Axel Oberemm

Identification of an enzymatically formed glutathione conjugate of the cyanobacterial hepatotoxin microcystin-LR: the first step of detoxication

Cyanobacterial toxins have adverse effects on mammals, birds and fish and are being increasingly recognised as a potent stress factor and health hazard factor in aquatic ecosystems. Microcystins, cyclic heptapeptides and a main group of the cyanotoxins are mainly retained within the producer cells during cyanobacterial bloom development. However, these toxins are released into the surrounding medium by senescence and lysis of the blooms. Any toxin present could then come into contact with a wide range of aquatic organisms including phytoplankton grazers, invertebrates, fish and aquatic plants. Recent studies showed the conversion of microcystin in animal liver to a more polar compound in correlation with a depletion of the glutathione pool of the cell. The present study shows the existence of a microcystin-LR glutathione conjugate formed enzymatically via soluble glutathione S-transferase in various aquatic organisms ranging from plants (Ceratophyllum demersum), invertebrates (Dreissena polymorpha, Daphnia magna) up to fish eggs and fish (Danio rerio). The main derived conjugate was characterized by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry yielding a mass of m/z 1302, which is equivalent to the mass assumed for a glutathione microcystin-LR conjugate. This conjugate appears to be the first step in the detoxication of a cyanobacterial toxin in aquatic organisms.

Uptake and effects of microcystin-LR on detoxication enzymes of early life stages of the zebra fish (Danio rerio)

The effects of cyanotoxins on fish have been studied mainly in adults, rather than in early life stages which could be more sensitive or, in view of their immobility, more readily affected. The uptake of microcystin‐LR by different early life stages of the zebra fish (Danio rerio) was investigated using 14C‐labelled microcystin‐LR. The effects on the activity of the detoxication enzymes, microsomal and soluble glutathione S‐transferases (GST), and glutathione peroxidase (GP‐X) were examined. There was a detectable uptake of microcystin from the first day of embryonic development up to 5 day old larvae. On average, an absorption of 0.5 ng microcystin for eggs and eleuthero‐embryos was calculated over the entire exposure time. Because of the differences in volume of the eggs and eleuthero‐embryos, there was an increase in the microcystin‐LR concentration between these stages. In the eggs, approximately 25% of the medium concentration was found, and in eleuthero‐embryos an equilibrium between fish and medium was reached.

Effects of cyanobacterial toxins and aqueous crude extracts of cyanobacteria on the development of fish and amphibians

The effects of cyanobacterial hepatotoxins microcystin‐LR, ‐RR, and ‐YR, of the neurotoxins saxitoxin and anatoxin‐a, and of crude aqueous extracts of cyanobacteria were determined on the embryos of fish and amphibians from cleavage up to advanced stages of embryonic development. No acute toxic effects were observed after exposure to microcystins at 0.5, 5, and 50 μg/L, but the timing of hatching was altered in rainbow trout: earlier hatching occurred with 0.5, 5, and 50 μg/L microcystin‐RR, 5 and 50 μg/L microcystin‐YR, and 50 μg/L microcystin‐LR; a minor delay in hatching was observed at 0.5 and 5 μg/L microcystin‐LR. In axolotl, a delay in feeding was observed (microcystin‐LR at 5 and 50 μg/L; microcystin‐YR at 50 μg/L). After termination of exposure and rearing in toxin‐free tanks, survival rate and growth were adversely affected in zebrafish larvae preexposed to microcystin‐LR at 5 and 50 μg/L. At the highest applied concentration of microcystin‐LR (10 mg/L), morphological effects were detected. Saxitoxin at 10 μg/L and above delayed hatching in zebrafish and led to malformations and mortalities at 500 μg/L. Hatching was also delayed in axolotl at 500 μg/L saxitoxin. Anatoxin‐a (400 μg/L) altered the heart rate in zebrafish, but no chronic effects were observed. Far more pronounced effects, compared to those obtained with pure toxins, were observed following exposure to various aqueous crude extracts of cyanobacteria from field samples and batch cultures: similar malformations combined with high mortalities and adverse effects on outer egg structures were observed concomitantly in all species. HPLC data showed that these effects cannot be attributed to microcystin content alone. ©1999 John Wiley & Sons, Inc. Environ Toxicol 14: 77–88, 1999

