Diversity and distribution of Microcystis (Cyanobacteria) oligopeptide chemotypes from natural communities studied by single-colony mass spectrometry

Martin Welker1,{dagger}, Matthias Brunke2,{ddagger}, Karina Preussel1,3, Indra Lippert3 and Hans von Döhren1

1 Technische Universität Berlin, Inst. Chemie, AG Biochemie und molekulare Biologie, Franklinstrasse 29, 10587 Berlin, Germany
2 Institut für Gewässerökologie und Binnenfischerei, Müggelseedamm 301, 12587 Berlin, Germany
3 Humboldt-Universität Berlin, Inst. Biologie, Luisenstrasse 53, 10117 Berlin, Germany

Correspondence
Martin Welker
martin.welker{at}chem.tu-berlin.de


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Microcystis sp. has been recognized in recent years as a producer of a high number of secondary metabolites. Among these, peptides that are produced by the non-ribosomal peptide synthetase pathway often show bioactivity or are toxic to humans. The production of particular peptides is specific for individual Microcystis clones, allowing their characterization as chemotypes by analysing the peptidome. The authors studied the in situ diversity of peptides and chemotypes in Microcystis communities from lakes in and around Berlin, Germany, by direct analysis of individual colonies by MALDI-TOF mass spectrometry. From 165 colonies analysed a total of 46 individual peptides could be identified, 21 of which have not been described previously. For six of the new peptides the structures could be elucidated from fragment patterns, while for others only a preliminary classification could be achieved. In most colonies, two to ten individual peptides were detected. In 19 colonies, 16 of which were identified as M. wesenbergii, no peptide metabolites could be detected. The peptide data of 146 colonies were subjected to an ordination (principal components analysis). The principal components were clearly formed by the microcystin variants Mcyst-LR, -RR and -YR, anabaenopeptins B and E/F, a putative microviridin, and a new cyanopeptolin. In the resulting ordination plots most colonies were grouped into five distinct groups, while 40 colonies scattered widely outside these groups. In some cases colonies from different lakes clustered closely, indicating the presence of similar chemotypes in the respective samples. With respect to colony morphology no clear correlation between a chemotype and a morphospecies could be established, but M. aeruginosa, for example, was found to produce predominantly microcystins. In contrast, M. ichthyoblabe colonies were mostly negative for microcystins and instead produced anabaenopeptins. The number of peptides detected in a limited number of samples and the various combinations of peptides in individual Microcystis colonies highlights the immense metabolic potential and diversity of this genus.


Abbreviations: CID, collision-induced dissociation; MALDI-TOF MS, matrix-associated laser desorption/ionization time of flight mass spectrometry; NRPS, non-ribosomal peptide synthetase; PCA, principal components analysis; PSD, post-source decay

{dagger}Present address: Institut Pasteur, Unité des Cyanobactéries, 28 rue du Dr Roux, 75724 Paris Cedex 15, France.

{ddagger}Present address: Landesamt für Natur und Umwelt, Hamburger Chaussee 25, 24220 Flintbek, Germany.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Blooms of Microcystis sp. are a familiar phenomenon encountered in eutrophic, temperate lakes during the summer. In most cases colonies with varying morphology can be distinguished microscopically. In a botanical approach Komárek & Anagnostidis (1999) distinguished 21 morphospecies in the genus Microcystis based on characteristics such as cell diameter, border and consistency of the mucilage, and number of cell division planes. Ten such morphospecies are reported from temperate waters. The major weakness of this morphological approach is the difficulty of maintaining the characteristic features when a strain is isolated from a field sample. Generally isolates cease to grow in colonies and instead form loose aggregates or grow as single cells, prohibiting the retrospective assignment of cultured strains to a species (Otsuka et al., 2000).

The genetic diversity of Microcystis (in field populations) seems to be much lower than the morphological diversity. In recent years sequence information has become available for a number of Microcystis strains, mainly from a part of the phycocyanin operon (intergenic spacer between cpcB and cpcA, IGS) and the 16S-23S-rRNA-DNA internal transcribed spacer (ITS). Both regions are considered highly variable in general, but for the genus Microcystis sequence homology was found to be very high (Otsuka et al., 1998, 1999), and it was proposed to unify Microcystis in a single species (M. aeruginosa) and to consider the other morphospecies as subspecies by application of the bacteriological code of nomenclature (Stanier et al., 1978; Otsuka et al., 2001). There is, however, at present no theoretically founded general concept about what the term ‘species' should mean in (cyano)bacteria comparable to the evolutionary species concept for ‘higher’ eukaryotes (Cohan, 2002). A third perspective to describe the diversity of Microcystis is related to the production of secondary metabolites. Microcystis – and cyanobacteria in general – can produce a broad variety of peptides of as yet unknown function for the producing organisms that are presumably synthesized by the non-ribosomal peptide synthetase (NRPS) pathway. The majority of peptides that have been isolated from Microcystis can be grouped into six classes according to structural characteristics, as summarized in Table 1. However, the number of known peptides is probably far from reflecting the total diversity of peptide compounds occurring in Microcystis. The gene clusters encoding the respective NRPSs are probably all in a size range of 25–60 kbp, dependent on the number of amino acid residues and modifying domains such as epimerization domains (von Döhren et al., 1997). Although only few data are available on the distribution of other NRPS genes (Christiansen et al., 2001) a correlation between peptide production (i.e. detectability) and presence of the respective genes can be assumed. Strains that are known to produce certain peptides do so for extended periods of time and independent of the culture conditions, giving the respective strains a distinct ‘peptide fingerprint’.


View this table:
[in this window]
[in a new window]
 
Table 1. Structural characteristics of six classes of cyanobacterial oligopeptides

 
Oligopeptides can be qualitatively detected rapidly and precisely by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). The advantage of MALDI-TOF MS compared to other chromatographic and/or mass spectral analyses is the fact that the size of the original sample can be very small and that the detection of many compounds can be achieved without a preceding separation/purification step. This allows the rapid analysis and typing of small samples like single colonies of Microcystis. This was done recently for Microcystis colonies originating from lake Wannsee, Berlin, Germany, with a focus on toxin (microcystin) production (Fastner et al., 2001). It was shown that the Microcystis bloom consisted of a number of distinct chemotypes, some of which were encountered over an extended period of time during the entire blooming season.

For the present study an investigation was conducted to compare Microcystis chemotype diversity in a number of lakes in and around Berlin during the peak blooming season. A major objective was to include as many individual peptides in the analysis as possible, which was enabled by an increased amount of data in a fragment pattern database held at AnagnosTec GmbH, Luckenwalde, Germany, and to improve the data analysis by subjecting the data matrix to ordination statistics.

