Comparative cellular toxicity of hydrophilic and hydrophobic microcystins on Caco-2 cells.

Microcystins (MC), cyanobacterial peptide hepatotoxins, comprise more than 100 different variants. They are rather polar molecules but some variants contain hydrophobic amino acid residues in the highly variable parts of the molecule. In MC-LF and MC-LW, the more hydrophobic phenylalanine (F) and tryptophan (W), respectively, have replaced arginine (R) in MC-LR. Depending on the structure, microcystins are expected to have different in vivo toxicity and bioavailability, but only a few studies have considered the toxic properties of the more hydrophobic variants. The present study shows that MC-LF and MC-LW have more pronounced cytotoxic effects on Caco-2 cells as compared to those of MC-LR. Treatment of Caco-2 cells with MC-LW and especially MC-LF showed clear apoptotic features including shrinkage and blebbing, and the cell–cell adhesion was lost. An obvious reduction of cell proliferation and viability, assessed as the activity of mitochondrial dehydrogenases, was observed with MC-LF, followed by MC-LW and MC-LR. Cytotoxicity was quantified by measuring lactate dehydrogenase leakage. The more hydrophobic MC-LW and MC-LF induced markedly enhanced lactate dehydrogenase leakage compared to controls and MC-LR, indicating that the plasma membrane was damaged. All of the three toxins examined inhibited protein phosphatase 1, with MC-LF and MC-LW to a weaker extent compared to MC-LR. The higher toxic potential of the more hydrophobic microcystins could not be explained by the biophysical experiments performed. Taken together, our data show that the more hydrophobic microcystin variants induce higher toxicity in Caco-2 cells.

Removal of cholera toxin from aqueous solution by probiotic bacteria.

Cholera remains a serious health problem, especially in developing countries where basic hygiene standards are not met. The symptoms of cholera are caused by cholera toxin, an enterotoxin, which is produced by the bacterium Vibrio cholerae. We have recently shown that human probiotic bacteria are capable of removing cyanobacterial toxins from aqueous solutions. In the present study we investigate the ability of the human probiotic bacteria, Lactobacillus rhamnosus strain GG (ATCC 53103) and Bifidobacterium longum 46 (DSM 14583), to remove cholera toxin from solution in vitro. Lactobacillus rhamnosus strain GG and Bifidobacterium longum 46 were able to remove 68% and 59% of cholera toxin from aqueous solutions during 18 h of incubation at 37 °C, respectively. The effect was dependent on bacterial concentration and L. rhamnosus GG was more effective at lower bacterial concentrations. No significant effect on cholera toxin concentration was observed when nonviable bacteria or bacterial supernatant was used.

Production and sedimentation of peptide toxins nodularin-R and microcystin-LR in the northern Baltic Sea.

This seven-year survey was primarily targeted to quantification of production of nodularin-R (NOD-R), a cyclic pentapeptide hepatotoxin, in Baltic Sea cyanobacteria waterblooms. Additionally, NOD-R and microcystin-LR (MC-LR; a cyclic heptapeptide toxin) sedimentation rates and NOD-R sediment storage were estimated. NOD-R production (70-2450 microg m(-3); approximately 1 kg km(-2) per season) and sedimentation rates (particles; 0.03-5.7 microg m(-2)d(-1); approximately 0.3kg km(-2) per season) were highly variable over space and time. Cell numbers of Nodularia spumigena did not correlate with NOD-R quantities. Dissolved NOD-R comprised 57-100% of total NOD-R in the predominantly senescent, low-intensity phytoplankton blooms and seston. Unprecedentedly intensive MC-LR sedimentation (0.56 microg m(-2)d(-1)) occurred in 2004. Hepatotoxin sedimentation rates highly exceeded those of anthropogenic xenobiotics. NOD-R storage in surficial sediments was 0.4-20 microg kg(-1) ( approximately 0.1 kg km(-2)). Loss of NOD-R within the chain consisting of phytoplankton, seston and soft sediments seemed very effective.