Toxicogenomic analysis of N-nitrosomorpholine induced changes in rat liver: Comparison of genomic and proteomic responses and anchoring to histopathological parameters

A common animal model of chemical hepatocarcinogenesis was used to examine the utility of transcriptomic and proteomic data to identify early biomarkers related to chemically induced carcinogenesis. N-nitrosomorpholine, a frequently used genotoxic model carcinogen, was applied via drinking water at 120 mg/L to male Wistar rats for 7 weeks followed by an exposure-free period of 43 weeks. Seven specimens of each treatment group (untreated control and 120 mg/L N-nitrosomorpholine in drinking water) were sacrificed at nine time points during and after N-nitrosomorpholine treatment. Individual samples from the liver were prepared for histological and toxicogenomic analyses. For histological detection of preneoplastic and neoplastic tissue areas, sections were stained using antibodies against the placental form of glutathione-S-transferase (GST-P). Gene and protein expression profiles of liver tissue homogenates were analyzed using RG-U34A Affymetrix rat gene chips and two-dimensional gel electrophoresis-based proteomics, respectively. In order to compare results obtained by histopathology, transcriptomics and proteomics, GST-P-stained liver sections were evaluated morphometrically, which revealed a parallel time course of the area fraction of preneoplastic lesions and gene plus protein expression patterns. On the transcriptional level, an increase of hepatic GST-P expression was detectable as early as 3 weeks after study onset. Comparing deregulated genes and proteins, eight species were identified which showed a corresponding expression profile on both expression levels. Functional analysis suggests that these genes and corresponding proteins may be useful as biomarkers of early hepatocarcinogenesis.

Proteomic responses of human intestinal Caco-2 cells exposed to silver nanoparticles and ionic silver

Even although quite a number of studies have been performed so far to demonstrate nanoparticle‐specific effects of substances in living systems, clear evidence of these effects is still under debate. The present study was designed as a comparative proteomic analysis of human intestinal cells exposed to a commercial silver nanoparticle reference material and ions from AgNO3. A two‐dimensional gel electrophoresis/MALDI mass spectrometry (MS)‐based proteomic analysis was conducted after 24‐h incubation of differentiated Caco‐2 cells with non‐cytotoxic and low cytotoxic silver concentrations (2.5 and 25 µg ml−1 nanosilver, 0.5 and 5 µg ml−1 AgNO3). Out of an overall number of 316 protein spots differentially expressed at a fold change of ≥ 1.4 or ≤ −1.4 in all treatments, 169 proteins could be identified. In total, 231 spots were specifically deregulated in particle‐treated groups compared with 41 spots, which were limited to AgNO3‐treatments. Forty‐four spots (14 %) were commonly deregulated by both types of treatment. A considerable fraction of the proteins differentially expressed after treatment with nanoparticles is related to protein folding, synthesis or modification of proteins as well as cellular assembly and organization. Overlays of networks obtained for particulate and ionic treatments showed matches, indicating common mechanisms of combined particle and ionic silver exposure and exclusive ionic silver treatment. However, proteomic responses of Caco‐2 cells treated with higher concentrations of silver species also showed some differences, for example regarding proteins related to fatty acid and energy metabolism, suggesting an induction of also some different molecular mechanisms for particle exposure and ionic treatment. Copyright

Sensitivity of early life stages of vendace, Coregonus albula, to acid pH in postmining lakes: An experimental approach