We were especially interested in the distribution of chemotypes in relation to the hydrological connectivity of the lakes and in the correlation between chemotypes and morphospecies.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Sampling and preparation of Microcystis colonies.
Plankton samples were taken from seven lakes in and around Berlin, Germany, where blooms of Microcystis have been reported in previous years. Lakes Müggelsee (MU), Wannsee (WA) and Schlachtensee (SC) are situated in Berlin. Müggelsee and Wannsee are riverine, polymictic lakes flushed by the rivers Spree and Havel, respectively. These two lakes are hydrologically connected by the river Spree, which joins the river Havel some 10 km upstream of lake Wannsee. Schlachtensee is a mesotrophic, dimictic lake located close to Wannsee in the forest of Grunewald but it is not connected naturally to the lakes in the course of the Havel. Lakes Wesensee (WS), Herrensee (HE), Pehlitzsee (PE) and Parsteiner See (PA) are located 40–80 km north-east of Berlin in the biosphere reservation Schorfheide-Chorin. All the lakes are considered mesotrophic to eutrophic and are polymictic with the exception of Schlachtensee.

In July and August 2000 a net sample (30 µm) was taken from each lake at near-shore sites by vertical hauls from about 1·5 m depth to the surface, and samples were stored in a cooling bag. In the laboratory individual colonies were immediately isolated from the samples under a dissecting microscope and washed in sterile water several times before being photographed with an analogue and a digital camera. Colonies were assigned to morphospecies according to Komárek & Anagnostidis (1999) during the isolation procedure and a second and third time independently from slides and digital images, respectively. The colonies were then placed on a stainless steel MALDI-TOF MS sample plate and allowed to dry at room temperature. During the isolation procedure the intention was to select representatively the different morphotypes present in the samples. As performance of the MS analysis depends also on the size of the colonies, smaller colonies (<50 µm diameter) were omitted to ensure good performance in MS analysis.

MS analysis.
Extraction of the cells was performed directly on the template with the matrix solution consisting of a saturated solution of {alpha}-cyanohydroxycinnamic acid (HCA) in a mixture of water, ethanol and acetonitrile (1 : 1 : 1) acidified with 0·1 % (v/v) trifluoroacetic acid.

MALDI-TOF MS analyses were performed on a Voyager-DE PRO Biospectrometry workstation (Applied Biosystems) equipped with a reflecting ion mirror, post-source decay (PSD) and collision-induced dissociation (CID) options as described previously (Welker et al., 2003). Peaks matching known masses were further analysed by fragmentation (PSD spectra; Kaufmann et al., 1993) and comparison of the resulting fragment spectra to theoretical fragment patterns (Erhard et al., 1999). Mass signals of unknown compounds with sufficient intensities (>103 counts in accumulated spectra) were analysed and fragment patterns were compared to a database containing fragment patterns of known and partly characterized cyanobacterial peptides. For some colonies mass spectra were obtained also in negative-ion extraction mode.

Since the naming of cyanobacterial oligopeptides is not done according to a broadly accepted nomenclature many synonyms exist for peptides with similar structural characteristics. Therefore in this study we refer to the characteristics listed in Table 1 for classes of peptides.

Data processing and statistical analysis.
MALDI-TOF MS can be considered only as semi-quantitative, although attempts have been made to use it for the quantification of a diversity of compounds including microcystins (Ling et al., 1998; Fastner et al., 2002). In our study, where unknown mixtures of a variety of compounds were analysed, effects influencing mass signal intensity like quenching or ionization efficiency could not be estimated. On the other hand, in the mass spectra patterns of relative signal intensities were evident. To allow a statistical analysis of the data by a multivariate technique the mass spectra were therefore processed without losing quantitative patterns (i.e. counts in mass spectra).

The detection of chlorophyll derivatives (pheophytin at m/z=871·6 Da and/or pheophorbide at m/z=593·3 Da) was set as minimum criterion for the inclusion of respective colonies in the further analysis. When no respective mass signals were obtained, the colonies were considered either as too small or already decaying at the time of preparation, or inefficiently extracted. All mass spectra were first processed with the de-isotoping option of the DataExplorer 4.0 software based on a generic peptide formula (Welker et al., 2002). Resulting mass spectra were checked for the presence of sodium and potassium adduct peaks, and in the cases where these could be identified, the respective monoisotopic counts were added to the protonated mass signal. In a next step all mass signals that could not be identified as a peptide metabolite were excluded, i.e. glycolipid signals, matrix-associated signals, or those that could not be identified as peptides with certainty.

Principal components analysis (PCA) is a form of indirect gradient analysis (Jongman et al., 1997) and was conducted to explore differences and gradients in the peptide occurrence in colonies. The compounds included in the PCA were selected when they met either of the following criteria: (1) present in at least one colony mass spectrum with a minimal relative intensity of 70 %, or (2) present in at least three colonies with a minimal relative intensity of 30 %. Prior to statistical analysis, the variability of the absolute intensities of the remaining peptide mass signals was reduced by normalizing peak heights to the interval [0;1]. The skewness of the data was reduced by a logarithmic transformation. The ordination was performed by using CANOCO 4.02 for Windows (ter Braak & Smilauer, 1998).

The sum of counts per particular peptides processed in this way was than considered as the colony-specific metabolic ‘fingerprint’.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In all samples several Microcystis morphospecies could be identified based on the criteria defined by Komárek & Anagnostidis (1999). Most morphospecies were found in all samples, but in some lakes certain morphospecies predominated, like M. wesenbergii in Herrensee and M. aeruginosa in Schlachtensee. M. viridis was rare in the size fraction that was suitable for MS analysis and only from one colony could a mass spectrum of satisfactory quality be obtained.

From a total of more than 190 colonies that were analysed by MS, 165 yielded mass spectra with identifiable mass signals of chlorophyll a derivatives, glycolipids or peptides. Mass signals in the spectra were exclusively recorded from singly charged ions. In 146 out of 165 at least one mass signal was recorded that could be identified as a peptide. The remaining 19 colonies gave no peptide mass signal; 16 of these colonies were M. wesenbergii, two M. botrys and one M. flos-aquae. In most cases, mass signals of several compounds per colony generally were intense enough to perform PSD fragmentation. For many colonies the laser energy had to be reduced sharply to avoid a signal overflow.

