Open Access

Convergent evolution of [D-Leucine1] microcystin-LR in taxonomically disparate cyanobacteria

  • Tânia Keiko Shishido1,
  • Ulla Kaasalainen2,
  • David P Fewer1,
  • Leo Rouhiainen1,
  • Jouni Jokela1,
  • Matti Wahlsten1,
  • Marli Fátima Fiore3,
  • João Sarkis Yunes4,
  • Jouko Rikkinen2 and
  • Kaarina Sivonen1Email author
BMC Evolutionary Biology201313:86

DOI: 10.1186/1471-2148-13-86

Received: 12 November 2012

Accepted: 9 April 2013

Published: 19 April 2013

Abstract

Background

Many important toxins and antibiotics are produced by non-ribosomal biosynthetic pathways. Microcystins are a chemically diverse family of potent peptide toxins and the end-products of a hybrid NRPS and PKS secondary metabolic pathway. They are produced by a variety of cyanobacteria and are responsible for the poisoning of humans as well as the deaths of wild and domestic animals around the world. The chemical diversity of the microcystin family is attributed to a number of genetic events that have resulted in the diversification of the pathway for microcystin assembly.

Results

Here, we show that independent evolutionary events affecting the substrate specificity of the microcystin biosynthetic pathway have resulted in convergence on a rare [D-Leu1] microcystin-LR chemical variant. We detected this rare microcystin variant from strains of the distantly related genera Microcystis, Nostoc, and Phormidium. Phylogenetic analysis performed using sequences of the catalytic domains within the mcy gene cluster demonstrated a clear recombination pattern in the adenylation domain phylogenetic tree. We found evidence for conversion of the gene encoding the McyA2 adenylation domain in strains of the genera Nostoc and Phormidium. However, point mutations affecting the substrate-binding sequence motifs of the McyA2 adenylation domain were associated with the change in substrate specificity in two strains of Microcystis. In addition to the main [D-Leu1] microcystin-LR variant, these two strains produced a new microcystin that was identified as [Met1] microcystin-LR.

Conclusions

Phylogenetic analysis demonstrated that both point mutations and gene conversion result in functional mcy gene clusters that produce the same rare [D-Leu1] variant of microcystin in strains of the genera Microcystis, Nostoc, and Phormidium. Engineering pathways to produce recombinant non-ribosomal peptides could provide new natural products or increase the activity of known compounds. Our results suggest that the replacement of entire adenylation domains could be a more successful strategy to obtain higher specificity in the modification of the non-ribosomal peptides than point mutations.

Keywords

Adenylation domain Phylogeny Substrate specificity Gene conversion Point mutations Recombination

Background

Non-ribosomal peptides are an important class of secondary metabolites produced by a range of bacteria and fungi [13]. These peptides have many biotechnological and pharmaceutical applications such as the antibiotics penicillin [4] and daptomycin [5], and the anticancer bleomycin [6]. They are synthesized on large modular non-ribosomal peptide synthetase (NRPS) and polyketide synthase (PKS) enzyme complexes. Each NRPS module is responsible for the recognition and incorporation of an amino acid during elongation of the peptide intermediate. The basic NRPS module has a condensation (C), adenylation (A), and peptidyl carrier protein (PCP) domains. The adenylation domain is responsible for the selection and activation of amino acids in the form of aminoacyl adenylates [7]. It is followed by the peptidyl carrier protein and condensation domains, the former holding the activated amino acid and the latter making a peptide bond between two adjacent amino acids [2]. In addition, auxiliary enzymes may be present and have activities such as the epimerization, cyclisation, N-methylation, formylation, and reduction of amino acids [1, 3]. The large variety of non-proteinogenic amino acids and hydroxyl acids that can be incorporated and further modified by tailoring enzymes allows the production of highly complex peptides.

Microcystins are the most frequently reported cyanobacterial toxins in aquatic blooms. They are small cyclic heptapeptides with extensive variation in amino acid residue composition and are commonly produced by planktonic strains in fresh and brackish water [8]. Microcystins are potent inhibitors of serine/threonine protein phosphatases (PP1 and PP2A) and can cause human intoxication, tumor promotion, and death [9]. The general structure of microcystin is cyclo(−D-Ala1-X2-D-MeAsp3-Z4-Adda5-D-Glu6-Mdha7-) (Figure 1a and b). A hybrid NRPS-PKS enzyme complex directs the synthesis of microcystin and is encoded in the 55–55.6 kb (mcy) gene cluster (Figure 1c) [1013].
https://static-content.springer.com/image/art%3A10.1186%2F1471-2148-13-86/MediaObjects/12862_2012_Article_2321_Fig1_HTML.jpg
Figure 1

Microcystin chemical structures and biosynthetic enzymes. (a) [D-Leu1] microcystin-LR and (b) [D-Met1] microcystin-LR. MeAsp: D-erythro-β-methylaspartic acid. Adda: (2S,3S,8S,9S)-3-,amino-9-methoxy-2,6,8-trimethyl-10-phenyldeca-4,6-dienoic acid. Mdha: N-methyldehydroalanine. X and Z are the highly variable positions. (c) NRPS and PKS involved in microcystin synthesis from Microcystis. The order of the enzymes corresponds to the assembly of microcystin (each position of the microcystin structure is indicated under each adenylation domain). N. N-methyltransferase, Ep. epimerization, TE. thioesterase domains are shown in white and the PCP. peptidyl carrier protein domain in black. Regions included in the phylogenetic analysis are indicated in grey, in addition of their respective PCP domains.

Phylogenetic studies indicate that the mcy gene cluster has an ancient origin among cyanobacteria [1417]. Other studies suggest that horizontal transfer, gene loss, and recombination events in the microcystin gene cluster explain the distribution and variation of the genes among the closely related Microcystis spp. [1820]. Recombination events affecting the adenylation domain have been described in genes encoding McyA1, McyB1 and McyC [16, 18, 19, 2125]. Positive selection acting on the adenylation domains of McyB1 and McyC was reported as the possible cause of the large number of microcystin variants produced by cyanobacteria [24]. Deletion of the entire N-methyltransferase domain of mcyA in Anabaena or point mutations in this gene in Microcystis were associated with the absence of N-methylation in the microcystins produced by these strains [23, 26]. Furthermore, recombination in the same region (mcyA 1 ) was related to the synthesis of microcystins containing 2-amino-2-butenoic acid (Dhb) in some strains of the genus Planktothrix[25]. Such genetic rearrangements, positive selection, and recombination events act to increase the chemical variability of microcystins found in nature. However, the production of the same rare microcystin variant in disparate taxonomic lineages of cyanobacteria raises questions about the genetic mechanism underlying this phenomenon [2731]. Evolutionary diversification of NRPS and PKS pathways is achieved through genetic mechanisms such as recombination, duplication, fusion or fission of genes, deletion or substitutions of domains, circular permutations, gene loss and horizontal gene transfer [32, 33]. Here, we show that the production of a rare [D-Leu1] microcystin (MC) variant in three distantly related genera of cyanobacteria is the result of three independent evolutionary events leading to convergence on the same chemical structure. There is an interest in engineering non-ribosomal peptide biosynthetic pathways in order to increase the production levels of known compounds or create new bioactive compounds [3]. Our results suggest that the replacement of entire adenylation domains might be a more successful strategy for changing substrate specificity in the engineering of NRPS than point mutations.

