Factors influencing nominal effective concentrations of chemical compounds in vitro: medium protein concentration

Time- and dose-dependent effects of roundup on human embryonic and placental cells

Roundup is the major herbicide used worldwide, in particular on genetically modified plants that have been designed to tolerate it. We have tested the toxicity and endocrine disruption potential of Roundup (Bioforce on human embryonic 293 and placental-derived JEG3 cells, but also on normal human placenta and equine testis. The cell lines have proven to be suitable to estimate hormonal activity and toxicity of pollutants. The median lethal dose (LD(50)) of Roundup with embryonic cells is 0.3% within 1 h in serum-free medium, and it decreases to reach 0.06% (containing among other compounds 1.27 mM glyphosate) after 72 h in the presence of serum. In these conditions, the embryonic cells appear to be 2-4 times more sensitive than the placental ones. In all instances, Roundup (generally used in agriculture at 1-2%, i.e., with 21-42 mM glyphosate) is more efficient than its active ingredient, glyphosate, suggesting a synergistic effect provoked by the adjuvants present in Roundup. We demonstrated that serum-free cultures, even on a short-term basis (1 h), reveal the xenobiotic impacts that are visible 1-2 days later in serum. We also document at lower non-overtly toxic doses, from 0.01% (with 210 microM glyphosate) in 24 h, that Roundup is an aromatase disruptor. The direct inhibition is temperature-dependent and is confirmed in different tissues and species (cell lines from placenta or embryonic kidney, equine testicular, or human fresh placental extracts). Furthermore, glyphosate acts directly as a partial inactivator on microsomal aromatase, independently of its acidity, and in a dose-dependent manner. The cytotoxic, and potentially endocrine-disrupting effects of Roundup are thus amplified with time. Taken together, these data suggest that Roundup exposure may affect human reproduction and fetal development in case of contamination. Chemical mixtures in formulations appear to be underestimated regarding their toxic or hormonal impact.

The improvement of in vitro cytotoxicity testing for the assessment of acute toxicity in fish

The use of fish cell line cytotoxicity tests as alternatives to acute lethality tests with fish is hampered by the clearly lower sensitivity of the fish cell line tests. Recently, it has been shown that this is not a unique feature of fish cells. In fact, the sensitivity of mammalian and human cell lines toward the cytotoxic actions of chemicals, in general, is comparable to that of fish cell lines. Reviewing some of our recent investigations, the objective of this paper is to show that the sensitivity of in vitro cytotoxicity testing and the correspondence between in vitro cytotoxic and acute fish toxic concentrations (LC50) can be increased, if: a) inhibition of cell growth instead of cell death is used as the endpoint; and b) the bioavailable free cytotoxic concentration (ECu50) of chemicals in vitro, instead of the nominal cytotoxic concentration (EC50), is used as the measure of cytotoxic potency. Based on these results, a pragmatic in vitro testing strategy for estimating the minimal aquatic toxic potency of chemicals is proposed.

Validation of a prediction model for estimating serum concentrations of chemicals which are equivalent to toxic concentrations in vitro

The objective of the present study was to evaluate the validity of a recently developed extrapolation model for the prediction of concentrations of chemicals in serum which are equivalent to in vitro effective nominal concentrations. Necessary input data are in vitro toxic concentrations and distribution relevant system and substance specific parameters, e.g. lipid volume fractions and albumin concentrations, octanol/water partition coefficients and specific binding to albumin. It was investigated whether the influence of human and bovine serum, respectively, on nominal cytotoxic potencies (EC(50)-values) of selected chemicals in vitro can be properly predicted using this algorithm. Cytotoxicity was determined as growth inhibition of proliferating Balb/c 3T3 cells after exposure for 72 h. Concentration-effect relationships were measured in the presence of 2% foetal bovine serum (FBS) and, additionally, 18% FBS or human serum (HS), or 1% (w/v) bovine (BSA) or human (HSA) albumin, respectively. Addition of HSA and BSA increased the EC(50)-values of the different chemicals by factors of 2.1 - 22 and 1.7 - 29, respectively. From these measurements values for the specific binding of the test compounds to BSA and HSA were derived. Addition of 18% HS increased the EC(50)-values by factors between 4.2 and 52, while addition of 18% FBS resulted only in 1.5 - 10.4-fold increases. A comparison of experimentally determined and calculated EC(50)-values revealed that the differing influence of human and bovine serum was quite well predicted by the extrapolation model. Deviations did not exceed the factor 3 and were in most cases lower than 2. It is concluded that the extrapolation model is quite well suited to predict equivalent concentrations in serum from in vitro effective concentrations.