Since 1990, many highly acidic lakes with very hard water have formed in Lusatia, eastern Germany, following the decommissioning of most open‐cast lignite mines. Even after neutralization of the water, the lakes may reacidify due to the high acidification potential of soil and groundwater. Hence, investigation of critical pH‐levels for the respective fish species is required when planning to use the lakes for fisheries. Sensitivity of early life stages of vendace, Coregonus albula, to low pH was determined in reconstituted water with the hydrochemical characteristics of the postmining lakes. Eggs were transferred to test solutions with pH‐values ranging from 3.50 to 7.40 either 10 min after artificial insemination (series A) or 7 h later (series B) and incubated using a static‐renewal procedure. Water hardening of the eggs at exposure pH (series A) led to a severe reduction in egg diameter at pH≤5.00 and to an earlier and stronger increase in egg mortality, but by the time of hatching, differences between both series were small. Hatching percentages at pH≤5.00 were very low and the eleutheroembryos died shortly after hatching. In both series, hatching percentages at pH 5.50 did not differ from those at pH 7.40. Feeding activity of vendace exposed to pH 5.50 was, however, reduced and the fish did not survive. Hence, early life stages of vendace can be regarded as very sensitive to low pH and associated elevated concentrations of Al, and maintenance of neutral pH is essential when postmining lakes are stocked with vendace. © 2000 John Wiley & Sons, Inc. Environ Toxicol 15: 214–224, 2000

Letter to the Editor: A regulatory view on the discussion on the role of alternative methods in the risk assessment of chemicals in the context of REACH

In a paper published recently in Archives of Toxicology, Lilienblum et al. present their view on available and ready to use alternative methods for the risk assessment of chemicals (Lilienblum et al. 2008). As they refer speciWcally to the new European legislation (REACH) which came into force mid 2007 we would like to comment some of the aspects of the paper from the perspective of regulators. First of all there is general agreement with the evaluation of the current status of alternative methods of the authors (Lilienblum et al. 2008). Considering complex endpoints, we consent that emphasis should be given to reWne tests with the aim to reduce the number of animals rather than to try to replace the tests by developing in vitro alternatives to the current testing methods. However, we would more clearly express the potential for the vitro methods for some of the endpoints which are discussed in the publication. For example, the in vitro dermal resorption test and the in vitro test on phototoxicity are both validated test methods and accepted at the regulatory level. Hence, in our view no in vivo tests are necessary to cover these endpoints. In addition, we see a high value in the available prediction tool for skin irritation and corrosion (ECB 2008) which is extremely helpful in a tiered approach. For classiWcation and labelling negative results can be used as they have clearly been identiWed in the validation study (Hulzebos et al. 2005). Hence, even under the GHS rules where two levels for positive results are required (which the prediction tool is not able to distinguish) the tool is nevertheless helpful considering that negative results for skin irritation and corrosion account for more than 70% of the cases in the EU New Chemicals Database. It can be assumed that using this alternative method would reduce the need for in vivo testing under REACH in 70% of 15,500 chemicals with a production volume below 10 tons per year. We also would put more emphasis in the potential the currently available in vitro test strategy can oVer for genotoxicity and mutagenicity testing. We agree with the assessment and the critique presented in the Lilienblum paper (Lilienblum et al. 2008). However, we would like to draw the attention to the fact that the currently used regulatory tiered testing approach reduces the necessity to perform in vivo testing to a small number of cases. Based on the numbers in the New Chemicals Data Base, the current approach to start with in vitro testing and to proceed with in vivo testing only in the positive in vitro cases was required in only around 25% of the cases which reduces the number of required in vivo tests under REACH to a quite remarkable low number of cases. Furthermore, we would like to give some considerations to the topic of toxicokinetic and metabolism which is extensively discussed in the Lilienblum paper. We agree with the assessment that toxicokinetic and metabolism is of importance and is a crucial issue and that metabolism often plays a key role in intraand inter-species diVerences (to document the regulatory need we refer to work of Bernauer et al. 2000, 2002, 2003, 2006; Abraham et al. 2005; Mielke et al. 2005; Bernauer and Gundert-Remy 2008). We agree with other scientists that the interpretation of testing results need additional data for extrapolation of the results to humans (Greim 2007) and toxicokinetic information is regarded as very helpful for the development of integrated testing strategies (ITS) which are propagated in the REACH guidance documents according to REACH legislation. In addition, toxicokinetic data can be used to support waiving arguments (Annex VIII, 8.6.1 90 days study; Annex IX, 8.7 studies on reprotoxiccity; Annex X, 8.4 germ cell mutageU. Gundert-Remy (&) · U. Bernauer · S. Madle · A. Oberemm · A. Schulte · H.-B. Richter-Reichhelm Berlin, Germany e-mail: ursula.gundert-remy@bfr.bund.de