Results

Production of the [D-Leu1]MC variant by taxonomically disparate cyanobacteria

We documented the production of the rare [D-Leu1]MC variant in morphologically disparate cyanobacteria (Figure 2). These cyanobacteria were assigned to the genera Nostoc, Phormidium and Microcystis. Nostoc sp. strain UK89IIa was identified based on morphology and 16S rRNA gene sequence similarity (Figures 2 and 3). The Phormidium sp. CENA270 strain has a thick sheath, tangled filaments and small (≤6.3 μm) cells that indicate the presence of similar characters with Phormidium, even though the cells in the trichomes are distinctly wider (5–6.3 μm) than their length (1–3 μm). The 16S rRNA gene sequence of strain CENA270 was 98% of Phormidium sp. DVL1003c, which also produces microcystin. These two strains form a clade with Phormidium, Lyngbya, Oscillatoria and Oscillatoriales strains in the 16S rRNA trees (Figure 3; Additional file 1: Figure S1). The two Microcystis strains had previously been identified [27, 34].
https://static-content.springer.com/image/art%3A10.1186%2F1471-2148-13-86/MediaObjects/12862_2012_Article_2321_Fig2_HTML.jpg
Figure 2

Photomicrographs of the studied strains. (a) Microcystis aeruginosa NPLJ-4. (b) Microcystis sp. RST 9501. (c) Nostoc sp. UK89IIa. (d) Phormidium sp. CENA270.

https://static-content.springer.com/image/art%3A10.1186%2F1471-2148-13-86/MediaObjects/12862_2012_Article_2321_Fig3_HTML.jpg
Figure 3

Phylogenetic analysis of the 16S rRNA gene. Maximum-likelihood tree based on the 16S rRNA gene. Bootstrap values above 50% from 1000 maximum-likelihood bootstrap replicates are given at the nodes. The studied strains are in bold and highlighted. Symbols for microcystin variants in position one: [D-Ala1]; ■ [D-Leu1]; [D-Met1]; [D-Ser1].

Nostoc sp. UK89IIa produced four variants of microcystins, while Phormidium sp. CENA270 produced five variants (Table 1). All of the detected microcystin variants produced by these two strains contained D-Leu, and none of the strains produced detectable levels of microcystins that contained D-Ala. The two Microcystis strains produced at least twelve variants altogether (Table 1, Additional file 1: Table S2 and Additional file 2: Table S1). However, 97% of the microcystin variants in Microcystis aeruginosa NPLJ-4 and 80% of the microcystin produced by Microcystis sp. RST 9501 contained D-Leu at position 1 (Table 1, Additional file 2: Table S1). We carried out further chemical analysis to characterize the new microcystin variants produced by these strains.
Table 1

Chemical variants of microcystin detected in LC-MS/MS analyses

Microcystin variant

[M+H]+(m/z)

Microcystis aeruginosa NPLJ-4

Microcystis sp. RST9501

Nostoc sp. UK89IIa

Phormidium sp. CENA270

[Leu1]MC-LR

1037

76

76

96

31

[Leu1, Asp3]MC-LR

1023

19

<1

2

18

[Met1]MC-LR

1055

1

18

-

-

[Met1, Asp3]MC-LR

1041

2

-

-

-

[Leu1, Dha7]MC-LR

1023

1

-

1

-

[Leu1]MC-HilR

1051

-

3

-

-

[Leu1]MC-HphR

1085

-

-

2

-

[Leu1]MC-LHar

1051

-

-

-

7

[Leu1]MC-RR

1080

-

-

-

31

[Leu1, Asp3]MC-RR

1066

-

-

-

13

Values given as the percentage of relative area of the peaks in the protonated molecular ion chromatograms for the main microcystins variants produced by the studied strain. Hil, homoisoleucine. Hph, homophenylalanine. Har, homoarginine.

In all known microcystin variants, the amino acid configuration is D in position 1 and L in positions 2 and 4 (Figure 1a and b). Therefore, the chirality of the amino acids in positions 1, 2 and 4 of the major microcystin variants produced by the studied strain was investigated by using deuterated acid hydrolysis reagents (DCl, D2O). This confirmed that the hydrolyzed microcystins contained D-Leu (Additional file 1: Table S2).

Production of [Met1]MC-LR by Microcystisspp

Mass spectrometry strongly suggested that Microcystis strains NPLJ-4 and RST 9501 produced new microcystin variants that contained Met instead of D-Leu (Table 1, Figure 1b). In order to confirm these results, Microcystis sp. RST 9501 cells were grown with 32S and 34S as the sole source of sulfur. LC-MS revealed an increase in [Met1]MC-LR protonated ion mass from m/z 1,055.5 (control containing 32S) to 1,057.5 (34S-labelled samples), indicating the presence of a sulfur atom in the microcystin, and the same differences could be visualized in the fragmentation analysis (Additional file 1: Figure S2). The product ion spectra of [Met1]MC-LR could be compared with the spectra of [D-Leu1]MC-LR and -RR (Additional file 1: Figures S2a and S3a and b). The results confirmed the presence of a new variant of [Met1]MC in the two Microcystis strains NPLJ-4 and RST 9501. However, Microcystis sp. RST 9501 produced 14 times more [Met1]MC-LR than strain NPLJ-4. On the other hand, Microcystis aeruginosa NPLJ-4 produced a larger diversity of microcystin variants, but most of them in trace amounts, such as [Val1]MC-LR and [Phe1]MC-LR.

Phylogenetic analysis of microcystin catalytic domains

In order to understand the order and timing of genetic events leading to the production of the rare [D-Leu1]MC variant, we conducted phylogenetic analysis of the mcy gene cluster. Phylogenetic trees based on concatenated mcyD and mcyE gene sequences and the housekeeping 16S rRNA gene were robust and found to have a similar pattern (Additional file 1: Figure S4). Strains producing the [D-Leu1]MC variant do not group together but instead group with strains that produce the [D-Ala1]MC variant in both trees (Additional file 1: Figure S4). We constructed alignments based on the NRPS catalytic domains encoded in each of the microcystin biosynthetic genes (Figures 1c, 4 and 5). The condensation domains and peptidyl carrier proteins of each module encoded in the mcy gene cluster grouped together according to their encoding gene and were placed in separate clades in the phylogenetic tree (Figure 4a and b). A similar pattern was also observed for the adenylation domains of McyG, McyE, and McyB2 (Figure 5), and the epimerization domains of McyA (Figure 4c). However, the adenylation domain sequences of McyA1, McyA2, McyB1, and McyC are mixed and do not form separate clades (Figure 5). A phylogenetic tree was constructed using adenylation domain sequences of McyA2 and other NRPSs obtained from BLASTp searches of the nr database at NCBI (Figure 6). The McyA2 adenylation domains of the Microcystis strains were grouped in a single well-supported clade, irrespective of whether they produced [D-Leu1]MCs or [D-Ala1]MCs variants. By contrast, the McyA2 adenylation domain of Phormidium sp. CENA270 and Nostoc sp. UK89IIa did not group together with other McyA2 adenylation domains. Instead, the Phormidium sp. CENA270 McyA2 adenylation domain was placed in the same clade of McyB1 adenylation domain of Microcystis strains. The McyA2 adenylation domain of Nostoc sp. UK89IIa grouped with adenylation domain sequences from Nostoc punctiforme PCC73102 and Nostoc sp. GSV224.
https://static-content.springer.com/image/art%3A10.1186%2F1471-2148-13-86/MediaObjects/12862_2012_Article_2321_Fig4_HTML.jpg
Figure 4

Conservative evolutionary history of domains surrounding the McyA 2 adenylation domain in the microcystin biosynthetic gene cluster. Maximum-likelihood tree based on amino acid sequences of the (a) condensation domain, (b) peptidyl carrier protein domain and (c) epimerization domain within the mcy gene cluster. Phylogenetic tree inferred using MEGA 5. Bootstrap values above 50 per cent from 1000 respectively neighbor-joining, maximum parsimony and maximum-likelihood bootstrap replicates are given at the nodes. The studied strains are in bold and indicated with *.

https://static-content.springer.com/image/art%3A10.1186%2F1471-2148-13-86/MediaObjects/12862_2012_Article_2321_Fig5_HTML.jpg
Figure 5

Evolutionary history of adenylation domains of microcystin biosynthetic gene cluster. Maximum-likelihood tree based on amino acids sequences of adenylation domain within mcy gene cluster. Phylogenetic tree inferred using MEGA 5. Bootstrap values above 50 per cent from 1000 respectively neighbor-joining, maximum parsimony and maximum-likelihood bootstrap replicates are given at the nodes. The studied strains are in bold and indicated with *.

https://static-content.springer.com/image/art%3A10.1186%2F1471-2148-13-86/MediaObjects/12862_2012_Article_2321_Fig6_HTML.jpg
Figure 6

Independent evolutionary history of McyA 2 adenylation domain. Maximum-likelihood tree based on amino acids sequences inferred using MEGA 5. Bootstrap values above 50 per cent from 1000 respectively neighbor-joining, maximum parsimony and maximum-likelihood bootstrap replicates are given at the nodes. The studied strains are in bold and indicated with *.