Cleavage of tensin during cytoskeleton disruption in YTX-induced apoptosis

Yessotoxin (YTX) is a marine algal toxin previously shown to induce apoptosis in L6 and BC3H1 myoblast cell lines. Disassembly of the F-actin cytoskeleton and cleavage of tensin, a cytoskeletal protein localised at the focal adhesion contacts, appear during this apoptotic process. Tensin binds to actin filaments at the focal adhesion contacts and it links the actin cytoskeleton to the extracellular matrix (ECM). This binding occurs via integrin receptors and it makes tensin a potential link between the actin cytoskeleton and signal transduction. This study evaluates disruption in the F-actin cytoskeleton and change of tensin in myoblast cell lines exposed to 100 nM YTX up to 72 h. YTX treatment cleaves tensin and makes it translocate to the cell centre. Tensin has normally a role in the maintenance of cell shape and YTX-treatment may therefore alter the shape of the cells. YTX exposure also induces formation of lamellas associated with pseudopodia. Alternative linkages and cytoskeletal proteins anchoring the actin filaments to focal contacts remain to be identified.

Proposal to improve vertebrate cell cultures to establish them as substitutes for the regulatory testing of chemicals and effluents using fish

Cultures of vertebrate cells are widely applied in mechanistic studies in human toxicology as well as in toxicity identification in ecotoxicology. As in vitro models, they display many advantages over whole animal experimentation, pertaining to such characteristics as availability, reproducibility and costs. As well, they satisfy the societal desire to reduce the number of animals in toxicology. For these reasons vertebrate cell models also appear to be a desirable replacement for animals in regulatory tests. Several vertebrate cell models are now accepted for regulatory purposes in human health sciences, with the test for photocytotoxicity using the 3T3 mouse cell line being one example. However, an in vitro alternative to whole animal tests has not yet been established for regulatory risk assessment in ecotoxicology. This review sets out to outline why such a replacement has not yet been possible and explores avenues to improve vertebrate cell cultures so that a replacement of whole animal tests could more likely be achieved. Inasmuch as fish is the most widely used non-mammalian vertebrate in risk assessment and regulation, focus will be on the replacement, by in vitro vertebrate models, of fish.

Comparison of the cytotoxicity of the MEIC reference chemicals measured in protein free and in complete culture medium

In order to investigate if a protein free cytotoxicity assay could improve the prediction of human acute toxicity, the cytotoxicity of 40 MEIC reference chemicals was measured by the neutral red uptake inhibition after 24h in protein free culture medium on rat hepatoma-derived Fa32 cells. The results were compared with the corresponding values obtained in complete culture medium, including 10% fetal calf serum. Potassium cyanide, arsenic trioxide, mercuric chloride, hexachlorophene and pentachlorophenol were much more cytotoxic in PF medium, as was the case to a lower extent for 16 other chemicals. The cytotoxicity of 8 chemicals was only changed to a limited extent when tested in PF medium, suggesting that serum proteins do not strongly interact with their cytotoxicity. Eleven other chemicals were less cytotoxic in PF medium, maybe because of too poor physiological conditions. Although a large number of differences in cytotoxicity were observed in function of the medium used for the assay, a good correlation was observed between both series of data (r(2)=0.946). The correlation between the cytotoxicity in PF medium and the human acute toxicity is lower (r(2)=0.647) than that in complete medium (r(2)=0.746). The results show that further research is necessary in order to improve the in vitro/in vivo correlations by introducing protein-dependent considerations.