Accumulation and effects of nodularin from a single and repeated oral doses of cyanobacterium Nodularia spumigena on flounder (Platichthys flesus L.).

Nodularin (NODLN) is a cyclic pentapeptide hepatotoxin produced by the cyanobacterium Nodularia spumigena, which occurs regularly in the Baltic Sea during the summer season. In this study flounder (Platichthys flesus L.) was orally exposed to NODLN either as a single dose or as three repeated doses 3 days apart. Liver and bile samples of the fish were taken 4 days after the last dose. Liver glutathione-S-transferase (GST) activity was also measured and the histopathology of the liver was investigated. The liver of the exposed fish was analyzed by liquid chromatography-mass spectrometry for NODLN concentration. The content of NODLN-like compounds in the bile was analyzed by enzyme-linked immunosorbent assay. NODLN exposure caused slightly incoherent liver architecture and degenerative cell changes in both groups. The mean liver GST activity was significantly higher in the repeatedly dosed flounders than in the singly dosed flounders or in the control. In conclusion, the significantly lower NODLN concentration and the increased GST activity in the liver of the repeatedly dosed flounders compared to the singly dosed flounders suggest that NODLN is rapidly detoxificated. The absence of NODLN glutathione conjugates and the low concentrations of NODLN-like compounds in the bile indicate that detoxification products disintegrate or they are rapidly excreted.

Specific strains of probiotic bacteria are efficient in removal of several different cyanobacterial toxins from solution.

The ability of specific strains of probiotic bacteria to remove the pure cyanobacterial peptide toxins microcystin-LR, -RR, -LF, and a combination of microcystins from the cyanobacterial extracts Microcystis PCC 7820 and NIES 107, as well as the cyanobacterial cytotoxin cylindrospermopsin, from aqueous solutions was assessed. The probiotic bacterial strains studied were Lactobacillus rhamnosus strains GG and LC-705, Bifidobacterium lactis strains 420 and Bb12 and Bifidobacterium longum 46, all previously shown to be effective in toxin removal. The maximum removal of microcystin-LR, 60.3%, was observed with L. rhamnosus GG, of microcystin-RR, 62.8%, and microcystin-LF, 77.4%, with L. rhamnosus LC-705, and of cylindrospermopsin, 31.6%, with B. longum 46 (toxin concentration 100mugL(-1), 37 degrees C, 24h). Several microcystins could be removed simultaneously as observed by removal of microcystins present in the cyanobacterial extracts. A combination of three probiotic strains enhanced their removal ability as compared to the removal properties of the individual strains. We conclude that specific strains of probiotic bacteria are effective in elimination of different cyanotoxins from solution.

Effect of glucose in removal of microcystin-LR by viable commercial probiotic strains and strains isolated from dadih fermented milk.

The removal of the cyanobacterial peptide toxin microcystin-LR at 4 and 37 degrees C by six commercial probiotic strains and Lactobacillus plantarum strains IS-10506 and IS-20506 isolated from dadih, Indonesian traditional fermented milk, was assessed in this study. The aim was to evaluate the main factors influencing the viability and metabolic activity of the probiotic strains, as well as their capacity to remove microcystin-LR. Both L. plantarum strains isolated from dadih, as well as Bifidobacterium lactis strains Bb12 and 420, were shown to be more resistant, and >85% remained viable in phosphate-buffered saline (PBS) solution for 48 h of incubation at either temperature, while the viability of the other four commercial bacteria decreased markedly over time. The effect of glucose on viability and removal of toxin was shown to be a strain-specific and strain-dependent property, but in general, the efficiency of microcystin-LR removal increased when glucose was added to the solution. A maximum removal of 95% was observed for L. plantarum strain IS-20506 (37 degrees C, 10 (11) colony-forming units mL(-1)) with 1-2% glucose supplementation and 75% in PBS alone.