Substrate specificity of the McyA2adenylation domain

Conservation of the tertiary structure of adenylation domains makes it possible to predict amino acid binding pockets and consequently the substrate specificity. All strains that produce the [D-Ala1]MC variants analyzed in this study have identical predicted McyA2 adenylation domain binding pocket sequences, with the exception of Planktothrix agardhii NIVA-CYA 126/8 (Table 2). There are only two conserved residues (D235 and K517) in the predicted binding pocket in strains producing [D-Ala1] or [D-Leu1]MC variants. However, Nostoc sp. UK89IIa and Phormidium sp. CENA270 have identical predicted binding pockets, differing substantially from those of strains producing [D-Ala1]MC variants (Table 2). Microcystis aeruginosa NPLJ-4 and Microcystis sp. RST 9501 have identical predicted binding pockets, which differ by three amino acids at positions 301, 330, and 331 in comparison to strains that activate L-Ala (Table 2). Strains producing [D-Ala1]MC variants have polar amino acids (Thr or Ser) at position 330 of the binding pocket, while strains producing [D-Leu1]MC variants have hydrophobic non-polar amino acids (Ile and Val) at this position (Table 2).
Table 2

Predicted binding pockets of the adenylation domain of McyA 2

Strain

Binding pocket

Score %(a)

Predicted(a)

Activated

 

235

236

239

278

299

301

322

330

331

517

Anabaena sp. 90

D

L

F

N

N

A

L

T

Y

K

100

Ala

Ala

Microcystis aeruginosa NIES-843

-

-

-

-

-

-

-

-

-

-

100

Ala

Ala

Microcystis aeruginosa PCC7806

-

-

-

-

-

-

-

-

-

-

100

Ala

Ala

Microcystis aeruginosa UV027

-

-

-

-

-

-

-

-

-

-

100

Ala

Ala

Microcystis aeruginosa K-139

-

-

-

-

-

-

-

-

-

-

100

Ala

Ala

Nostoc sp. 152

-

-

-

-

-

-

-

-

-

-

100

Ala

Ala*

Phormidium sp. DVL1003c

-

-

-

-

-

-

-

-

-

-

100

Ala

Ala

Planktothrix rubescens NIVA-CYA 98

-

-

-

-

-

-

-

-

-

-

100

Ala

Ala

Planktothrix agardhii NIVA-CYA 126/8

-

-

-

-

-

-

-

S

-

-

90

Ala

Ala

Microcystis sp. RST 9501

-

-

-

-

-

G

-

I

C

-

70

Cys

Leu, Met

Microcystis aeruginosa NPLJ-4

-

-

-

-

-

G

-

I

C

-

70

Cys

Leu, Met ±

Nostoc sp. UK89IIa

-

A

W

F

L

G

N

V

V

-

100

Leu

Leu

Phormidium sp. CENA270

-

A

W

F

L

G

N

V

V

-

100

Leu

Leu

The adenylation domain of McyA2 is responsible for amino acid activation in position one of microcystins.

(a) Prediction obtained using the NRPSpredictor2 program [72, 73]. Trace amounts of *Serine and ± Valine or Phenylalanine detected in LC-MS/MS.

Genetic variations in the McyA2adenylation domain

Microcystis spp. NPLJ-4 and RST 9501, Nostoc sp. UK89IIa and Phormidium sp. CENA270 produce the same rare [D-Leu1]MC variant [27, 31]. Recombination events affecting the substrate specificity of the mcyA 2 gene were detected in Nostoc sp. UK89IIa and Phormidium sp. CENA270 by four different methods (Table 3). Breakpoints with statistical support within the mcyA gene were identified in UK89IIa and CENA270 (Figure 7a). The identified predicted breakpoints were visualized in a recombination breakpoint distribution plot (Figure 7b). A lengthy region replaced in a recombination event in the mcyA gene in Nostoc sp. UK89IIa (1029 bp) and Phormidium sp. CENA270 (167 and 707 bp) was observed in the present study (Figure 7a). The predicted binding pockets responsible for amino acid selection and activation were found within this region in both cases.
Table 3

Recombination events affecting the substrate specificity of the adenylation domain of the mcyA 2 gene

Strain

Parents*

Average P-value

 

Major

Minor

RDP

GENECON

BootScan

MaxChi

Chimaera

SiScan

3Seq

Nostoc sp. UK89IIa

N152

M9501

1.423x10-9

5.882x10-9

2.306x10-3

5.938x10-12

7.331x10-3

7.789x10-16

2.192x10-24

Phormidium sp. CENA270

±P1003c

±Ana90

1.242x10-6

1.793x10-18

3.361x10-6

6.848x10-14

1.200x10-3

7.894x10-27

2.410x10-105

 

¦P1003c

¦M027

4.209x10-3

-

-

2.274x10-5

9.115x10-4

6.746x10-12

-

Analysis was performed in the RDP3 program and breakpoint positions are indicated in Figure 7a. The P-value cutoff was chosen as 0.05 and the best signals for recombination are associated with the lowest P-values.

* M7806 = Microcystis aeruginosa PCC7806; M139 = Microcystis aeruginosa K-139; N152 = Nostoc sp. 152; M9501 = Microcystis sp. RST 9501; P1003e = Phormidium sp. DVL1003c; Ana90 = Anabaena sp. 90; M027 = Microcystis aeruginosa UV027. Breakpoint positions: ± 432 (undetermined) to 1139 and ¦ 216 to 383 (undetermined).

https://static-content.springer.com/image/art%3A10.1186%2F1471-2148-13-86/MediaObjects/12862_2012_Article_2321_Fig7_HTML.jpg
Figure 7

Recombination breakpoints in the mcyA 2 gene detected by the RDP3 program. (a) Breakpoint positions detected in three cyanobacterial strains. The position in the alignment is in parentheses. *The breakpoint position is undetermined. (b) Potential recombination hotspots within the alignment of mcyA 2 genes. The black line corresponds to the breakpoint count within a moving 200-nt-long window. White and grey areas respectively indicate local 99% and 95% confidence intervals of the hot/cold spot test. Upper and lower broken lines respectively indicate 99% and 95% confidence thresholds for the global hotspot test.

No evidence for recombination was detected in the mcyA 2 gene from Microcystis aeruginosa NPLJ-4 or Microcystis sp. RST 9501. However, the McyA2 adenylation domain of Microcystis spp. NPLJ-4 and RST 9501 differed by 17 amino acids residues compared to other McyA2 adenylation domains of Microcystis strains that produce microcystins containing D-Ala (Additional file 1: Figure S5). Five of the 17 amino acids residues were predicted to be located within 8 Å of the substrate, and three of them were considered to be within the binding pocket. Mutants based on Microcystis sp. RST 9501 McyA2 adenylation domain were constructed to analyze the effect of these three amino acids residues on binding of the predicted substrate amino acids. Surprisingly, the results obtained in the ATP-pyrophosphate exchange assay showed the highest specificity for L-Val in all the mutants and the wild type (Additional file 1: Figure S6). Furthermore, the results indicated a higher activation of L-Ile instead of the expected L-Leu and L-Ala. The high levels of miscognate activation indicate that the catalytic efficiency of the enzymes in recognizing nonpolar amino acid was generally higher, although some specificity was observed for L-Tyr (Additional file 1: Figure S6).