In most of the mass spectra a number of peaks with high intensity was recorded (Fig. 1) and individual compounds could be identified clearly. Besides mass signals of known compounds, in many colonies a number of less intense mass signals were recorded that could not be assigned to any known (peptide) mass. Differences in mass spectra of individual colonies from a single sample were immediately evident (Fig. 1). In some colonies, such as MU023 and MU037, the same peptides could be detected, e.g. Mcyst-LR, Mcyst-RR, cyanopeptolin W, and a chlorinated variant of cyanopeptolin W. In MU037, anabaenopeptin E/F and kasumigamide were additionally detected but were absent in MU023. The latter two peptides in turn were detected in a colony, MU035, which was found negative for microcystins and cyanopeptolins. Several mass signals occurred frequently but could not be characterized unambiguously and therefore were excluded from the further analysis, e.g. M+H=1174·6 Da or M+H=1270·6 Da.



View larger version (34K):
[in this window]
[in a new window]
 
Fig. 1. MALDI-TOF mass spectra of Microcystis colonies isolated from Müggelsee in August 2000. Colony ID is indicated to the right above each trace. Numbers in plain type refer to masses (M+H) of identified peaks (Table 2); numbers in italics refer to mass signals that occurred repeatedly but could not be further characterized.

 
Other mass spectra, on the other hand, were similar to each other also with respect to relative peak intensities (Fig. 2). A chemotype that produced mainly the linear tetrapeptide kasumigamide and an oxygen-deficient variant of it was found in lakes Müggelsee, Pehlitzsee and Parsteiner See. Nearly identical mass spectra were recorded for several colonies from Schlachtensee that produced Mcyst-LR, Mcyst-RR and a new peptide that could be classified as a putative cyanopeptolin (M+H=1054·5 Da).



View larger version (27K):
[in this window]
[in a new window]
 
Fig. 2. MALDI-TOF mass spectra of Microcystis colonies isolated from Müggelsee (MU), Pehlitzsee (PE), Parsteiner See (PA) and Schlachtensee (SC) in August/September 2000. Colony ID is indicated to the right above each trace. Numbers refer to masses (M+H) of identified peaks (Table 2).

 
PSD fragment spectra were obtained at least once for each peptide listed in Table 2. From 46 peptides, 23 could be identified as previously described metabolites. For six peptides an amino acid sequence could be proposed; 13 peptides could be identified as members of a known peptide class, while the remaining four peptides could not be characterized further with certainty. In some cases, a distinction between two structural variants with identical masses could not be made for all colonies in which the respective mass signals were recorded, as for anabaenopeptins E and F, for cyanopeptolins A and D, and for demethylMcyst variants, e.g. [Asp3]Mcyst-LR and [Dha7]Mcyst-LR.


View this table:
[in this window]
[in a new window]
 
Table 2. Peptide compounds detected in colonies of Microcystis sp. from lakes in Berlin/Brandenburg by MALDI-TOF MS

Compounds marked with an asterisk are new peptides, either congeners of a known peptide class (Table 1) or peptides that could not be assigned to a peptide class. When ‘new’ is given as reference, the structure was elucidated by PSD fragmentation in the present study and the amino acid sequence is mentioned in the text. The numbers following the name then refer to the molecular mass [M]. References: (1) Lawton et al. (1999); (2) Murakami et al. (1995); (3) Matsuda et al. (1996); (4) Neumann et al. (1997); (5) Ishida et al., 2000; (6) Ishida & Murakami (2000); (7) Murakami et al. (1997); (8) Shin et al. (1998); (9) Erhard et al. (1999); (10) Itou et al. (1999); (11) Martin et al. (1993); (12) Erhard 1999; (13) Harada et al. (1991); (14) Okino et al. (1993); (15) Botes et al. (1985); (16) Kiviranta et al. (1992); (17) Kusumi et al. (1987); (18) Namikoshi et al. (1995); (19) Ishida et al. (1997); (20) Tsukamoto et al. (1993); (21) Harada et al. (1993).

 
New variants of the cyanopeptolin class were identified by series of fragments containing Ahp (Table 1). For two new variants, the data for structure elucidation are shown in Fig. 3. In the original mass spectrum recorded in positive-ion extraction mode four mono-isotopic peaks were recorded (see Fig. 1, MU029). Two of them, 896·5 Da and 930·4 Da, differed by 33·9 Da, corresponding to the net difference of a chlorine atom. The mass signal of the latter peptide had an isotopic distribution typical for singly chlorinated compounds with a pseudomolecular peak at 932·4 Da. Associated peaks were recorded with a mass difference of 18 Da, corresponding to a water abstraction that occurs with many cyanopeptolins (M. Welker, unpublished). In negative-ion extraction mode two intense peaks were recorded with a mass difference of 78 Da from the [M+H]+ peaks, corresponding to the net mass of a sulphate group that is abstracted in positive-ion extraction mode. By PSD fragmentation the amino acid sequence could be elucidated by assigning detected fragment peaks to theoretical fragments. Immonium ions (or fragments thereof) were detected of arginine (70 Da), leucine (86 Da), phenylalanine (120 Da) and methyltyrosine (150 Da). Together with the fragments 243 Da ([Ahp+Phe]–H2O+H) and 215 Da ([Ahp+Phe]–H2O–CO+H) this indicated a cyanopeptolin structure. Further masses could be assigned to fragments: 765 Da, M–Leu–H2O+H; 694 Da, M–Arg–H2O–CO+H; 672 Da, M–MTyr–H2O–CO+H; 588 Da, M–MTyr–Leu–H2O+H; 553 Da, M–Phe–MTyr–H2O+H; 525 Da, M–Phe–MTyr–H2O–CO+H; 439 Da, [Phe+MTyr+Leu]–H2O+2H and [Ahp+Phe+MTyr]–H2O–CO+2H; 420 Da, [Phe+MTyr+Leu]–H2O–CO+2H and [Ahp+Phe+MTyr]–H2O–CO+2H; 345 Da, [MTyr+Leu+Thr]–H2O–CO+H; 328 Da, [Arg+Thr+GA]–H2O+H; 307 Da, [Phe+MTyr]–H2O+2H; 172 Da, [Thr+GA]–H2O+H. MTyr refers to an N-methylated tyrosine residue. Combining the mass spectral information with that on published cyanopeptolins, we propose the structure given schematically in Fig. 3(d) for the two peptide compounds.