Removal of microcystin-LR by strains of metabolically active probiotic bacteria.

The ability of specific strains of probiotic bacteria to remove the cyanobacterial peptide toxin microcystin-LR from aqueous solutions was assessed. Lactobacillus rhamnosus strains GG and LC-705, Bifidobacterium longum 46, Bifidobacterium lactis 420 and Bifidobacterium lactis Bb12 were shown to be the most effective in toxin removal among 11 tested strains. The highest removal percentage of microcystin-LR was 58.1%, observed with B. lactis Bb12 (toxin concentration 100 microg L(-1), 10(10) CFU mL(-1), 37 degrees C, 24 h). Freshly cultured bacteria were shown to be more efficient in microcystin removal than lyophilized or nonviable bacteria. Removal of microcystin-LR was shown to be dependent on both temperature and bacterial concentration. It is concluded that some of the tested strains have good potential in removing microcystins from aqueous solutions.

Analysis of nodularin-R in eider (Somateria mollissima), roach (Rutilus rutilus L.), and flounder (Platichthys flesus L.) liver and muscle samples from the western Gulf of Finland, northern Baltic Sea.

Nodularin (NODLN) is a cyanobacterial hepatotoxin that may cause toxic effects at very low exposure levels. The NODLN-producing cyanobacterium Nodularia spumigena forms massive blooms in the northern Baltic Sea, especially during the summer. We analyzed liver and muscle (edible meat) samples from common eider (Somateria mollissima), roach (Rutilus rutilus L.), and flounder (Platichthys flesus L.) for NODLN-R by liquid chromatography/mass spectrometry (LC-MS) and enzyme-linked immunosorbent assay (ELISA). Thirty eiders, 11 roach, and 15 flounders were caught from the western Gulf of Finland between September 2002 and October 2004. Eiders from April to June 2003 were found dead. The majority of samples were analyzed by LC-MS and ELISA from the same sample extracts (water:methanol:n-butanol, 75:20:5, v:v:v). Nodularin was detected in 27 eiders, nine roach, and eight flounders. Eider liver samples contained NODLN up to approximately 200 microg/kg dry weight and muscle samples at approximately 20 microg/kg dry weight, roach liver samples 20 to 900 microg NODLN/kg dry weight and muscle samples 2 to 200 microg NODLN/kg dry weight, and flounder liver samples approximately 5 to 1,100 microg NODLN/kg dry weight and muscle samples up to 100 microg NODLN/kg dry weight. The NODLN concentrations found in individual muscle samples of flounders, eiders, and roach (1-200 microg NODLN/kg dry wt) indicate that screening and risk assessment of NODLN in Baltic Sea edible fish and wildlife are required for the protection of consumer's health.

Production of antibodies against microcystin-RR for the assessment of purified microcystins and cyanobacterial environmental samples.

Microcystins (MC) are cyanobacterial hepatotoxins responsible for animal-poisoning and human health incidents. Immunoassays provide a sensitive means to detect these toxins, although cross-reactivity characteristics of different antibodies are variable, and most antibodies have been produced against MC-LR. Here, we have produced the first polyclonal antibodies against the commonly occurring variant, MC-RR, and compared them with MC-LR antibodies for the analysis of purified MCs and cyanobacterial environmental samples. Both antisera cross-reacted with all MCs tested, and with the related cyanobacterial hepatotoxin nodularin-R, but not with non-toxic cyanobacterial peptides. In general, better cross-reactivity characteristics were observed with the MC-RR antisera and limits of quantification were lower for most variants, with all MCs tested and nodularin-R having limits of quantification of 0.31 nM or below. The antisera had different affinities to mixtures containing pooled MC-LR and MC-RR, with MC-LR antisera underestimating total MC concentration when MC-RR represented over 70% of the total MC pool. Both antisera correlated well with HPLC-UV data when incorporated into ELISAs to screen previously characterised environmental samples from Aland, Finland. MC-RR antisera are useful for screening samples containing multiple MCs, and particularly for samples primarily containing MC-RR variants.