Discussion

Identification of the rare [D-Leu1]MC and other microcystin variants

Microcystins form a large family of cyclic toxins characterized by a highly conserved chemical structure with an extensive amino acid composition at the two variable positions, two and four (Additional file 1: Figure S7 and Table S3). In this study, we detected rare [D-Leu1]MC variants from strains of the distantly related genera Microcystis, Nostoc, and Phormidium. Almost all microcystins reported in the literature contain D-Ala in position 1 [8]. Previously, microcystins containing D-Leu [2731], Gly [35], and D-Ser [36, 37] have also been reported. The [D-Leu1]MC variant has previously been found from Microcystis aeruginosa NPLJ-4, Microcystis sp. RST 9501, and water blooms from Brazil and Canada dominated by Microcystis strains [2731, 34]. The production of microcystins in lichen thalli by Nostoc symbionts has previously been reported [37, 38]. Nostoc strains isolated from lichen symbiosis produce a large variety of microcystins including the [D-Leu1]MC variant [31].

Microcystins are best known from aquatic habitats, where they are frequently reported from blooms. Although microcystins are more commonly detected in planktonic strains, terrestrial and benthic strains have also been reported to be producers of these compounds [8, 39]. Benthic environmental samples containing microcystins have been reported from Switzerland [40], Spain [41], Australia [42], and Antarctica [43]. Microcystin production in isolated cyanobacterial strains from benthic environments has been reported from Egypt [44], New Zealand [45], and the USA [46]. It is not always clear which cyanobacterium produces the toxin in benthic mats of cyanobacteria. A strain of the genus Phormidium was isolated from the walls of a reservoir in the USA and shown to produce a range of microcystin variants, all of which contained D-Ala (Additional file 1: Table S3) [46]. Our results demonstrated that Phormidium sp. CENA 270, isolated from a pond in the northeast of Brazil, also produces microcystins but with D-Leu in place of D-Ala. In the phylogenetic analysis of the 16S rRNA gene, the two Phormidium strains cluster together with Lyngbya, Oscillatoria, Phormidium and Oscillatoriales (Additional file 1: Figure S1). The biomass of benthic strains can go unnoticed to the casual observer but be massive enough to cause animal poisonings [40, 45]. Moreover, when the strains lyse, microcystins are released into the water, which suggests that the analysis of toxic benthic cyanobacteria is also important in water-quality management.

The chemical structure of microcystins is highly conserved, with variation at X and Z positions (Figure 1a and b) resulting in over 86 reported variants [39]. In this study, we demonstrated that Microcystis strains NPLJ-4 and RST 9501 produce new microcystin variants containing methionine in addition to the rare [D-Leu1]MC variants. Met is also present in oscillamide B [47] and in microcystin-M(O)R and -YM [48]. Nodularia spumigena strains also produce nodulapeptins, which commonly contain Met [49, 50]. According to analysis using the NRPS Norine database, Met contains a methylthiol group and is rare in non-ribosomal peptides [51]. The presence of the highly active sulfhydryl group in the thiol group could explain the scarcity of secondary metabolites containing Met or Cys. If amino acid recognition by the McyA2 adenylation domain is not strict, the incorporation of Met instead of Leu is logical because of the similar size and hydrophobicity of the side chains. Here, we demonstrated that methionine is incorporated in the microcystins produced by Microcystis strains from brackish water. Microcystin variants are constantly being discovered, making the microcystin family extremely diverse, and posing a challenge for the detection of microcystins from water samples.

Convergence on [D-Leu1]MC variant chemical structure

Phylogenetic analysis of the McyA2 adenylation domains provided evidence for independent evolutionary events affecting the substrate specificity of the enzyme in three disparate genera of cyanobacteria (Figures 5 and 6). Breakpoint analysis suggests the replacement of almost the entire substrate specificity-conferring portion of the adenylation domain in Phormidium sp. CENA270 and Nostoc sp. UK89IIa. These gene conversions dramatically altered the predicted substrate specificity of the McyA2 adenylation domain in these strains and are linked to the synthesis of the [D-Leu1]MC variant. However, point mutations affecting the substrate specificity of the McyA2 adenylation domain in Microcystis strains NPLJ-4 and RST 9501 led to the synthesis of the [D-Leu1]MC and [Met1]MC variants.

Phormidium sp. CENA270 and Nostoc sp. UK89IIa are not grouped together with other McyA2 adenylation domains (Figure 6). The McyA2 adenylation domain of Phormidium sp. CENA270 grouped with the McyB1 adenylation domain from Microcystis strains, which produce microcystin variants containing D-Ala [10, 52]. The adenylation domains of NosA1 and NosC1 from the nostopeptolide gene cluster are placed in the same clade with the McyA2 adenylation domain from Nostoc sp. UK89IIa (Figure 6). They are involved in the incorporation of Ile/Leu/Val and Leu, respectively, in Nostoc sp. GSV224 [53].

Genetic variation in the microcystin synthetases can be visualized in the phylogenetic trees showing two different patterns. While the amino acids of condensation, peptidyl carrier protein, and epimerization domain regions can be grouped according to the enzyme sequence (McyA, McyB, McyC, McyE, and McyG grouped together), the adenylation domain phylogeny clearly indicates recombination (Figures 4 and 5). Recombination in adenylation domains has previously been described for the adenylation and condensation domains of McyB1 and McyC [16]. The recombination and positive selection in the McyB1 adenylation domain are involved in the high variability of amino acids incorporated at position 2 of the microcystin [16, 18, 22, 24, 25]. These genetic events have been related to the increase in the chemical diversity of microcystin. Interestingly, our results show that these different evolutionary events are involved in the convergence of the [D-Leu1]MC-LR.

Nevertheless, the selective forces behind this convergent evolution remain unclear. Competition in brackish water and different seasonal periods have possibly acted as selective forces. The chemical diversity of microcystins could be related to protein phosphatase inhibition as a form of chemical defense, for example against predators. Previous studies have indicated that microcystins can affect some predators, acting as metal chelators, in gene regulation, or in the inter- and intra-specific signaling [5459]. However, microcystins join a large number of secondary metabolites produced by different organisms that have no assigned biological function. According to the most accepted view, these compounds are produced due their ecological or physiological function and benefits for the producer organisms [60]. However, more information is still needed concerning the advantages in the production of these secondary metabolites. The biological role of a mixture of different bioactive compounds produced by the same strain would be interesting to study.

Prediction of McyA2adenylation domain substrate specificity

The eight to ten amino acid residues forming the adenylation domain binding pocket are the main determinants of substrate specificity [7, 61, 62]. In our study, Phormidium sp. CENA270 and Nostoc sp. UK89IIa were shown to produce [D-Leu1]MC variants and have identical binding pocket sequences (Table 2). Such amino acids signatures had been already described as presenting Leu specificity [7]. The Microcystis strains NPLJ-4 and RST 9501 differ in the binding pocket positions 301, 330, and 331 from the strains producing [D-Ala1]MCs. Residues at positions 301 and 330 are regarded to be less variable than at position 331 [7].

The adenylation domain binding pocket of Microcystis strains NPLJ-4 and RST 9501 has three different amino acid residues and a broader diversity of microcystin variants at position 1. Despite the fact that almost the entire binding pocket of Phormidium sp. CENA270 and Nostoc sp. UK89IIa differs from the other studied strains, only microcystin variants containing Leu at position 1 were detected. Re-engineering of non-ribosomal peptides has been a challenge in order to synthesize new peptides or to increase the activity of known compounds. The engineering of NRPSs to change substrate specificity can in some cases be achieved by point mutations. However, our results suggest that the replacement of entire domains might be a more successful strategy for producing a single product.

Replacement of almost the entire McyA2 adenylation domain in Nostoc sp. UK89IIa and Phormidium sp. CENA270 resulted in specificity towards Leu. Neither strain produced detectable levels of microcystin variants that contain other amino acids at this position. The recombination detected in the mcyA 2 gene of these strains affects the substrate-conferring portion of the McyA2 adenylation domain, which is important for the selection and activation of amino acids [7, 61, 62]. Previously, it has been reported that recombination among different adenylation domains from mcyB1 and mcyC genes has led to a change in amino acid activation [24].