An alternative approach for the safety evaluation of new and existing chemicals, an exercise in integrated testing

Various in vitro and in silico methods without animals were applied to 10 substances listed on ELINCS with a complete VIIA base-set available at NOTOX. The hazard assessment for these substances was performed on basis of available non-animal data, QSAR, PBBK-modelling and additional, new in vitro testing was applied. Based on these data predictions on fish toxicity, acute toxicity, skin- and eye-irritation, sensitisation, and toxicity after repeated dosing were made. The predictions were compared with the outcome of the in vivo tests. Nine out of ten predictions on fish LC(50) proved to be correct. For skin- and eye-irritation 70% was predicted correctly. Sensitisation was predicted correctly for 7 out of 10 substances, but three false negatives were found. Acute oral toxicity (LD(50)) and repeated dose toxicity were less successful (5 out of 10 and 2 out of 10 correct predictions, respectively); application of the PBBK model proved successful. Acute dermal toxicity was predicted correctly in 9 out of 10 cases. In general an over-estimation of systemic toxicity was found, which can be explained by an over-prediction of cytotoxicity and worst case assumptions on absorption and binding to (plasma) proteins. This integrated approach leads to a 38% reduction of laboratory animals.

Differential effects of glyphosate and roundup on human placental cells and aromatase

Roundup is a glyphosate-based herbicide used worldwide, including on most genetically modified plants that have been designed to tolerate it. Its residues may thus enter the food chain, and glyphosate is found as a contaminant in rivers. Some agricultural workers using glyphosate have pregnancy problems, but its mechanism of action in mammals is questioned. Here we show that glyphosate is toxic to human placental JEG3 cells within 18 hr with concentrations lower than those found with agricultural use, and this effect increases with concentration and time or in the presence of Roundup adjuvants. Surprisingly, Roundup is always more toxic than its active ingredient. We tested the effects of glyphosate and Roundup at lower nontoxic concentrations on aromatase, the enzyme responsible for estrogen synthesis. The glyphosate-based herbicide disrupts aromatase activity and mRNA levels and interacts with the active site of the purified enzyme, but the effects of glyphosate are facilitated by the Roundup formulation in microsomes or in cell culture. We conclude that endocrine and toxic effects of Roundup, not just glyphosate, can be observed in mammals. We suggest that the presence of Roundup adjuvants enhances glyphosate bioavailability and/or bioaccumulation.

Microglia is activated by astrocytes in trimethyltin intoxication

Microglia participates in most acute and chronic neuropathologies and its activation appears to involve interactions with neurons and other glial cells. Trimethyltin (TMT)-induced brain damage is a well-characterized model of neurodegeneration, in which microglial activation occurs before neuronal degeneration. The aim of this in vitro study was to investigate the role of astroglia in TMT-induced microgliosis by using nitric oxide (NO), inducible NO synthase (iNOS), and morphological changes as parameters for microglial activation. Our investigation discusses (a) whether microglial cells can be activated directly by TMT; (b) if astroglial cells are capable of triggering or modulating microglial activation; (c) how the morphology and survival of microglia and astrocytes are affected by TMT treatment; and (d) whether microglial-astroglial interactions depend on direct cell contact or on soluble factors. Our results show that microglia are more vulnerable to TMT than astrocytes are and cannot be activated directly by TMT with regard to the examined parameters. In bilayer coculture with viable astroglial cells, microglia produce NO in significant amounts at subcytotoxic concentrations of TMT (20 micromol/l). At these TMT concentrations, microglial cells in coculture convert into small round cells without cell processes, whereas flat, fibroblast-like astrocytes convert into thin process bearing stellate cells with a dense and compact cell body. We conclude that astrocytes trigger microglial activation after treatment with TMT, although the mechanisms of this interaction remain unknown.

Sunscreen penetration of human skin and related keratinocyte toxicity after topical application