View larger version (36K):
[in this window]
[in a new window]
 
Fig. 3. Partial mass spectra of a Microcystis colony (for the full spectrum see Fig. 1). (a) Mass spectrum in negative-ion extraction mode. (b) Positive-ion extraction mode, with numbers referring to [M+H]+. (c) PSD-CID fragment spectrum of the precursor ion mass signal M+H=878·45 Da. (d) Proposed amino acid sequences of the two compounds detected in (a) and (b); ‘O’ refers to the ester bond between the {beta}-hydroxy group of threonine and the carboxy group of Leu and GA to glyceric acid. Arrows in (a) and (b) annotate differences in masses due to loss or addition of the atoms or molecules indicated. For an assignment of fragment mass signals to amino acid sequences see text.

 
Similar structure determination was accomplished for the chlorinated variant of cyanopeptolin W (M+H=1013·5 Da). In the respective fragment spectra, part of the signals was identical to those observed with cyanopeptolin W while part of the signals was shifted by 34 Da. The chlorination could be subsequently localized at the N-methyltyrosine moiety of the molecule. Both variants always co-occurred, but only in colonies from lakes Müggelsee and Wannsee.

A new cyanopeptolin with a mass of M+H=973·5 Da (cyanopeptolin 972) was repeatedly detected and analysed by PSD from colonies originating from three sampled lakes. The proposed amino acid sequence of this peptide is [Arg-Ahp-Leu/Ile-MTyr-Val-O-Thr]-Asp-HA (HA, hexanonic acid).

Another new cyanopeptolin detected in colonies from lakes Parsteiner See, Wesensee and Wannsee had a mass of M+H=1007·4 Da (cyanopeptolin 1006) and the proposed amino acid sequence is [Arg-Ahp-Phe-MTyr-Val-O-Thr]-Asp-HA.

New aeruginosins were recognized by intense fragment peaks m/z=140 Da and m/z=122 Da, indicative of the Choi-immonium ion and its dehydrated derivative. For one aeruginosin (M+H=603·3 Da) an amino acid sequence can be proposed as [Hpla-Leu-Choi-Argininal] whereas for another aeruginosin (M+H=679·3 Da) a full structure elucidation was not possible.

Individual peptides or peptide classes were detected with pronounced differences in frequency (Table 3). The most common individual peptides were microcystin variants that occurred in 30 % (Mcyst-YR) to more than 60 % (Mcyst-LR) of all colonies. Differences among sampled lakes were considerable: in more than 90 % of the colonies from Schlachtensee Mcyst-LR was detected while in colonies of Herrensee the proportion was below 20 %. In Herrensee, however, the Microcystis community was dominated by M. wesenbergii, which did not produce any detectable peptides (10 out of 16) colonies. The frequency of other individual peptides in colonies from the sampled lakes differed more than observed for microcystins. For example, anabaenopeptins B and E/F were not detected in any colony of lakes Herrensee, Parsteiner See, Pehlitzsee and Schlachtensee, while some 40 % of the colonies from Müggelsee and Wannsee produced these peptides. A similar pattern was found for the chlorinated cyanopeptolin W, which occurred exclusively in lakes Müggelsee and Wannsee. Other peptides, such as aeruginosin 602 and cyanopeptolin 1020, were found in colonies from most of the lakes but with a lower frequency compared to microcystins.


View this table:
[in this window]
[in a new window]
 
Table 3. Frequency of occurrence as a percentage of individual peptides (upper part of table) or peptide classes (lower part) in Microcystis colonies isolated from lakes near and in Berlin

All values are given as a percentage of n colonies. Only the mass spectral data of those colonies that fulfilled the criteria described in the text were included. ND, Not detected in any colony of that sample.

 
When regarding classes of peptides instead of individual congeners the picture changes considerably. Microcystins were still the most frequent peptides but followed shortly by cyanopeptolins, which were, with 19 congeners, the most diverse of the peptide classes in our dataset. For anabaenopeptins as a peptide class the pattern of frequency is very similar to that of two of its congeners, indicating the low structural variability of detected members of that class. Aeruginosins and microviridins were both detected in about 20 % of the colonies and in colonies from all samples except for microviridins, which were not detected in Herrensee. The least abundant class of peptides was microginins, detected only in some 8 % of the colonies.

The number of individual peptides detected in single colonies varied considerably, from zero to ten, with a median of three. Likewise the number of peptide classes varied. The number of peptide classes (Table 1) that were detected in a single colony ranged from zero (11 % of the colonies) to five (1·2 % of the colonies). Besides these extremes, the distribution was the following: one class of peptides in 14 %, two classes in 40 %, three classes in 25 % and four classes in 7·2 % of the 165 colonies. Thus a majority of more than 60 % of the colonies produced peptides of two or three peptide classes.

Co-production of peptides from different classes showed some patterns of preference (Table 4). Microcystins, for example, were co-produced with cyanopeptolins in 57 % of the colonies. Microcystins and cyanopeptolins in turn were coproduced with microviridins in 20 % and 19 % of the colonies, respectively. Rare combinations were microcystins and microginins (1 %), aeruginosins and microviridins (4 %), and aeruginosins and anabaenopeptins (4 %).


View this table:
[in this window]
[in a new window]
 
Table 4. Co-occurrence of oligopeptides of different classes in Microcystis colonies from seven lakes around Berlin

Absolute numbers of a total of 165 colonies are given in the upper part, and percentages in the lower part.

 
In Fig. 4 the results of the PCA are presented in the form of ordination plots of factors 1 and 2 (F1xF2) showing the colony scores with assignment of origin (Fig. 4a), and the colony scores with assignment of morphospecies (Fig. 4b). The peptide loadings (plot not shown) were dominated by a few peptides that were therefore the main determinants of the colony chemotypes. In the negative F1 (referred to as ‘F1–’) area the three main Mcyst variants (-LR, -RR and -YR) were the predominant peptides. Anabaenopeptins B and E/F together with microcystilide A, microginin FR1/3 and cyanopeptolin 1048 (F1+/F2– sector) were inversely correlated with Mcyst-YR and -LR. In the F1+/F2+ sector and along the F1+ axis a number of peptides with low loadings were grouped, of which kasumigamide, aeruginosin 602 and cyanopeptolin A/D were the most determinant ones.



View larger version (31K):
[in this window]
[in a new window]
 
Fig. 4. Principal component analysis ordination plots (F1xF2) of individual Microcystis colonies based on their peptide content as analysed by MALDI-TOF MS. The data matrix consisted of peptide occurrence data in 146 Microcystis colonies collected from seven lakes around Berlin, Germany. Colonies are labelled with reference to their origin (a) and to their morphospecies (b).