We designed an experiment in order to test whether point mutations at positions 301, 330, and 331 could change the substrate specificity of the adenylation domain. However, single amino acids changes did not have the expected results. All the constructs and the wild type were found to activate valine in ATP-pyrophosphate (PPi) exchange assays. A previous study [7] demonstrated that in the case of single or multiple mutations, the specificity of the wild type is not lost, but there is an increase in new substrate specificity. A comparison of adenylation domains from Microcystis strains that activate Ala and Leu reveals that several amino acid residues differ between them (Additional file 1: Figure S5). Of these different amino acid residues, five are 8 Å or less distant from the substrate and only three belong to the binding pocket. Although it is predicted that amino acid residues in the binding pocket are involved in selectivity, the catalytic efficiency could also be affected by the tertiary structure and proteinogenic surrounding area of the adenylation domain [63]. Promiscuity of the enzymes, allowing them to activate different substrates, could also be involved in the high variability of microcystin variants. Promiscuous activation of amino acids with a hydrophobic side chain by TycA, involved in the synthesis of the antibiotic tyrocidine A, has been reported [64]. Moreover, adenylation domains activating multiple substrates have been described from the fengycin [65], lychenysin [66], nostopeptolide [53], and cyanopeptolin [67] biosynthetic pathways.

Conclusion

Our study revealed that independent gene conversion events and point mutations led to the production of the same microcystin variant by strains belonging to three different cyanobacterial orders. The large chemical diversity of microcystins is proposed to be mostly the result of genetic rearrangements, positive selection, and recombination acting to increase structural diversity. Furthermore, the replacement of the entire adenylation domain seems to result in a more specific change in non-ribosomal peptides than point mutation. New variants of [Met1]MCs were detected in Microcystis strains NPLJ-4 and RST 9501. Our study also revealed a new cyanobacterial strain (Phormidium sp CENA270) producing a rare variant of the potent hepatotoxic microcystin. This finding expands on the recent increase in the detection of microcystin-producing terrestrial and benthic cyanobacterial strains.

Methods

Studied strains and culture conditions

Phormidium sp. CENA270 was isolated from a pond during the rainy season in Paulista (Paraiba, Brazil) and maintained in BG-11 medium [68], but supplemented with half the described amount of combined nitrogen. Nostoc sp. UK89IIa was isolated from the lichen Peltigera neopolydactyla sampled in Laukaa (Finland) and maintained in Z8 medium [69] lacking a source of combined nitrogen. Microcystis strains NPLJ-4 and RST 9501 were respectively isolated from the brackish water of Jacarepaguá Lagoon, Rio de Janeiro [34], and the Patos Lagoon estuary, Rio Grande do Sul [27], in Brazil and maintained in Z8 medium [69] with and without a source of combined nitrogen. The biomass for chemical and molecular analysis was obtained by growing the strains in 2 x 3 L of culture medium under constant light of 3–8 μmol m-2 s-1 and a temperature of 24 ± 1°C. Microcystis sp. RST 9501 was grown in Z8 medium replaced with stable isotope 34S-labeled MgSO4.7H2O (catalogue no. IS7080; 90 atom % 34S; Icon) for the detection and identification of sulfur-containing microcystins.

Sequencing and phylogenetic analysis

Freeze-dried cells (6–50 mg) were used for DNA extraction with a DNeasy Plant mini kit (Qiagen). Two different sizes of glass beads (180 μm and 425–600 μm, Sigma-Aldrich) were added and the cells were disrupted by shaking at 6.5 m s-1 for 60 seconds in 3 batches using a FastPrep (M.P. Biomedicals). An additional 1 h incubation at 100°C was necessary in order to extract Phormidium sp. CENA270 genomic DNA due the thick sheath surrounding the trichomes of this strain. PCR reaction conditions are described in Additional file 3. The PCR products were cloned in pCR®2.1-TOPO (TOPO TA Cloning, Invitrogen) with the following modifications from the manufacturer’s instructions: the entire cloning reaction was used for transformation in 25 μL of TOP10 competent cells, 100 μL of SOC medium was added to the mixture after heat shock and incubated for 20 min in shaker (160 RPM) at 37°C, and the entire reaction volume was plated in LB plates containing 150 μg mL-1 of ampicillin.

Plasmid extraction was performed using the QIAprep Spin Miniprep kit (Qiagen) and cycle sequencing was carried out using an ABI Prism® BigDye® Terminator v3.1 cycle sequencing kit. The oligonucleotide primers used to sequence PCR products are given in Additional file 1: Table S4. Sanger sequencing was performed in an ABI PRISM 310 Genetic Analyzer. Contigs were aligned in the program Phred/Phrap/Consed (Philip Green, University of Washington, Seattle, USA), accepting bases with quality >20. BLASTn was used to search the nr database at NCBI for related strains to be included in phylogenetic analysis. GenBank accession numbers are indicated in Additional file 1: Table S5. Maximum-likelihood trees were constructed in MEGA 5.0 [70]. The best substitution model for each sequence was chosen based on analysis in MEGA 5.0 (K2 + G for the small tree of 16S rRNA; GTR + G + I for mcyD concatenated with mcyE; K2 + G + I for long tree of 16S rRNA; JTT + G + I for condensation, adenylation, and epimerization domains sequences of McyA; and JTT + G for peptidyl carrier protein domain sequences of McyA). Neighbor-joining and maximum parsimony trees were constructed using the respective methods: JTT + G and CNI on random trees. The domains present in the McyA amino acid sequences were detected in the program PKS/NRPS Analysis [71]. Adenylation domain substrate specificity prediction was performed using the program NRPSpredictor2 [72, 73].

Recombination test

The nucleotide sequences of 13 cyanobacterial strains (Additional file 1: Table S5) corresponding to second adenylation domain of mcyA (mcyA 2 ) were aligned in ClustalW (MEGA 5.0). The recombination analyses were performed in the program RDP3 [74, 75], which implements different methods to detect recombination. The methods used in this study were: original RDP [76], BOOTSCAN [77], GENECONV [78], MAXCHI [79], CHIMAERA [80], SISCAN [81] and 3Seq [82]. Default parameters were used and a P-value cutoff was chosen as 0.05. Nucleotide sequence comparison was performed in BLASTn (NCBI database). A breakpoint distribution was plotted, allowing the visualization of recombination hotspots [74]. The phylogenetic compatibility matrix was calculated by TreeOrder scan in the program SSE [83].

McyA2adenylation domain mutation and heterologous expression

A fragment containing the mcyA 2 adenylation domain was obtained from Microcystis sp. RST 9501 by PCR using the primers RSTPETF and RSTPETR (Additional file 1: Table S6). The PCR reaction conditions are described in Additional file 3. The 1,646 bp DNA fragment was cloned in pET101/D-TOPO (Invitrogen) and transformed in TOP10 chemically competent Escherichia coli cells. Mutants were obtained by site-directed mutation with PCR mutagenesis using specific primers (Additional file 1: Table S6). Detailed information on mutagenesis is presented in Additional file 3. Mutant G301A contains an Ala instead of Gly at position 301, mutant I330T a Thr instead of Ile at position 330, mutant C331Y a Tyr instead of Cys at position 331, and mutant G301A, I330T, C331Y has all three mutations. The presence of the desired mutation was verified by sequencing. Mutated and wild type adenylation domains were cloned into pFN18A (HaloTag® 7) T7 Flexi® vector (Promega, WI, USA) and transformed in Escherichia coli KRX competent cells (Promega) following the manufacturer’s instructions. The adenylation domain was heterologously expressed and purified using the HaloTag® Protein Purification System (Promega). The ATP-pyrophosphate exchange assay was performed as previously described [49].

Chemical analysis

Identification of microcystin variants was performed using an Agilent 1100 Series LC/MSD Trap XCT Plus high-performance liquid chromatograph mass spectrometer (Agilent Technologies). Freeze-dried biomass was extracted with 1 mL of methanol in a FastPrep homogenizer (M.P. Biomedicals). The supernatant was injected to a Luna C18(2) column (150 × 2.1 mm, 5 μm, Phenomenex) for the LC-MS analysis. Electrospray ionization in positive mode was used and the product ion spectra of protonated microcystins were analyzed to identify the structure of the variant.