Oxidation of the cyanobacterial hepatotoxin microcystin-LR by chlorine dioxide: influence of natural organic matter.

Cyanobacteria (blue-green algae) are known producers of cytotoxic, hepatotoxic, and neurotoxic compounds with severe acute and chronic effects on vertebrates. Successful removal of these toxins in drinking water treatment is therefore of importance for public health. In the present work the oxidation of the cyanobacterial hepatotoxin microcystin-LR (MC-LR) by chlorine dioxide (ClO2) was studied at natural microcystin concentrations (10 microg L(-1)) and normal ClO2 dosages (1 mg L(-1)) in the absence and presence of natural organic matter (NOM). ClO2 was found to be rapidly consumed by fulvic and humic acids, leaving less residual ClO2 to oxidize MC-LR. Predicted decrease rates in MC-LR concentration correlated highly with experimental data both in pure water and in the presence of NOM. Rate constants determined at high ClO2 and MC-LR concentrations in pure water could be used to predict the oxidation of MC-LR at natural concentrations. Toxicity tests with a protein phosphatase inhibition assay on reaction solutions and high-performance liquid chromatography fractions revealed that PP1 enzyme inhibition emerged only from intact MC-LR, while the oxidation products, dihydroxy isomers of MC-LR, were nontoxic even at unnaturally high concentrations.

Trophic transfer of cyanobacterial toxins from zooplankton to planktivores: consequences for pike larvae and mysid shrimps.

The aim of this study was to evaluate the potentially harmful effects of zooplankton preexposed to cyanobacteria on two planktivorous animals: a fish larva (pike, Esox lucius) and a mysid shrimp (Neomysis integer). The planktivores were fed zooplankton from a natural community that had been preexposed to cell-free extract or to purified toxin (nodularin) of the cyanobacterium Nodularia spumigena, and the growth, feeding, and pellet production of the planktivores, as well as the toxin content of the pellets, were measured. In addition, radiolabeled nodularin ((3)H-dihydronodularin) was used in separate experiments to measure the vector transfer of nodularin from zooplankton to their predators. During 11-day exposures, dissolved nodularin was transferred to pike larvae and N. integer via zooplankton at very low rates of accumulation. Treatment with N. spumigena extract decreased the ingestion and feces production rates of pike larvae. With purified nodularin alone, no such effect could be observed. No effect on molting cycle length, fecal pellet production, C:N ratio, or growth of N. integer was detected. The results suggest that dissolved cyanobacterial toxins released during bloom decay can have a negative impact on feeding and, hence, on the growth of fish larvae via zooplankton, even without direct contact between cyanobacteria and the fish.

Quantitative LC-ESI-MS analyses of microcystins and nodularin-R in animal tissue--matrix effects and method validation.

The matrix effects and signal response in LC-MS analysis of six microcystins and nodularin-R were studied in mussels and liver samples from the common eider and rainbow trout. The instrumentation used in the study was a triple quadrupole MS with electrospray ionization. The results from the spiked tissue samples showed that both signal suppression and enhancement occurred. The recorded matrix effects were not severe; all studied toxins could be detected with sufficient limit of detection in all matrices. The results indicate, however, that matrix effects must be monitored for accurate quantification of microcystin and nodularin in tissue samples. Matrix effects can be studied with standard additions in the studied matrix, as was done in this study. Solid-phase extraction (SPE) resulted in a lower limit of detection compared to no cleanup in the sample preparation. SPE also prolonged the chromatographic stability. SPE cleanup is therefore strongly recommeded. Also described in this article are the chromatographic and mass spectrometric details of glutathione and cysteine conjugates, which are the detoxification products of the toxins. LC-MS analysis is suitable for detoxification studies of microcystins and nodularins. Cysteine conjugate was identified as the main detoxification product in a mussel sample that was exposed to toxic cyanobacteria in an aquarium.