 
In the factorial F1xF2 plane (Fig. 4a) colony scores were located in several groups, some of which were not very distinct, and overlapping. Four main groups were characterized by microcystin variants (groups I, II, III and IV) in variable combinations and together with other peptides. Group I is characterized by a high content of Mcyst-RR and -LR and cyanopeptolin 1053. Especially the colonies from Schlachtensee (n=17, see also Fig. 2) clustered tightly. Intermingled were colonies from lakes Parsteiner See (n=5), Pehlitzsee (n=3), Wesensee (n=1) and Wannsee (n=1), indicating the presence of the same chemotype in these lakes. Group II is characterized by high contents of Mcyst variants -LR, -RR, -YR and -(H4)YR together with cyanopeptolin 1020, cyanopeptolin 1006 and microviridin 1667, but with a less dense clustering when compared to group I. Colonies grouped there mainly originated from Wannsee and Pehlitzsee (9 each). The colonies in group III were characterized by high contents of Mcyst-LR and -YR and microviridin 1694. The major contribution to this group came from Wesensee (6), Wannsee, Müggelsee and Schlachtensee (3, 3 and 2 colonies, respectively). Group IV was formed by colonies from Schlachtensee (10), Wannsee (2) and Parsteiner See (4). This group was determined by Mcyst-LR and -demethyl-LR, and aeruginopeptin 95B.

A distinct cluster of colonies without microcystins was found in group V, which is characterized by high contents of anabaenopeptins B and E/F and of microcystilide A. Nine of the colonies originated from Wannsee, four from Müggelsee, and two from Wesensee. Beside these main groups some 40 colonies scattered on the plot or formed small groups that were distinguishable only after enlargement of the respective area (in the F1xF2 scores). The colonies that are found in the enlarged area contained neither microcystins nor anabaenopeptins. One small group (VI) was formed by colonies containing kasumigamide (Fig. 2) and another one by colonies containing aeruginosin 602 and micropeptin B (VII). Other colonies that seemed not to be linked to any larger distinct group either contained rare peptides or contained peptides in a rare combination.

The colonies of all lakes scattered strongly along the PCA factors; no clear correlation of chemotypes to particular lakes was evident and differences in chemotype diversity between lakes thus were gradual. The majority of colonies from Schlachtensee, as one extreme, clustered tightly in groups I and IV (n=17 and n=10, respectively), while only two were placed in group III and two not associated to any group. On the other hand, out of 25 colonies from Müggelsee only nine were placed in groups I to V and the remaining 16 colonies scattered widely. Fig. 4(b) shows the same F1xF2 factorial plot with the colonies labelled with respect to their morphospecies. The few M. wesenbergii morphotypes with detectable peptides (n=5) were distributed irregularly, giving no indication of a related chemotype. M. flos-aquae colonies were placed in groups I, II and V, and with three in group VI. About half of M. ichthyoblabe colonies were placed in group V and only 20 % were found in groups I–IV. For M. aeruginosa morphotypes the pattern was reversed, with 77 % of the colonies placed in groups I–IV and none in group V, and the remaining 23 % distributed widely. A similar pattern was obtained for M. botrys, with 67 % of the colonies in groups I–IV, and none in group V.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Origin of the detected peptides
The MS approach was successful in detecting a multitude of known and new peptides from very small samples of cyanobacterial cells. Most mass spectra were satisfactory with respect to signal intensity and accuracy, i.e. absolute intensities of more than 104 counts and baseline separation of isotopic peaks up to a mass range of about 1100 Da. The possibility to further characterize a single mass signal by PSD fragmentation from the same sample without a further purification step testified to the analytical power of MALDI-TOF MS for peptide analysis (Li et al., 1999). Nonetheless, a residual uncertainty concerning the identity of the peptides could not be avoided with the analytical set-up for several less intense mass signals, since the structural diversity of cyanobacterial oligopeptides is so high that molecular mass alone is not sufficient for an unambiguous identification of an individual compound in many cases. For most Microcystis colonies some hundred cells seem to be sufficient to detect the major peptide metabolites. With respect to the multitude of detected peptides, quenching effects during the ionization phase are likely to have influenced the actual appearance of the resulting mass spectra. Detectability of individual peptides depends partly on the efficiency with which they can be protonated (Karas et al., 2000). As shown in Fig. 3, part of a molecule can be abstracted in the flight path in dependence of the ion extraction mode. On the other hand, the mass spectra were not dominated by only a few peptide mass signals, and individual peptides generated intense mass signals in one colony while producing much smaller ones in others, for example anabaenopeptin E/F in MU032 and MU037.

When field samples are analysed for peptides it has to be addressed whether the detected peptides (and other metabolites) actually originate from a specific cyanobacterial taxon, in our case Microcystis, or potentially from other organisms that can not be separated mechanically in the sample preparation process. Microcystis colonies from natural waters are hardly ever ‘pure’: colonies are populated by heterotrophic prokaryotes (Brunberg, 1999), other cyanobacteria, diatoms, green algae, or protozoa like Vorticella. Nonetheless, for several reasons we are confident that the peptides that we detected by MALDI-TOF MS are indeed produced by Microcystis and that the peptide patterns are metabolic ‘fingerprints' of Microcystis clones. Firstly, nearly all the peptides we detected are either known or are congeners of known peptides that have been isolated from axenic and unialgal Microcystis strains. Secondly, similar peptides have not been reported from eukarytic algae or protozoa so far and although heterotrophic bacteria in principle can be considered to produce oligopeptides, the peptides that were detected in Microcystis are structurally different from these.

The production of several non-ribosomal peptides by an individual clone has been shown repeatedly for axenic strains of Microcystis, with varying combinations of peptide classes (Martin et al., 1993; Fastner et al., 2001). There is, of course, a residual uncertainty whether the colonies we isolated and analysed were indeed clonal or inseparable conglomerates of several clones. By applying several washing steps and visual control a potential respective bias was minimized.

Diversity of peptide structures and chemotypes
Considerable efforts have been undertaken in recent decades to isolate and identify cyanobacterial peptides, mainly for pharmacological purposes, and a wealth of structures has been published (Namikoshi & Rinehart, 1996; Burja et al., 2001). Nonetheless, the number of individual peptides, many of them not described previously, that could be detected in a limited number of samples was surprising. For cyanopeptolins, for example, some 70 congeners are known. Still, that number seemingly reflects only a small part of the actual number of naturally occurring congeners of this peptide class: in 165 colonies 12 new structural variants of cyanopeptolins alone were detected. A similar situation was encountered for microviridins, where we found four putative new variants.

The highest number of individual peptide signals detected in a colony was ten, while the median was three. Accordingly, the number of peptide classes detected in individual colonies has to be regarded as a minimum number, although the production of two or three peptide classes seems to be common in Microcystis isolates.