The remains of the methanol extracts were mixed with water and dichloromethane in equivalent proportions (1:1:1). The hydrophilic upper phase was diluted with water and passed through a preconditioned SPE column (strata 8B-S100-UBJ, Phenomenex). Microcystin-containing fractions were recovered in 1 mL of methanol. Microcystins were isolated and purified by HPLC (HP 1100 Series modular chromatograph, Agilent Technologies). The microcystin fractions were hydrolyzed with 200 μL of 6 M deuterium chloride (catalog no. 543047; 35 wt % solution in D2O, 99 atom % D, Sigma-Aldrich) and prepared for enantiomeric amino acid analysis using FDAA (Pierce) as a Marfey reagent, as described previously [84]. Chiral analysis of the amino acids alanine, leucine, arginine, and homoarginine was carried out using LC-MS. Amino acids from the common [D-Ala1] MC-LR and -RR variants produced by Anabaena sp. 90 were analyzed as references. Detailed information on the chemical analysis is provided in Additional file 3.

Abbreviations

NRPS: 

Non-ribosomal peptide synthetase

PKS: 

Polyketide synthase

NRPS-PKS: 

Hybrid non-ribosomal peptide synthetase and polyketide synthase

A: 

Adenylation domain

C: 

Condensation domain

PCP: 

Peptidyl carrier protein domains

E: 

Epimerization domain

MC: 

Microcystin.

Declarations

Acknowledgments

This work was supported by Academy of Finland to KS (118637). TKS was funded by the Helsinki Graduate Program in Biotechnology and Molecular Biology, São Paulo Research Foundation (2009/13455-0), Centre for International Mobility (TM-09-6506) and Finnish Cultural Foundation. UK was funded by the Emil Aaltonen Foundation.

Authors’ Affiliations

(1)
Department of Food and Environmental Sciences, Division of Microbiology and Biotechnology, University of Helsinki, Viikki Biocenter (Viikinkaari 9)
(2)
Department of Biosciences, University of Helsinki, Viikki Biocenter (Viikinkaari 1)
(3)
Center for Nuclear Energy in Agriculture, University of São Paulo
(4)
Unidade de Pesquisas em Cianobactérias, Federal University of Rio Grande