Oxidation of the cyanobacterial hepatotoxin microcystin-LR by chlorine dioxide: reaction kinetics, characterization, and toxicity of reaction products.

Cyanobacteria are known producers of cytotoxins, hepatotoxins, and neurotoxins. The main toxins are microcystins, cyclic heptapeptide hepatotoxins, produced by strains of several cyanobacterial genera frequently found in eutrophied freshwaters. Due to the acute and chronic toxicity of microcystins, successful removal of these toxins in drinking water treatment processes is of increasing concern. In the present work the kinetics of microcystin-LR (MC-LR) oxidation by chlorine dioxide (ClO2) was studied with UV-spectrometry and high performance liquid chromatography (HPLC). Characterization of reaction products was performed with mass spectrometric (MS) analysis, while the toxicity of reaction products was tested with a protein phosphatase inhibition assay (PPIA). The main reaction products formed, dihydroxy isomers of MC-LR as identified by MS, were nontoxic according to the PPIA. The overall rate constant k for the reaction between MC-LR and ClO2 at 293 K and pH 5.65 was modest, k = 1.24 M(-1) s(-1), suggesting that ClO2 is not a suitable oxidant for the degradation of microcystins in drinking water treatment processes.

Eiders (Somateria mollissima) obtain nodularin, a cyanobacterial hepatotoxin, in Baltic Sea food web.

Nodularin (NODLN) is a cyclic pentapeptide hepatotoxin produced by the cyanobacterium Nodularia spumigena, which occurs regularly in the Baltic Sea during the summer season. Nodularia blooms have caused several animal kills in the Baltic Sea area, and NODLN has been found in mussels and fish caught from the northern Baltic Sea in 1996 to 2002. We analyzed liver samples of common eider (Somateria mollissima) for NODLN by enzyme-linked immunosorbent assay (ELISA) and liquid chromatography/mass spectrometry (LC-MS). Eiders feed extensively on mussel and can be exposed to NODLN by contaminated mussels. Fifteen eiders were shot and collected from three different sites in the western Gulf of Finland (northern Baltic Sea) in August and September 2002. Analyses by ELISA and LC-MS showed that eider liver samples contained 3 to 180 microg NODLN/kg dry weight and 0.1 to 5.8 microg NODLN/liver (dry wt). This is the first documentation of NODLN in seabirds and additional evidence for the transfer of NODLN in different parts of the Baltic Sea food web.

Interaction between microcystins of different hydrophobicities and lipid monolayers.

Microcystins (MC) are a group of amphiphatic peptide hepatotoxins and protein phosphatase inhibitors produced by certain cyanobacteria (blue-green algae). Microcystins are believed to require an active transport mechanism to penetrate the plasma membranes of animal cells. In this study the surface barostat technique showed that two more hydrophobic microcystins MC-LF, containing Leu and Phe, and MC-LW, containing Leu and Trp, had a higher surface activity on an egg phosphatidylcholine-cholesterol (7:3, molar ratio) monolayer as compared to that of a more hydrophilic variant MC-LR, containing Leu and Arg. Fluorescence anisotropy measurements of 1-[4-(trimethylamine)phenyl]-hexa-1,3,5-trien (TMA-DPH) were used to assess changes in the fluidity or lipid packing of model membranes in the presence of toxins. All three toxins caused a decrease in the steady-state anisotropy of TMA-DPH, suggesting that the toxins interacted with the membranes. The change in anisotropy was more pronounced for MC-LF and MC-LW than for MC-LR. Moreover, the fluorescence emission maximum of Trp in MC-LW was shifted slightly towards a shorter wavelength and the intensity was enhanced when allowed to interact with lipid vesicles, suggesting that the single Trp in MC-LW moved into a more unpolar environment when interacting with the vesicles. The differences between hydrophilic and hydrophobic microcystins could result in changes in organotropism, toxicokinetics and bioaccumulation.