The diversity of chemotypes and their distribution in and among sampled lakes showed considerable differences, although the PCA ordination results were determined by the occurrence of a few quantitatively important peptides in the colonies. Considering the biochemistry of non-ribosomal peptide synthesis (i.e. the presence of NRPS gene clusters) a more complex picture arises. For example a clone that produces two microcystin variants and a third (non-microcystin) peptide clearly differs from a clone producing the same microcystins but another additional peptide from a different class.

Distribution of chemotypes in samples and relation to morphospecies
Among the Microcystis samples were some with high chemotype diversity like Wannsee and Müggelsee and some with low chemotype diversity like Schlachtensee und Wesensee. The high diversity and similarity in Müggelsee and Wannsee was probably due to the hydrological connection of these two lakes by the rivers Spree and Havel. Schlachtensee, in contrast, is hydrologically isolated, being fed mainly by groundwater. However, the assumption that isolation of the ecosystem leads to a reduction in diversity following competitive exclusion should be examined carefully while so little is known about the interaction of Microcystis clones and the role of peptide metabolites. A striking correlation of a morphospecies to a single group in the PCA plot or chemotype was not obtained for any morphospecies. Only for M. ichthyoblabe was a vague correlation to group V evident. When all groups with microcystin-producing colonies were combined it became evident that, for example, M. aeruginosa was predominantly toxic, while M. ichthyoblabe was not, supporting previous findings (Fastner et al., 2001). But for all morphospecies microcystin-producing and -non-producing strains were found (except for M. viridis), and taking into account the total number of peptides no clear correlation of morphospecies to a single chemotype is obtained. Thus, peptide production is arguably, to a certain degree, independent of the genes determining the morphology. Nonetheless, if only the production of mirocystins is considered a prediction could be made for some morphospecies with fair accuracy.

What is the adaptive value of cyanobacterial oligopeptides?
The proximate explanation for the presence of diverse peptide structures that differ sometimes only in one amino acid residue (e.g. microcystins -LR and -RR) or even only in a single atom (e.g. chlorine) is related to the non-ribosomal peptide synthesis. The activation and incorporation of different amino acids is dependent on the amino acid residues that line the binding pocket of the adenylation domain of individual modules (Stachelhaus et al., 1999; Challis et al., 2000). Further variation in peptide structures is achieved by N-methylation and chlorination, for example, which are not performed on all the synthesized molecules.

The production of a diversity of peptides by individual Microcystis clones and the physiological/ecological function of non-ribosomal peptides in cyanobacteria are not well understood yet. A number of hypotheses on their role as grazing protection, allelochemicals or infochemicals have been proposed. It has to be considered that the production of structurally diverse peptides seems to be more beneficial compared to a single or a few structural variant(s) since natural selection did not lead to the dominance of a few chemotypes, not even in a narrow geographical range as in our study. Individual peptides or peptide classes seem not to be essential for the fitness of Microcystis clones – otherwise the communities should be dominated by a few chemotypes or by chemotypes producing certain peptides. But producing and non-producing strains obviously generally coexist in most environments without any clear superiority of one over the other.

For further studies on chemotype diversity, sequence information on the respective gene clusters is required to evaluate the correlation of genes and peptide diversity. A good correlation between the presence of microcystins and detection of the mcyB gene in Microcystis colonies has been established (Kurmayer et al., 2002).

Several studies have shown the presence of the mcy gene in strains that cluster more or less independently in phylogenetic trees (Neilan et al., 1995, 1997; Otsuka et al., 1999). This recently led to the hypothesis that the distribution of mcy genes among Microcystis strains is only partly the result of vertical gene transfer and to a certain degree due to horizontal gene transfer, lateral recombination and gene loss (Mikalsen et al., 2003), processes that might also be important factors for the distribution of other non-ribosomal peptides among Microcystis clones. For other genes it has been hypothesized that horizontal gene transfer may occur within taxonomic units like genera (Rudi et al., 1998; Barker et al., 2000).

At present, there are no data available on clonal dynamics in field populations, and the methods required to follow the dynamics of clones in experiments and under field conditions, like real time-PCR (Kurmayer & Kutzenberger, 2003) or DGGE (Janse et al., 2003), have emerged only recently. It can be concluded that a prediction or a modelling of peptide (namely Mcyst) occurrence and dynamics is currently only possible in combination with long-term data series (Welker et al., 2003), owing to the lack of understanding of the interaction between Microcystis clones with differing peptide patterns.


   ACKNOWLEDGEMENTS
 
Thanks to Thomas Hartzendorf for help in colony picking and processing. Jírí Komárek gave useful advice on the classification of morphospecies, which is greatly appreciated. Marcel Erhard and Jutta Fastner are both thanked for their useful comments on mass spectra and interpretation of the data. Further, suggestions made by the reviewers were highly appreciated. This study was funded by the Deutsche Forschungsgemeinschaft (DFG-grant Do 270/10-3).


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Barker, G. L. A., Handley, B. A., Vacharapiyasophon, P., Stevens, J. R. & Hayes, P. K. (2000). Allele-specific PCR shows that genetic exchange occurs among genetically diverse Nodularia (Cyanobacteria) filaments in the Baltic Sea. Microbiology 146, 2865–2875.[Abstract/Free Full Text]

Botes, D. P., Wessels, P. L., Kruger, H., Runnegar, M. T. C., Satikarn, S., Smith, R. J., Barba, J. C. J. & Williams, D. H. (1985). Structural studies on cyanoginosins-LR, -YR, -YA, and -YM, peptide toxins from Microcystis aeruginosa. J Chem Soc Perkin Trans 1, 2747–2748.[CrossRef]

Brunberg, A. K. (1999). Contribution of bacteria in the mucilage of Microcystis spp. (Cyanobacteria) to benthic and pelagic bacterial production in a hypereutrophic lake. FEMS Microbiol Lett 29, 13–22.

Burja, A. M., Banaigs, B., Abou-Mansour, E., Burgess, J. G. & Wright, P. C. (2001). Marine cyanobacteria – a prolific source of natural products. Tetrahedron 57, 9347–9377.[CrossRef]

Challis, G. L., Ravel, J. & Townsend, C. A. (2000). Predictive, structure-based model of amino acid recognition by nonribosomal peptide synthetase adenylation domains. Chem Biol 7, 211–224.[CrossRef][Medline]

Christiansen, G., Dittmann, E., Ordorika, L. V., Rippka, R., Herdman, M. & Börner, T. (2001). Nonribosomal peptide synthetase genes occur in most cyanobacterial genera as evidenced by their distribution in axenic strains of the PCC. Arch Microbiol 178, 452–458.[CrossRef]

Cohan, F. M. (2002). What are bacterial species? Annu Rev Microbiol 56, 457–487.[CrossRef][Medline]

Erhard, M. (1999). Matrix assisted laser desorption/ionisation-time of flight Massenspektrometrie von Cyanobakterien und Untersuchungen zur Biosynthese und Funktion der Microcystine. Thesis, Technische Universität Berlin.

