concentration but did not affect expression of GCS. In conclusion, loss of glutathione precedes cell death in PC12 cells. However, at sublethal doses the surviving fraction respond to oxidative stress by increasing glutathione levels, where this is achieved, at least in part, at the gene level through upregulation of GCS. Whilst NGF does protect against oxidative toxicity, this is not achieved through upregulation of GCS or glutathione.

Evaluation of the importance of astrocytes when screening for acute toxicity in neuronal cell systems

Reliable, high throughput, in vitro preliminary screening batteries have the potential to greatly accelerate the rate at which regulatory neurotoxicity data is generated. This study evaluated the importance of astrocytes when predicting acute toxic potential using a neuronal screening battery of pure neuronal (NT2.N) and astrocytic (NT2.A) and integrated neuronal/astrocytic (NT2.N/A) cell systems derived from the human NT2.D1 cell line, using biochemical endpoints (mitochondrial membrane potential (MMP) depolarisation and ATP and GSH depletion). Following exposure for 72 h, the known acute human neurotoxicants trimethyltin-chloride, chloroquine and 6-hydroxydopamine were frequently capable of disrupting biochemical processes in all of the cell systems at non-cytotoxic concentrations. Astrocytes provide key metabolic and protective support to neurons during toxic challenge in vivo and generally the astrocyte containing cell systems showed increased tolerance to toxicant insult compared with the NT2.N mono-culture in vitro. Whilst there was no consistent relationship between MMP, ATP and GSH log IC(50) values for the NT2.N/A and NT2.A cell systems, these data did provide preliminary evidence of modulation of the acute neuronal toxic response by astrocytes. In conclusion, the suitability of NT2 neurons and astrocytes as cell systems for acute toxicity screening deserves further investigation.

The case for aluminium as the neurotoxin in sporadic Alzheimer's disease

The flash-pattern evoked potential difference (F - P) in man increases with age (93 subjects), correlates with decreasing cognitive ability and when it exceeds a unique critical level the subject is clinically diagnosed as having Alzheimer's disease. Aluminium accumulates in the human brain with age, increases the F - P value close to the critical value in a dose dependent manner, and at such a rate that normal environmental exposure to aluminium accounts for all or nearly all the F - P increases in man. Aluminium neurotoxicity is therefore a major cause of sporadic Alzheimer's disease.

Single-cell ELISA and flow cytometry as methods for highlighting potential neuronal and astrocytic toxicant specificity

The timeline imposed by recent worldwide chemical legislation is not amenable to conventional in vivo toxicity testing, requiring the development of rapid, economical in vitro screening strategies which have acceptable predictive capacities. When acquiring regulatory neurotoxicity data, distinction on whether a toxic agent affects neurons and/or astrocytes is essential. This study evaluated neurofilament (NF) and glial fibrillary acidic protein (GFAP) directed single-cell (S-C) ELISA and flow cytometry as methods for distinguishing cell-specific cytoskeletal responses, using the established human NT2 neuronal/astrocytic (NT2.N/A) co-culture model and a range of neurotoxic (acrylamide, atropine, caffeine, chloroquine, nicotine) and non-neurotoxic (chloramphenicol, rifampicin, verapamil) test chemicals. NF and GFAP directed flow cytometry was able to identify several of the test chemicals as being specifically neurotoxic (chloroquine, nicotine) or astrocytoxic (atropine, chloramphenicol) via quantification of cell death in the NT2.N/A model at cytotoxic concentrations using the resazurin cytotoxicity assay. Those neurotoxicants with low associated cytotoxicity are the most significant in terms of potential hazard to the human nervous system. The NF and GFAP directed S-C ELISA data predominantly demonstrated the known neurotoxicants only to affect the neuronal and/or astrocytic cytoskeleton in the NT2.N/A cell model at concentrations below those affecting cell viability. This report concluded that NF and GFAP directed S-C ELISA and flow cytometric methods may prove to be valuable additions to an in vitro screening strategy for differentiating cytotoxicity from specific neuronal and/or astrocytic toxicity. Further work using the NT2.N/A model and a broader array of toxicants is appropriate in order to confirm the applicability of these methods.

Investigation of aspects of reactive gliosis in human astrocytoma cell lines:application for toxicity assessment

The process of astrogliosis, or reactive gliosis, is a typical response of astrocytes to a wide range of physical and chemical injuries. The up-regulation of the astrocyte specific glial fibrillary acidic protein (GFAP) is a hallmark of reactive gliosis and is widely used as a marker to identify the response. In order to develop a reliable, sensitive and high throughput astrocyte toxicity assay that is more relevant to the human response than existing animal cell based models, the U251-MG, U373-MG and CCF-STTG 1 human astrocytoma cell lines were investigated for their ability to exhibit reactive-like changes following exposure to ethanol, chloroquine diphosphate, trimethyltin chloride and acrylamide. Cytotoxicity analysis showed that the astrocytic cells were generally more resistant to the cytotoxic effects of the agents than the SH-SY5Y neuroblastoma cells. Retinoic acid induced differentiation of the SH-SY5Y line was also seen to confer some degree of resistance to toxicant exposure, particularly in the case of ethanol. Using a cell based ELISA for GFAP together with concurrent assays for metabolic activity and cell number, each of the three cell lines responded to toxicant exposure by an increase in GFAP immunoreactivity (GFAP-IR), or by increased metabolic activity. Ethanol, chloroquine diphosphate, trimethyltin chloride and bacterial lipopolysaccharide all induced either GFAP or MTT increases depending upon the cell line, dose and exposure time. Preliminary investigations of additional aspects of astrocytic injury indicated that IL-6, but not TNF-α. or nitric oxide, is released following exposure to each of the compounds, with the exception of acrylamide. It is clear that these human astrocytoma cell lines are capable of responding to toxicant exposure in a manner typical of reactive gliosis and are therefore a valuable cellular model in the assessment of in vitro neurotoxicity.