Comparative proteomic analysis of hepatic effects induced by nanosilver, silver ions and nanoparticle coating in rats

The presence of nano-scaled particles in food and food-related products has drawn attention to the oral uptake of nanoparticles and their interactions with biological systems. In the present study, we used a toxicoproteomics approach to allow for the untargeted experimental identification and comparative analysis of cellular responses in rat liver after repeated-dose treatment with silver nanoparticles, ions, and the coating matrix used for particle stabilization. The proteomic analysis revealed treatment-related effects caused by exposure to silver in particulate and ionic form. Both silver species induced similar patterns of signaling and metabolic alterations. Silver-induced cellular alterations comprised, amongst others, proteins involved in metal homeostasis, oxidative stress response, and energy metabolism. However, we discovered that secondary nano-scaled structures were formed from ionic silver. Furthermore, also the coating matrix alone gave rise to the formation of nano-scaled particles. The present data confirm, complement, and extend previous knowledge on silver toxicity in rodent liver by providing a comprehensive proteomic data set. The observation of secondary particle formation from non-particle controls underlines the difficulties in separating particle-, ion-, and matrix coating-related effects in biological systems. Awareness of this issue will support proper evaluation of nanotoxicology-related data in the future.

Proteomic effects of repeated-dose oral exposure to 2-monochloropropanediol and its dipalmitate in rat testes.

2- and 3-monochloropropanediol (2-MCPD) and their fatty acid esters are food contaminants which are concomitantly formed upon thermal treatment of foodstuff containing fats and salt. Exposure to 2- or 3-MCPD thus results, for example, from refined vegetable oils, in instant meals or infant formula, as well as in cereals or pastries. The molecular mechanisms of 2-MCPD toxicity are poorly understood. Here, we performed a comprehensive proteomic analysis of 2-MCDP-induced alterations in the testes from rats following oral administration of 10 mg/kg body weight per day 2-MCPD, or an equimolar dose of 2-MCPD dipalmitate as a representative 2-MCPD fatty acid ester. In the absence of overt histopathologically detectable toxicity, moderate alterations in cellular proteomic signatures were recorded. The observations are in line with the assumption that the molecular mechanisms of 2-MCPD and 3-MCPD toxicity differ. Observed proteomic alterations point towards effects of 2-MCPD on mitogen-dependent signaling and mitochondrial energy utilization. Presented data for the first time provide insight into proteomic effects of 2-MCPD in testicular tissue.

Comparative proteomic analysis of silver nanoparticle effects in human liver and intestinal cells: comparison of nanosilver toxicity in liver and intestinal cells

Consumers are orally exposed to nanoparticulate or soluble species of the non‐essential element silver due to its use in food contact materials or as a food additive. Potential toxicity of silver nanoparticles has gained special scientific attention. A fraction of ingested ionic or particulate silver is taken up in the intestine and transported to the liver, where it may induce oxidative stress and elicit subsequent adverse responses. Here, we present a comprehensive analysis of global proteomic changes induced in human Hep G2 hepatocarcinoma cells by different concentrations of AgPURE silver nanoparticles or by corresponding concentrations of ionic silver. Bioinformatic analysis of proteomic data confirms and substantiates previous findings on silver‐induced alterations related to redox stress, mitochondrial dysfunction, intermediary metabolism, inflammatory responses, posttranslational protein modification and other cellular parameters. Similarities between the effects exerted by the two silver species are in line with the assumption that silver ions released from nanoparticles substantially contribute to their toxicity. Moreover, a comparative bioinformatic evaluation of proteomic effects in hepatic and intestinal cells exerted either by silver nanoparticles or bionic silver is presented. Our results show that, despite remarkable differences at the level of affected proteins in the different cell lines, highly similar biological consequences, corresponding to previous in vivo findings, can be deduced by applying appropriate bioinformatic data mining.