Erhard, M., von Döhren, H. & Jungblut, P. (1999). Rapid identification of the new anabaenopeptin G from Planktothrix agardhii HUB 011 using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Rapid Commun Mass Spectrom 13, 337–343.[CrossRef][Medline]

Fastner, J., Erhard, M. & von Döhren, H. (2001). Determination of oligopeptide diversity within a natural population of Microcystis spp. (Cyanobacteria) by typing single colonies by matrix-assisted laser desorption ionization-time of flight mass spectrometry. Appl Environ Microbiol 67, 5069–5076.[Abstract/Free Full Text]

Fastner, J., Codd, G. A., Metcalf, J. S., Woitke, P., Wiedner, C. & Utkilen, H. (2002). An international intercomparision exercise for the determination of purified microcystin-LR and microcystins in cyanobacterial field material. Anal Bioanal Chem 374, 437–444.[CrossRef][Medline]

Harada, K.-I., Ogawa, K., Matsuura, K. & 7 other authors (1991). Isolation of two toxic heptapeptide microcystins from an axenic strain of Microcystis aeruginosa, K-139. Toxicon 29, 479–489.[CrossRef][Medline]

Harada, K.-I., Mayumi, T., Shimada, T., Suzuki, M., Kondo, F. & Watanabe, M. F. (1993). Occurrence of four depsipeptides, aeruginopeptins, together with microcystins from toxic cyanobacteria. Tetrahedron Lett 34, 6091–6094.[CrossRef]

Ishida, K. & Murakami, M. (2000). Kasumigamide, an antialgal peptide from the cyanobacterium Microcystis aeruginosa. J Org Chem 65, 5898–5900.[CrossRef][Medline]

Ishida, K., Matsuda, H., Murakami, M. & Yamaguchi, K. (1997). Micropeptins 478-A and -B, plasmin inhibitors from the cyanobacterium Microcystis aeruginosa. J Nat Prod 60, 184–187.[CrossRef]

Ishida, K., Kato, T., Murakami, M., Watanabe, M. & Watanabe, M. F. (2000). Microginins, zinc metalloproteases inhibitors from the cyanobacterium Microcystis aeruginosa. Tetrahedron 56, 8643–8656.[CrossRef]

Itou, Y., Suzuki, S., Ishida, K. & Murakami, M. (1999). Anabaenopeptins G and H, potent carboxypeptidase A inhibitors from the cyanobacterium Oscillatoria agardhii (NIES-595). Bioorg Med Chem Lett 9, 1243–1246.[CrossRef][Medline]

Janse, I., Meima, M., Kardinaal, W. E. A. & Zwart, G. (2003). High-resolution differentiation of cyanobacteria by using rRNA-internal transcribed spacer denaturing gradient gel electophoresis. Appl Environ Microbiol 69, 6634–6643.[Abstract/Free Full Text]

Jongman, R. H. G., ter Braak, C. J. F. & van Tongeren, O. R. F. (1997). Data Analysis in Community and Landscape Analysis. Cambridge: Cambridge University Press.

Karas, M., Glückmann, M. & Schäfer, J. (2000). Ionization in matrix-assisted laser desorption/ionization: singly charged molecular ions are the lucky survivors. J Mass Spectrom 35, 1–12.[CrossRef][Medline]

Kaufmann, R., Spengler, B. & Lützenkirchen, F. (1993). Mass spectrometric sequencing of linear peptides by product-ion analysis in a reflectron time-of-flight mass spectrometer using matrix-assisted laser desorption ionization. Rapid Commun Mass Spectrom 7, 902–910.[Medline]

Kiviranta, J., Namikoshi, M., Sivonen, K., Evans, W. R., Carmichael, W. W. & Rinehart, K. L. (1992). Structure determination and toxicity of a new microcystin from Microcystis aeruginosa strain 205. Toxicon 30, 1093–1098.[CrossRef][Medline]

Komárek, J. & Anagnostidis, K. (1999). Cyanoprokaryota 1. Teil: Chroococcales. In Süwasserflora von Mitteleuropa. Edited by H. Ettl, G. Gärtner, H. Heynig & D. Mollenhauer. Jena: Gustav Fischer.

Kurmayer, R. & Kutzenberger, T. (2003). Application of real-time PCR for quantification of microcystin genotypes in a population of the cyanobacterium Microcystis sp. Appl Environ Microbiol 69, 6723–6730.[Abstract/Free Full Text]

Kurmayer, R., Dittmann, E., Fastner, J. & Chorus, I. (2002). Diversity of microcystin genes within a population of the toxic cyanobacterium Microcystis spp. in Lake Wannsee (Berlin, Germany). Microb Ecol 43, 107–118.[CrossRef][Medline]

Kusumi, T., Ooi, T., Watanabe, M. M., Takahashi, H. & Kakisawa, H. (1987). Cyanoviridin RR, a toxin from the cyanobacterium (blue-green alga) Microcystis viridis. Tetrahedron Lett 28, 4695–4698.[CrossRef]

Lawton, L. A., Morris, L. A. & Jaspars, M. (1999). A bioactive modified peptide, aeruginosinamide, isolated from the cyanobacterium Microcystis aeruginosa. J Org Chem 64, 5329–5332.[CrossRef]

Li, L., Garden, R. W., Romanova, E. V. & Sweedler, J. V. (1999). In situ sequencing of peptides from biological tissues and single cells using MALDI-PSD/CID analysis. Anal Chem 71, 5451–5458.[CrossRef][Medline]

Ling, Y.-C., Lin, L. & Chen, Y.-T. (1998). Quantitative analysis of antibiotics by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Rapid Commun Mass Spectrom 12, 317–327.[CrossRef][Medline]

Martin, C., Oberer, L., Ino, T., König, W. A., Busch, M. & Weckesser, J. (1993). Cyanopeptolins, new depsipeptides from the cyanobacterium Microcystis sp. PCC 7806. J Antibiot 46, 1550–1556.[Medline]

