Uwe Neumann

Optimised extraction of microcystins from field samples : A comparison of different solvents and procedures

Cyanobacterial cells of water blooms from various lakes dominated by Microcystis spp., Planktothrix agardhii or Planktothrix rubescens as well as three cultures of cyanobacteria were extracted with either pure methanol, water, 5% acetic acid, 75% aqueous methanol or a sequential extraction using methanol followed by water. It was found that methanol did not extract microcystins quantitatively from lyophilised cyanobacterial field samples. Quantitative extraction could only be obtained by the use of 75% methanol or the sequential extraction. For the extraction of microcystins from lyophilised cyanobacterial field samples we recommend the use of 75% methanol because of better reproducibility and faster procedure compared to the sequential extraction.

Microcystins (hepatotoxic heptapeptides) in german fresh water bodies

In 1995 and 1996 a total of 55 German water bodies with different trophic states were investigated for the presence of potentially microcystin‐producing cyanobacteria. The seston biomass of over 500 samples was analyzed by HPLC to assess microcystin content. The highest microcystin content per dry weight was found when Planktothrix rubescens was dominant, followed by Planktothrix agardhii and Microcystis spp. The microcystin to chlorophyll‐a ratio mostly varied between 0.1 and 0.5, with maxima of 1–2. In over 70% of the samples from pelagic water, particulate microcystin concentrations were below 10 μg L−1. By contrast, spatial and temporal concentrations varied by 4 orders of magnitude (2–25,000 μg L−1) at bathing sites in 1997 during a mass development of Microcystis spp. The seasonal peak of development of Microcystis spp. occurred from June to September and of Planktothrix agardhii from September to November. Planktothrix rubescens, however, was almost perennial. The seasonal levels of these cyanobacteria were clearly reflected by microcystin concentrations in the water bodies. ©1999 John Wiley & Sons, Inc. Environ Toxicol 14: 13–22, 1999

Co-Occurrence of Non-toxic (Cyanopeptolin) and Toxic (Microcystin) Peptides in a Bloom of Microcystis sp. from a Chilean Lake

A cyanobacterial bloom occurring in 1998 in lake Tres Pascualas (Concepción/Chile) was found to be dominated by Microcystis sp. The bloom contained both non-toxic (cyanopeptolin-type) and hepatotoxic (microcystin-type) peptides. Cyanopeptolin structure of the non-toxic peptides (called cyanopeptolin VW-1 and VW-2, respectively) was revealed by matrix assisted laser desorption ionization mass spectrometry (MALDI-TOF-MS) of whole cells, showing dominant molecular ions at m/z = 975 and m/z 995, respectively. On post source decay (PSD), both cyanopeptolins showed fragments deriving from Ahp-Phe-MTyr (3-amino-6-hydroxy-2-piperidone), the characteristic partial structure of cyanopeptolins. The amounts of each of the two cyanopeptolins could only roughly be estimated to be >0.1% of bloom material dry weight. In addition the blooms contained microcystins (20 microg/g bloom dry weight as determined by RP-HPLC, 13 microg/g according to ELISA determination). MALDI-TOF-MS revealed several structural variants of microcystin: MCYST-RR (microcystin with Arg and Arg, indicated by m/z 1,038 and confirmed by PSD revealing a m/z = 135 fragment deriving from the Adda side chain, MCYST-FR (microcystin with Phe and Arg, indicated by m/z = 1,015). The presence of [Asp(3)]-MCYST-LR (microcystin with Leu and Arg, Asp non-methylated, indicated by m/z 981), and [Asp(3)]-MCYST-YR (microcystin with Tyr and Arg, Asp non-methylated, indicated by m/z 1,031) were likely. The relative amounts of the peptides varied between February, April, and May. Whole cell extracts from the bloom material revealed specific enzyme inhibitory activities. The serin-proteases trypsin, plasmin, elastase were inhibited, assumable due to the cyanopeptolins found. Elastase and the cysteine-protease papain were not inhibited, inhibitions of protein kinase and glutathione S-transferase (GST) were low. Strong inhibition was observed with protein-phosphatase-1, likely due to the microcystins present in the samples.