References

  1. Walsh CT, Chen H, Keating TA, Hubbard BK, Losey HC, Luo L, Marshall CG, Miller DA, Patel HM: Tailoring enzymes that modify nonribosomal peptides during and after chain elongation on NRPS assembly lines. Curr Opin Chem Biol. 2001, 5: 525-534.PubMedView Article
  2. Finking R, Marahiel MA: Biosynthesis of nonribosomal peptides. Annu Rev Microbiol. 2004, 58: 453-488.PubMedView Article
  3. Sieber SA, Marahiel MA: Molecular mechanisms underlying nonribosomal peptide synthesis: approaches to new antibiotics. Chem Rev. 2005, 105: 715-738.PubMedView Article
  4. Kallow W, Neuhof T, Arezi B, Jungblut P, Von Döhren H: Penicillin biosynthesis: intermediates of biosynthesis of delta-alpha-aminoadipyl-cysteinyl-valine formed by ACV synthetase from Acremonium chrysogenum. FEBS Lett. 1997, 414: 74-78.PubMedView Article
  5. Steenbergen J, Alder J, Thorne G, Tally F: Daptomycin: a lipopeptide antibiotic for the treatment of serious gram-positive infections. J Antimicrob Chemother. 2005, 55: 283-288.PubMedView Article
  6. Shen B, Du L, Sanchez C, Edwards D, Chen M, Murrell J: The biosynthetic gene cluster for the anticancer drug bleomycin from Streptomyces verticillus ATCC 15003 as a model for hybrid peptide-polyketide natural product biosynthesis. J Ind Microbiol Biotechnol. 2001, 27: 378-385.PubMedView Article
  7. Stachelhaus T, Mootz HD, Marahiel MA: The specificity-conferring code of adenylation domains in nonribosomal peptide synthetases. Chem Biol. 1999, 6: 493-505.PubMedView Article
  8. Sivonen K, Jones G: Cyanobacterial toxins. Toxic cyanobacteria in water. Edited by: Chorus I, Bartram J. 1999, London: E & FN Spon, 41-111.
  9. Kuiper-Goodman T, Falconer IR, Fitzgerald DJ: Human health aspects. Toxic Cyanobacteria in Water: a Guide to their Public Health Consequences, Monitoring and Management. Edited by: Chorus I, Bartram J. 1999, London: E and FN Spon, 113-153.
  10. Tillett D, Dittmann E, Erhard M, Von Döhren H, Börner T, Neilan BA: Structural organization of microcystin biosynthesis in Microcystis aeruginosa PCC7806: an integrated peptide polyketide synthetase system. Chem Biol. 2000, 7: 753-764.PubMedView Article
  11. Christiansen G, Fastner J, Erhard M, Börner T, Dittmann E: Microcystin biosynthesis in Planktothrix: Genes, Evolution, and Manipulation. J Bacteriol. 2003, 185: 564-572.PubMed CentralPubMedView Article
  12. Rouhiainen L, Vakkilainen T, Siemer BL, Buikema W, Haselkorn R, Sivonen K: Genes coding for hepatotoxic heptapeptides (microcystins) in the cyanobacterium Anabaena strain 90. Appl Environ Microbiol. 2004, 70: 686-692.PubMed CentralPubMedView Article
  13. Dittmann E, Börner T: Genetic contributions to the risk assessment of microcystin in the environment. Toxicol Appl Pharmacol. 2005, 203: 192-200.PubMedView Article
  14. Rantala A, Fewer DP, Hisbergues M, Rouhiainen L, Vaitomaa J, Borner T, Sivonen K: Phylogenetic evidence for the early evolution of microcystin synthesis. Proc Natl Acad Sci USA. 2004, 101: 568-573.PubMed CentralPubMedView Article
  15. Jungblut AD, Neilan BA: Molecular identification and evolution of the cyclic peptide hepatotoxins, microcystin and nodularin, synthetase genes in three orders of cyanobacteria. Arch Microbiol. 2006, 185: 107-114.PubMedView Article
  16. Fewer DP, Rouhiainen L, Jokela J, Wahlsten M, Laakso K, Wang H, Sivonen K: Recurrent adenylation domain replacement in the microcystin synthetase gene cluster. BMC Evol Biol. 2007, 7: 183.PubMed CentralPubMedView Article
  17. Christiansen G, Molitor C, Philmus B, Kurmayer R: Nontoxic strains of cyanobacteria are the result of major gene deletion events induced by a transposable element. Mol Biol Evol. 2008, 25: 1695-1704.PubMed CentralPubMedView Article
  18. Mikalsen B, Boison G, Skulberg OM, Fastner J, Davies W, Gabrielsen TM, Rudi K, Jacobsen KS: Natural variation in the microcystin synthetase operon mcyABC and impact on microcystin production in Microcystis strains. J Bacteriol. 2003, 185: 2774-2785.PubMed CentralPubMedView Article
  19. Tanabe Y, Kaya KT, Watanabe MM: Evidence for recombination in the microcystin synthetase (mcy) genes of toxic cyanobacteria Microcystis spp. J Mol Evol. 2004, 58: 633-641.PubMedView Article
  20. Tanabe Y, Sano T, Kasai F, Watanabe MM: Recombination, cryptic clades and neutral molecular divergence of the microcystin synthetase (mcy) genes of toxic cyanobacterium Microcystis aeruginosa. BMC Evol Biol. 2009, 9: 115.PubMed CentralPubMedView Article
  21. Kurmayer R, Christiansen G, Gumpenberger M, Fastner J: Genetic identification of microcystin ecotypes in toxic cyanobacteria of the genus Planktothrix. Microbiology. 2005, 151: 1525-1533.PubMedView Article
  22. Kurmayer R, Gumpenberger M: Diversity of microcystin genotypes among populations of the filamentous cyanobacteria Planktothrix rubescens and Planktothrix agardhii. Mol Ecol. 2006, 15: 3849-3861.PubMed CentralPubMedView Article
  23. Tooming-Klunderud A, Mikalsen B, Kristensen T, Jakobsen KS: The mosaic structure of the mcyABC operon in Microcystis. Microbiology. 2008, 154: 1886-1899.PubMedView Article
  24. Tooming-Klunderud A, Fewer DP, Rohrlack T, Jokela J, Rouhiainen L, Sivonen K, Kristensen T, Jakobsen KS: Evidence for positive selection acting on microcystin synthetase adenylation domains in three cyanobacterial genera. BMC Evol Biol. 2008, 8: 256.PubMed CentralPubMedView Article
  25. Christiansen G, Yoshida WY, Blom JF, Portmann C, Gademann K, Hemscheidt T, Kurmayer R: Isolation and structure determination of two microcystins and sequence comparison of the McyABC adenylation domains in Planktothrix species. J Nat Prod. 2008, 71: 1881-1886.PubMed CentralPubMedView Article
  26. Fewer DP, Tooming-Klunderud A, Jokela J, Wahlsten M, Rouhiainen L, Kristensen T, Rohrlack T, Jakobsen KS, Sivonen K: Natural occurrence of microcystin synthetase deletion mutants capable of producing microcystins in strains of the genus Anabaena (Cyanobacteria). Microbiology. 2008, 154: 1007-1014.PubMedView Article
  27. Matthiensen A, Beattie KA, Yunes JS, Kaya K, Codd GA: [D-Leu1]Microcystin-LR, from the cyanobacterium Microcystis RST 9501 and from a Microcystis bloom in the Patos Lagoon estuary, Brazil. Phytochemistry. 2000, 55: 383-387.PubMedView Article
  28. Park H, Namikoshi M, Brittain SM, Carmichael WW, Murphy T: [D-Leu(1)] microcystin-LR, a new microcystin isolated from waterbloom in a Canadian prairie lake. Toxicon. 2001, 39: 855-862.PubMedView Article
  29. Schripsema J, Dagnino D: Spectral assignments and reference data complete assignment of the NMR spectra of [D-Leu1]-microcystin-LR and analysis of its solution structure. Magn Reson Chem. 2002, 40: 614-617.View Article
  30. Albuquerque EC, Melo LFC, Franco TT: Use of solid-phase extraction, high-performance liquid chromatography, and MALDI-TOF identification for [D-Leu1] MCYST-LR analysis in treated water: validation of the analytical methodology. Can J Anal Sci Spectrosc. 2007, 52: 1-10.
  31. Kaasalainen U, Fewer DP, Jokela J, Wahlsten M, Sivonen K, Rikkinen J: Cyanobacteria produce a high variety of hepatotoxic peptides in symbiosis. Proc Natl Acad Sci USA. 2012, 109: 5886-5891.PubMed CentralPubMedView Article
  32. Amoutzias GD, Van de Peer Y, Mossialos D: Evolution and taxonomic distribution of nonribosomal peptide and polyketide synthases. Future Microbiol. 2008, 3: 361-370.PubMedView Article
  33. Dittmann E, Fewer DP, Neilan BA: Cyanobacterial toxins: biosynthetic routes and evolutionary roots. FEMS Microbiol Rev. 2013, 37: 23-43.PubMedView Article
  34. Soares RM, Magalhaes VF, Azevedo SMFO: Accumulation and depuration of microcystins (cyanobacteria hepatotoxins) in Tilapia rendalli (Cichlidae) under laboratory conditions. Aquat Toxicol. 2004, 70: 1-10.PubMedView Article
  35. Wood SA, Mountfort D, Selwood AI, Holland PT, Puddick J, Cary SC: Widespread distribution and identification of eight novel microcystins in Antarctic cyanobacterial mats. Appl Environ Microbiol. 2008, 74: 7243-7251.PubMed CentralPubMedView Article
  36. Sivonen K, Namikoshi M, Evans WR, Fardig M, Carmichael WW, Rinehart KL: Three new microcystins, cyclic heptapeptide hepatotoxins, from Nostoc sp. strain 152. Chem Res Toxicol. 1992, 5: 464-469.PubMedView Article
  37. Oksanen I, Jokela J, Fewer DP, Wahlsten M, Rikkinen J, Sivonen K: Discovery of rare and highly toxic microcystins from lichen-associated cyanobacterium Nostoc sp. strain IO-102-I. Appl Environ Microbiol. 2004, 70: 5756-5763.PubMed CentralPubMedView Article
  38. Kaasalainen U, Jokela J, Fewer DP, Sivonen K, Rikkinen J: Microcystin production in the tripartite cyanolichen Peltigera leucophlebia. Mol Plant Microbe Interact. 2009, 22: 695-702.PubMedView Article
  39. Sivonen K: Cyanobacterial Toxins. Encyclopedia of Microbiology. Edited by: Moselio S. 2009, Oxford: Elsevier, 290-307.View Article
  40. Mez K, Beattie K, Codd G, Hanselmann K, Hauser B, Naegeli H, Preisig H: Identification of a microcystin in benthic cyanobacteria linked to cattle deaths on alpine pastures in Switzerland. Eur J Phycol. 1997, 32: 111-117.View Article