Matsuda, H., Okino, T., Murakami, M. & Yamaguchi, K. (1996). Aeruginosins 102-A and B, new thrombin inhibitors from the cyanobacterium Microcystis viridis (NIES-102). Tetrahedron 52, 14501–14506.[CrossRef]

Mikalsen, B., Boison, G., Skulberg, O. M., Fastner, J., Davies, W., Gabrielsen, T. M., Rudi, K. & Jakobsen, K. S. (2003). Natural variation in the microcystin synthetase operon mcyABC and impact on microcystin production in Microcystis strains. J Bacteriol 185, 2774–2785.[Abstract/Free Full Text]

Murakami, M., Ishida, K., Okino, T., Okita, Y., Matsuda, H. & Yamaguchi, K. (1995). Aeruginosins 98-A and B, trypsin inhibitors from the blue-green alga Microcystis aeruginosa (NIES-98). Tetrahedron Lett 36, 2785–2788.[CrossRef]

Murakami, M., Shin, H. J., Matsuda, H., Ishida, K. & Yamaguchi, K. (1997). A cyclic peptide, anabaenopeptin B, from the cyanobacterium Oscillatoria agardhii. Phytochemistry 44, 449–452.[CrossRef]

Namikoshi, M. & Rinehart, K. L. (1996). Bioactive compounds produced by cyanobacteria. J Ind Microbiol 17, 373–384.

Namikoshi, M., Sun, F., Choi, B. W., Rinehart, K. L., Carmichael, W. W., Evans, W. R. & Beasley, V. R. (1995). Seven more microcystins from Homer Lake cells: application of the general method for structure assignment of peptides with -dehydroamino acid unit(s). J Org Chem 60, 3671–3679.

Neilan, B. A., Jacobs, D. & Goodman, A. E. (1995). Genetic diversity and phylogeny of toxic cyanobacteria determined by DNA polymorphisms within the phycocyanin locus. Appl Environ Microbiol 61, 3875–3883.[Abstract]

Neilan, B. A., Jacobs, D., DelDot, T., Blackall, L. L., Hawkins, P. R., Cox, P. T. & Goodman, A. E. (1997). rRNA sequences and evolutionary relationships among toxic and nontoxic cyanobacteria of the genus Microcystis. Int J Syst Bacteriol 47, 693–697.[Abstract/Free Full Text]

Neumann, U., Forchert, A., Flury, T. & Weckesser, J. (1997). Microginin FR1, a linear peptide from a water bloom of Microcystis aeruginosa. FEMS Microbiol Lett 153, 475–478.[CrossRef]

Okino, T., Murakami, M., Haraguchi, R., Munekata, H. & Matsuda, H. (1993). Micropeptins A and B, plasmin and trypsin inhibitors from the blue green alga Microcystis aeruginosa. Tetrahedron Lett 34, 8131–8134.[CrossRef]

Otsuka, S., Suda, S., Li, R. H., Watanabe, M., Oyaizu, H., Matsumoto, S. & Watanabe, M. M. (1998). 16S rDNA sequences and phylogenetic analyses of Microcystis strains with and without phycoerythrin. FEMS Microbiol Lett 164, 119–124.[CrossRef]

Otsuka, S., Suda, S., Li, R. H., Watanabe, M., Oyaizu, H., Matsumoto, S. & Watanabe, M. M. (1999). Phylogenetic relationships between toxic and non-toxic strains of the genus Microcystis based on 16S to 23S internal transcribed spacer sequence. FEMS Microbiol Lett 172, 15–21.[CrossRef][Medline]

Otsuka, S., Suda, S., Li, R., Matsumoto, S. & Watanabe, M. M. (2000). Morphological variability of colonies of Microcystis morphospecies in culture. J Gen Appl Microbiol 46, 39–50.[Medline]

Otsuka, S., Suda, S., Shibata, S., Oyaizu, H., Matsumoto, S. & Watanabe, M. M. (2001). A proposal for the unification of five species of the cyanobacterial genus Microcystis Kutzing ex Lemmermann 1907 under the Rules of the Bacteriological Code. Int J Syst Evol Microbiol 51, 873–879.[Abstract]

Rudi, K., Skulberg, O. M. & Jakobsen, K. S. (1998). Evolution of cyanobacteria by exchange of genetic material among phyletically related strains. J Bacteriol 180, 3453–3461.[Abstract/Free Full Text]

Shin, H. J., Matsuda, H., Murakami, M. & Yamaguchi, K. (1998). Anabaenopeptins E and F, two new cyclic peptides from the cyanobacterium Oscillatoria agardhii (NIES-204). J Nat Prod 60, 139–141.

Stachelhaus, T., Mootz, H. D. & Marahiel, M. A. (1999). The specificity-conferring code of adenylation domains in nonribosomal peptide synthetases. Chem Biol 6, 493–505.[CrossRef][Medline]

Stanier, R. Y., Sistrom, W. R., Hansen, T. A. & 9 other authors (1978). Proposal to place the nomenclature of the cyanobacteria (blue-green algae) under the rules of the international code of nomenclature of bacteria. Int J Syst Bacteriol 28, 335–336.

ter Braak, C. J. F. & Smilauer, P. (1998). CANOCO reference manual and user's guide to CANOCO for Windows. Ithaca, USA: Microcomputer Power.

Tsukamoto, S., Painuly, P., Young, K., Yang, X. & Shimizu, Y. (1993). Microcystilide A: a novel cell-differentiation-promoting depsipeptide from Microcystis aeruginosa NO-15-1840. J Am Chem Soc 115, 11046–11047.

von Döhren, H., Keller, U., Vater, J. & Zocher, R. (1997). Multifunctional peptide synthetases. Chem Rev 97, 2675–2705.[CrossRef][Medline]

Welker, M., Fastner, J., Erhard, M. & von Döhren, H. (2002). Application of MALDI-TOF MS in cyanotoxin research. Environ Toxicol 17, 367–374.[CrossRef][Medline]

Welker, M., von Döhren, H., Täuscher, H., Steinberg, C. E. W. & Erhard, M. (2003). Toxic Microcystis in shallow lake Müggelsee (Germany) – dynamic, distribution, diversity. Arch Hydrobiol 157, 227–248.[CrossRef]

Received 28 November 2003; revised 2 February 2004; accepted 9 February 2004.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Welker, M.
Articles by von Döhren, H.
Articles citing this Article
PubMed
PubMed Citation
Articles by Welker, M.
Articles by von Döhren, H.
Agricola
Articles by Welker, M.
Articles by von Döhren, H.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2004 Society for General Microbiology.