Sunscreen skin penetration and safety assessment should be considered together in order to ensure that in vitro cytotoxicity studies examine relevant doses of these organic chemical UV filters to which viable epidermal cells are realistically exposed. In this study, we sought to determine whether sufficient topically applied sunscreens penetrated into human viable epidermis to put the local keratinocyte cell populations at risk of toxicity. The penetration and retention of five commonly used sunscreen agents (avobenzone, octinoxate, octocrylene, oxybenzone and padimate O) in human skin was evaluated after application in mineral oil to isolated human epidermal membranes. Sunscreen concentration-human keratinocyte culture response curves were then defined using changes in cell morphology and proliferation (DNA synthesis using radiolabelled thymidine uptake studies) as evidence of sunscreens causing toxicity. Following 24 h of human epidermal exposure to sunscreens, detectable amounts of all sunscreens were present in the stratum corneum and viable epidermis, with epidermal penetration most evident with oxybenzone. The concentrations of each sunscreen found in human viable epidermis after topical application, adjusting for skin partitioning and binding effects, were at least 5-fold lower, based on levels detected in viable epidermal cells, than those appearing to cause toxicity in cultured human keratinocytes. It is concluded that the human viable epidermal levels of sunscreens are too low to cause any significant toxicity to the underlying human keratinocytes.

Impact of bioavailability on the correlation between in vitro cytotoxic and in vivo acute fish toxic concentrations of chemicals

The lower sensitivity of in vitro cytotoxicity assays currently restricts their use as alternative to the fish acute toxicity assays for hazard assessment of chemicals in the aquatic environment. In vitro cytotoxic potencies mostly refer to nominal concentrations. The main objective of the present study was to investigate, whether a reduced availability of chemicals in vitro can account for the lower sensitivity of in vitro toxicity test systems. For this purpose, the bioavailable free fractions of the nominal cytotoxic concentrations (EC50) of chemicals determined with a cytotoxicity test system using Balb/c 3T3 cells and the corresponding free cytotoxic concentrations (ECu50) were calculated. The algorithm applied is based on a previously developed simple equilibrium distribution model for chemicals in cell cultures with serum-supplemented culture media. This model considers the distribution of chemicals between water, lipids and serum albumin. The algorithm requires the relative lipid volume of the test system, the octanol-water partition coefficient (K(ow)) and the in vitro albumin-bound fraction of the chemicals. The latter was determined from EC50-measurements in the presence of different albumin concentrations with the Balb/c 3T3 test system. Organic chemicals covering a wide range of cytotoxic potency (EC50: 0.16-527000 microM) and lipophilicity (logK(ow): -5.0-6.96) were selected, for which fish acute toxicity data (LC50-values) from at least one of the three fish species, medaka, rainbow trout and fathead minnow, respectively, were available. The availability of several chemicals was shown to be extensively reduced either by partitioning into lipids or by serum albumin binding, or due to both mechanisms. Reduction of bioavailability became more important with increasing cytotoxic potency. The sensitivity of the Balb/c 3T3 cytotoxicity assay and the correspondence between in vivo and in vitro toxic potencies were increased when the free cytotoxic concentrations instead of the nominal cytotoxic concentrations were used as measure of cytotoxic potency. The few remaining prominent differences between cytotoxic and acute toxic concentrations can be explained by a more specific mechanism of acute toxic action than basal cytotoxicity. It is concluded that the frequently observed low sensitivity of in vitro cytotoxicity test systems, compared to fish acute toxicity assays, at least in part, can be explained by differences in the availability of chemicals in vitro and in vivo. Moreover, neglecting these differences systematically causes a bias of the correlation between in vivo and in vitro toxic potencies of chemicals. Taking them into account, however, increases the predictivity of the in vitro assays.

The prediction of human acute systemic toxicity by the EDIT/MEIC in vitro test battery: the importance of protein binding and of partitioning into lipids

The aim of the two studies presented in this paper was to further improve the predictability of the original Multicentre Evaluation of In Vitro Cytotoxicity (MEIC) in vitro test battery for acute systemic toxicity. In the first study, whether a protein-free cytotoxicity assay could improve the prediction of human acute systemic toxicity was investigated. The cytotoxicity of 39 MEIC reference chemicals was measured by the neutral red uptake inhibition test after 30 minutes in phosphate-buffered saline (PBS), with hepatoma-derived Fa32 cells. The results were compared with the corresponding values obtained in complete culture medium, including 10% fetal calf serum. Mercuric chloride and hexachlorophene were much more cytotoxic in PBS, as was the case, to a lesser extent, for seven other chemicals. Potassium cyanide and eight other chemicals were less cytotoxic in PBS than in complete culture medium, probably because of poor physiological conditions. The correlation between the cytotoxicity measured in PBS and human acute toxicity was rather low, but became of the same order as for other assays, when mercuric chloride and hexachlorophene were withdrawn from the comparison. In the second study, modelling of human lethal blood concentrations by using the results of the three cell line tests of the original MEIC test battery were complemented by logP (octanol-water partition coefficient) values. The introduction of logP into the modelling did not improve the correlations, but some improvement of both R(2) and Q(2) was obtained by expanding the logP values with logP(2) values. The highest R(2) (0.84) and Q(2) (0.80) values were obtained for a model in which both experimental and calculated (ambiguous) logP values were used. When only experimental logP values were used, the corresponding values were 0.80 and 0.78. These two studies showed that including protein binding and the partition of chemicals in the MEIC in vitro test battery is important, in order to improve the predictability of the results obtained.