  41. Aboal M, Puig MA: Intracellular and dissolved microcystin in reservoirs of the river Segura basin, Murcia, SE Spain. Toxicon. 2005, 45: 509-518.PubMedView Article
  42. Dasey M, Ryan N, Wilson J, McGregor G, Fabbro L, Neilan BA, Burns BP, Kankaanpää H, Morrison LF, Codd GA, Rissik D, Bowling L: Investigations into the taxonomy, toxicity and ecology of benthic cyanobacterial accumulations in Myall Lake, Australia. Mar Freshw Res. 2005, 56: 45-55.View Article
  43. Hitzfeld BC, Lampert CS, Spaeth N, Mountfort D, Kaspar H, Dietrich DR: Toxin production in cyanobacterial mats from ponds on the McMurdo Ice Shelf, Antarctica. Toxicon. 2000, 38: 1731-1748.PubMedView Article
  44. Mohamed ZA, El-Sharouny HM, Ali WSM: Microcystin production in benthic mats of cyanobacteria in the Nile River and irrigation canals. Egypt. Toxicon. 2006, 47: 584-590.View Article
  45. Wood SA, Heath MW, Holland PT, Munday R, McGregor GB, Ryan KG: Identification of a benthic microcystin-producing filamentous cyanobacterium (Oscillatoriales) associated with a dog poisoning in New Zealand. Toxicon. 2010, 55: 897-903.PubMedView Article
  46. Izaguirre G, Jungblut AD, Neilan BA: Benthic cyanobacteria (Oscillatoriaceae) that produce microcystin-LR, isolated from four reservoirs in southern California. Water Res. 2007, 41: 492-498.PubMedView Article
  47. Sano T, Usui K, Ueda H, Osada H, Kaya K: Isolation of new protein phosphatase inhibitors from two cyanobacteria species. Planktothrix spp. J Nat Prod. 2001, 64: 1052-1055.View Article
  48. Namikoshi M, Rinehart KL, Sakai R, Stotts RR, Dahlem AM, Beasley VR, Carmichael WW, Evans WR: Identification of 12 hepatotoxins from a Homer Lake bloom of the cyanobacteriaMicrocystis aeruginosa, Microcystis viridis, and Microcystis wesenbergii: nine new microcystins. J Org Chem. 1992, 57: 866-872.View Article
  49. Rouhiainen L, Jokela J, Fewer DP, Urmann M, Sivonen K: Two alternative starter modules for the non-ribosomal biosynthesis of specific anabaenopeptin variants in Anabaena (Cyanobacteria). Chem Biol. 2010, 17: 265-273.PubMedView Article
  50. Fujii K, Sivonen K, Adachi K, Noguchi K, Sano H, Hirayama K, Suzuki M, Harada K-I: Comparative study of toxic and non-toxic cyanobacterial products: novel peptides from toxic Nodularia spumigena AV1. Tetrahedron Lett. 1997, 38: 5525-5528.View Article
  51. Caboche S, Leclere V, Pupin M, Kucherov G, Jacques P: Diversity of monomers in nonribosomal peptides: towards the prediction of origin and biological activity. J Bacteriol. 2010, 192: 5143-5150.PubMed CentralPubMedView Article
  52. Ohtake A, Shirai M, Aida T, Mori N, Harada K-I, Matsuura K, Suzuki M, Nakano M: Toxicity of Microcystis species isolated from natural blooms and purification of the toxin. Appl Environ Microbiol. 1989, 55: 3202-3207.PubMed CentralPubMed
  53. Hoffmann D, Hevel JM, Moore RE, Moore BS: Sequence analysis and biochemical characterization of the nostopeptolide A biosynthetic gene cluster from Nostoc sp. GSV224. Gene. 2003, 311: 171-180.PubMedView Article
  54. Utkilen H, Gjølme N: Iron-stimulated toxin production in Microcystis aeruginosa. Appl Environ Microbiol. 1995, 61: 797-800.PubMed CentralPubMed
  55. Rohrlack T, Dittmann E, Henning M, Borner T, Kohl JG: Role of microcystins in poisoning and food ingestion inhibition of Daphnia galeata caused by the cyanobacterium Microcystis aeruginosa. Appl Environ Microbiol. 1999, 65: 737-739.PubMed CentralPubMed
  56. Rohrlack T, Dittmann E, Börner T, Christoffersen K: Effects of cell-bound microcystins on survival and feeding of Daphnia spp. Appl Environ Microbiol. 2001, 67: 3523-3529.PubMed CentralPubMedView Article
  57. Dittmann E, Erhard M, Kaebernick M, Scheler C, Neilan BA, Von Döhren H, Börner T: Altered expression of two light-dependent genes in a microcystin-lacking mutant of Microcystis aeruginosa PCC 7806. Microbiology. 2001, 147: 3113-3119.PubMedView Article
  58. Babica P, Bláha L, Maršálek B: Exploring the natural role of microcystins – a review of effects on photoautotrophic organisms. J Phycol. 2006, 42: 9-20.View Article
  59. Schatz D, Keren Y, Vardi A, Sukenik A, Carmeli S, Börner T, Dittmann E, Kaplan A: Towards clarification of the biological role of microcystins, a family of cyanobacterial toxins. Environ Microbiol. 2007, 4: 965-970.View Article
  60. Jenke-Kodama H, Dittmann E: Evolution of metabolic diversity: insights from microbial polyketide synthases. Phytochemistry. 2009, 70: 1858-1866.PubMedView Article
  61. Challis GL, Ravel J, Townsend CA: Predictive, structure-based model of amino acid recognition by nonribosomal peptide synthetase adenylation domains. Chem Biol. 2000, 7: 211-224.PubMedView Article
  62. Conti E, Stachelhaus T, Marahiel MA, Brick P: Structural basis for the activation of phenylalanine in the non-ribosomal biosynthesis of gramicidin S. EMBO J. 1997, 16: 4174-4183.PubMed CentralPubMedView Article
  63. Eppelmann K, Stachelhaus T, Marahiel MA: Exploitation of the selectivity-conferring code of nonribosomal peptide synthetases for the rational design of novel peptide antibiotics. Biochemistry. 2002, 41: 9718-9726.PubMedView Article
  64. Villiers BRM, Hollfelder F: Mapping the limits of substrate specificity of the adenylation domain of TycA. Chembiochem. 2009, 10: 671-682.PubMedView Article
  65. Lin TP, Chen CL, Chang LK, Tschen JSM, Liu ST: Functional and transcriptional analyses of a fengycin synthetase gene, fen C, from Bacillus subtilis. J Bacteriol. 1999, 181: 5060-5067.PubMed CentralPubMed
  66. Konz D, Doekel S, Marahiel MA: Molecular and biochemical characterization of the protein template controlling biosynthesis of the lipopeptide lichenysin. J Bacteriol. 1999, 181: 133-140.PubMed CentralPubMed
  67. Rounge TB, Rohrlack T, Kristensen T, Jakobsen KS: Recombination and selectional forces in cyanopeptolin NRPS operons from highly similar, but geographically remote Planktothrix strains. BMC Microbiol. 2008, 8: 141.PubMed CentralPubMedView Article
  68. Allen MB: Simple conditions for growth of unicellular blue-green algae on plates. J Phycol. 1968, 4: 1-4.View Article
  69. Kotai J: Instructions for Preparation of Modified Nutrient Solution Z8 for Algae. Norwegian Institute for Water Research, publication B-11/69. 1972, Oslo: Blindern
  70. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S: MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol. 2011, 28: 2731-2739.PubMed CentralPubMedView Article
  71. Bachmann BO, Ravel J: In silico prediction of microbial secondary metabolic pathways from DNA sequence data. Methods Enzymol. 2009, 458: 181-217.PubMedView Article
  72. Rausch C, Weber T, Kohlbacher O, Wohlleben W, Hudson DH: Specificity prediction of adenylation domains in nonribosomal peptide synthetases (NRPS) using transductive support vector machines (TSVMs). Nucleic Acids Res. 2005, 33: 5799-5808.PubMed CentralPubMedView Article
  73. Röttig M, Medema MH, Bin K, Weber T, Rausch C, Kohbacher O: NRPSpredictor2—a web server for predicting NRPS adenylation domain specificity. Nucleic Acids Res. 2011, 39: W362-W367.PubMed CentralPubMedView Article
  74. Heath L, van der Walt E, Varsani A, Martin DP: Recombination patterns in aphthoviruses mirror those found in other picornaviruses. J Virol. 2006, 80: 11827-11832.PubMed CentralPubMedView Article
  75. Martin DP, Lemey P, Lott M, Moulton V, Posada D, Lefeuvre P: RDP3: a flexible and fast computer program for analyzing recombination. Bioinformatics. 2010, 26: 2462-2463.PubMed CentralPubMedView Article
  76. Martin D, Rybicki E: RDP: detection of recombination amongst aligned sequences. Bioinformatics. 2000, 16: 562-563.PubMedView Article
  77. Martin DP, Posada D, Crandall KA, Williamson C: A modified BOOTSCAN algorithm for automated identification of recombinant sequences and recombination breakpoints. AIDS Res Hum Retroviruses. 2005, 21: 98-102.PubMedView Article
  78. Padidam M, Sawyer S, Fauquet CM: Possible emergence of new geminiviruses by frequent recombination. Virology. 1999, 265: 218-225.PubMedView Article
  79. Maynard Smith J: Analyzing the mosaic structure of genes. J Mol Evol. 1992, 34: 126-129.
  80. Posada D, Crandall KA: Evaluation of methods for detecting recombination from DNA sequences: Computer simulations. Proc Natl Acad Sci USA. 2001, 98: 13757-13762.PubMed CentralPubMedView Article
  81. Gibbs MJ, Armstrong JS, Gibbs AJ: Sister-Scanning: a Monte Carlo procedure for assessing signals in recombinant sequences. Bioinformatics. 2000, 16: 573-582.PubMedView Article
  82. Boni MF, Posada D, Feldman MW: An exact nonparametric method for inferring mosaic structure in sequence triplets. Genetics. 2007, 176: 1035-1047.PubMed CentralPubMedView Article
  83. Simmonds P: SSE: A nucleotide and amino acid sequence analysis platform. BMC Res Notes. 2012, 5: 50.PubMed CentralPubMedView Article
  84. Jokela J, Herfindal L, Wahlsten M, Permi P, Selheim F, Vasconcelos V, Døskeland SO, Sivonen K: A novel cyanobacterial nostocyclopeptide is a potent antitoxin against microcystins. Chembiochem. 2010, 11: 1594-1599.PubMedView Article

Copyright

© Shishido et al.; licensee BioMed Central Ltd. 2013

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.