Estimation of the microcystin content in cyanobacterial field samples from German lakes using the colorimetric protein–phosphatase inhibition assay and RP‐HPLC

The protein–phosphatase (PP‐1) inhibition potency of microcystin was employed to estimate the microcystin content in cyanobacterial field samples from three German eutrophic lakes collected mainly in summer 1996. The microcystin content of lyophilized samples extracted with aqueous methanol was determined on the basis of the colorimetric PP‐1 inhibition assay with p‐nitrophenyl phosphate as substrate. With the test system described, an IC50 of 1 ng/mL reaction volume (∼0.6 nM PP‐1) was determined for microcystin‐LR standard, corresponding to 20 ng microcystin‐LR per milliliter sample volume. The microcystin content of field samples estimated on the basis of the PP‐1 inhibition assay was compared to that obtained by reversed phase high performance liquid chromatography (RP‐HPLC), and a good correlation was observed (r=0.9202, P<0.0001). The mean difference in microcystin content based on PP‐1 inhibition was determined to be 27.9±22.9% (n=50) related to measurements based on RP‐HPLC analysis. There was no observation of false positive or false negative results in this study. The results confirm the suitability of the PP‐1 inhibition assay with regard to a preliminary screening of cyanobacterial field samples for microcystin. ©1999 John Wiley & Sons, Inc. Environ Toxicol 14: 23–29, 1999

Lipopolysaccharide and porin of Roseobacter denitrificans, confirming its phylogenetic relationship to the α-3 subgroup of proteobacteria

Roseobacter denitrificans has rough (R)-type lipopolysaccharide, containing 2-keto-3-deoxyoctonate but no hepatoses. Its lipid A has a glucosamine-containing, phosphorylated backbone. It contains the rare 3-oxotetradecanoic (3-oxomyristic) acid as the only amide-bound fatty acid and ester-bound 3-hydroxydecanoic acid, this pattern being characteristic for the alpha-3 subgroup of Proteobacteria. Treatment of the major outer-membrane protein (porin, apparent molecular mass 88 kDa) of Roseobacter denitrificans with EDTA (2 mM, 30 degrees C, 20 min) resulted in the dissociation of the oligomers into monomers (apparent molecular mass 35 kDa). EDTA-sensitive dissociation has so far been observed only within the alpha-3 subgroup of Proteobacteria. The 12 N-terminal amino acids of the monomers exhibit sequence homology with the porins of Rhodobacter capsulatus, Rhodobacter sphaeroides and Rhodopseudomonas blastica. Renaming of Roseobacter denitrificans as Rhodobacter denitrificans is suggested.

Porin and porin-associated protein (PAP) of Rhodospirillum rubrum FR1

The porin of Rhodospirillum rubrum FR1 was found in the outer membrane as a complex with a relatively small (32 kDa) porin-associated protein (PAP). The porin moiety of the complex consisted of a trimer which revealed a mainly β-sheet structure, while the porin-PAP complex also contained a significant α-helical portion. Isolated PAP exhibited a mostly α-helical structure. The porin-PAP complex had the same single-channel conductivity (2.8 nS) as the isolated porin, demonstrating that PAP did not affect the porin channel size. Differential extraction of the cell wall fraction with N,N-dimethyldodecylamine N-oxide revealed that PAP stabilized the porin in the outer membrane. EDTA caused dissociation of the porin-PAP complex, indicating that divalent cations were involved in formation of this complex.

Characterization of Porin from Roseobacter denitrificans

Porin from Roseobacter denitrificans was isolated and purified to homogeneity. The pore characteristics from this marine bacterium were compared to those of its phylogenetically closely related freshwater bacteria Rhodobacter capsulatus 37b4, Rhodobacter sphaeroides and Rhodopseudomonas blastica. The porin formed weakly cation-selective, general diffusion pores in lipid bilayer membranes. High transmembrane potentials caused channel closing in steps that were of one or two thirds of the initial on-steps indicating that the porin of R. denitrificans comprised three more or less independent channels similar to PhoE and OmpC of Escherichia coli and the porin of Rhodobacter capsulatus. Prediction of the secondary structure of the 36 N-terminal amino acid residues indicated two transmembrane β-strands similar to those of the porins of Rhodobacter capsulatus 37b4 and Rhodopseudomonas blastica. Differences of the single channel conductivities between the porin of R. denitrificans and those of the related freshwater bacteria show that R. denitrificans evolved porin channels that are well adapted to the marine habitat.