Cytotoxic potency of trialkyltins to C6 glioma cells in vitro: impact of exposure conditions

Because of a possible role of astrocytes in trialkyltin-induced neurotoxicity in vivo various studies have been performed using cultures of astrocytes or glioma cells in vitro. With respect to cytotoxic potencies of trialkyltins these studies gave rather divergent results. Therefore the aim of the present study was to clarify whether variations of experimental conditions could be responsible for the differences of the cytotoxic activities of trimethyltin (TMT), triethyltin (TET) and tributyltin (TBT). Experiments were performed with rat C6 glioma cells. Toxicity was determined by measuring the reduction of the cell protein content. Cultures of proliferating and growth-arrested cells did not differ in their sensitivity. Exposure duration (1-72 h) had a strong but differing influence on the cytotoxic potency of the trialkyltins. After short exposure times the potencies differed largely (TMT < TET < TBT), whereas they became more and more similar with increasing exposure duration. The potency-time relationships for TMT and TET could be described by the equation: EC50 = k x t(-n), while for TBT an incipient value (EC50, infinity) had to be included: EC50 = EC50, infinity + k x t(-n). Addition of serum albumin to the culture medium decreased the cytotoxic potency of the trialkyltins. However, the impact of protein binding on their bioavailability was relatively low. The cytotoxic potency of the alkyltins was not dependent on the concentration of C6 cells. Taken together, neither differences in exposure conditions nor in the proliferative status of the cells are sufficient to account for the discrepancies in published results for trialkyltin cytotoxicity to astrocytes. Instead they may--at least partially--be explained by differing sensitivities of the endpoints used. Furthermore, C6 glioma cells respond considerably more sensitively to trialkytins than primary astrocytes, which questions their applicability as models for astrocyte toxicity.

Evidence for delayed cytotoxicity effects following exposure of rat hepatoma-derived Fa32 cells: implications for predicting human acute toxicity

The delayed cytotoxicity of the Multicentre Evaluation of In vitro Cytotoxicity (MEIC) reference chemicals was investigated in rat hepatoma-derived Fa32 cells. The cells were treated for 24 h with the test chemicals, and were than further cultured for 5 days in normal culture medium. The cytotoxicity was measured by the neutral red uptake inhibition, and the results were quantified by determining the NI50del. This is the concentration of test compound required to decrease the neutral red uptake with 50% compared with control cells. The results were compared with the acute NI50, the corresponding value measured immediately after 24 h treatment of the cells. On a total of 44 chemicals, nine showed delayed cytotoxicity (NI50del lower than or equal to NI50), 11 a probably delayed, and 24 no delayed cytotoxicity (NI50del more than 1.5 x NI50). When the NI50del was compared with human toxicity, a correlation coefficient r2=0.761 was obtained. For the same series of 44 chemicals this correlation was clearly higher than that for human hepatoma-derived Hep G2 cells (r2=0.695). The Hep G2 assay was the best acute in vitro assay for the prediction of human toxicity within the MEIC study. Consequently, the delayed cytotoxicity assay on cultured Fa32 cells has the best prediction value so far obtained for the human toxicity.

In vitro-in vivo extrapolation: estimation of human serum concentrations of chemicals equivalent to cytotoxic concentrations in vitro

In the present study an extrapolation model for estimating serum concentrations of chemicals equivalent to in vitro effective concentrations is developed and applied to median cytotoxic concentrations (EC(50)) determined in vitro. Nominal concentrations of a chemical in serum and in vitro are regarded as equivalent, if they result in the same aqueous concentration of the unbound form. The algorithm used is based on equilibrium distribution and requires albumin binding data, the octanol-water partition coefficient (K(ow)), and the albumin concentrations and lipid volume fractions in vitro and in serum. The chemicals studied cover wide ranges of cytotoxic potency (EC(50): 2.5-530,000 microM) and lipophilicity (logK(ow): -5 to 7). Their albumin binding characteristics have been determined by means of an in vitro cytotoxicity test as described previously. The equivalent serum concentrations of 19 of the 33 compounds investigated, having high protein binding and/or lipophilicity, were substantially higher than the EC(50)-values, by factors of 2.5-58. Prominent deviations between the equivalent nominal concentrations in serum and in vitro were largely restricted to chemicals with higher cytotoxic potency (EC(50)< or =1000 microM). The results suggest that estimates of equivalent serum concentrations based on in vitro data are robust for chemicals with low lipophilicity (logK(ow)< or =2) and low potency (EC(50)>1000 microM). With more potent chemicals or those with higher lipophilicity partitioning into lipids and/or binding to serum proteins have to be taken into account when estimating in vivo serum concentrations equivalent to in vitro effective concentrations.

Serum albumin binding at cytotoxic concentrations of chemicals as determined with a cell proliferation assay

The aim of the present study was to measure the influence of albumin binding on cytotoxic concentrations of chemicals and to determine binding parameters which can be used for quantitative in vitro-in vivo extrapolations. Protein binding parameters were determined from cytotoxic potencies measured with Balb/c 3T3 cells cultured in the presence of 18 and 600 microM bovine serum albumin (BSA). A subset of 27 chemicals from the Multicenter Evaluation of In Vitro Cytotoxicity (MEIC) project was investigated. At 18 microM BSA the EC(50)-values ranged from 2.54 microM (As(III)) to 527 mM (ethylene glycol). Increasing the BSA concentration either decreased the cytotoxic potency (12 compounds) by factors up to 34 (pentachlorophenol), had no effect (14 compounds), or increased the cytotoxicity (paraquat). Calculated molar ratios of binding ranged from 0.05 (Hg(2+)) to 4.8 moles per mole albumin (acetylic salicylic acid). At 18 microM BSA fractional binding of most of these compounds was low (<25%) but increased up to > or =90% (hexachlorophene, mercuric chloride, thioridazine, pentachlorophenol) at 600 microM BSA. The results obtained in general were compatible with available protein binding data and can be used to calculate equipotent concentrations of chemicals in biological systems containing different albumin concentrations.

Impact of protein binding on the availability and cytotoxic potency of organochlorine pesticides and chlorophenols in vitro

In vitro toxicity data are generally based on nominal concentrations and thus depend on both activity and availability of a compound. The aims of the present study were to examine the influence of protein binding on the cytotoxicity of selected organochlorine pesticides and chlorophenols in Balb/c 3T3 cell cultures and to determine parameters of protein binding which can be used to estimate protein bound fractions and to model distribution in vitro. EC(50)-values derived from concentration-effect relationships determined in the presence of various concentrations of bovine serum albumin (BSA) were linearly correlated to BSA concentration. Increasing the BSA concentration from about 1.2 to 40 mg/ml increased the EC(50)-values by factors between 3.4 and 34.4. Molar ratios of substance bound to albumin ranged from 0.11 to 2.42. Calculated fractions bound to albumin in the normal growth medium were 0.075-0.17 (p,p'-DDT, p,p'-DDE, dieldrin, lindane), 0.09-0.1 (4-mono- and 2,4-dichlorophenol), 0.68 (2,4,5-trichlorphenol) and almost 1.0 (pentachlorophenol). At 40 mg/ml BSA any compound was largely bound to albumin (fractions bound > or = 0.74). Distribution modelling revealed that the availability of the highly hydrophobic organochlorines additionally was significantly reduced by partitioning into lipids. The results clearly demonstrate that nominal and relative toxic potencies of organochlorine pesticides and chlorophenols determined in vitro are substantially influenced by effects of protein binding on availability.