OUP user menu

On the presence of peptide synthetase and polyketide synthase genes in the cyanobacterial genus Nodularia

Michelle C. Moffitt , Brett A. Neilan
DOI: http://dx.doi.org/10.1111/j.1574-6968.2001.tb10566.x 207-214 First published online: 1 March 2001

Abstract

Nodularin is a hepatotoxin produced by the bloom-forming cyanobacterial species Nodularia spumigena. Putative peptide synthetase and polyketide synthase genes were detected in toxic strains of Nodularia by degenerate PCR. Using specific primer sets, peptide synthetase and polyketide synthase gene homologues were detected in nodularin-producing strains indicating a possible role of peptide synthetase and polyketide synthase enzyme complexes in the biosynthesis of nodularin. Strains of Nodularia isolated from around the world were also analyzed for the production of nodularin by the protein phosphatase 2A inhibition assay. The protein phosphatase inhibition assay and the molecular detection of peptide synthetase and polyketide synthase genes in Nodularia may be useful techniques for the assessment of nodularin-producing cyanobacteria in the environment.

Keywords
  • Nodularin
  • Peptide synthetase
  • Polyketide synthase
  • Cyanobacterium
  • Nodularia

1 Introduction

The cyanobacteria produce a wide range of secondary metabolites, including hepatotoxins produced by a number of bloom-forming genera such as Microcystis, Cylindrospermopsis and Nodularia [1].

Nodularia blooms are common within the estuaries and coastal lagoons of southern Australia [2,3] and occur annually during summer months in the Baltic Sea [4]. Nodularia blooms are highly toxic due to the associated production of the cyclic pentapeptide hepatotoxin nodularin. Nodularin production has been shown to be restricted to the planktic bloom-forming species Nodularia spumigena by phylogenetic analysis [5]. Nodularin is similar in structure to the cyclic heptapeptide microcystin [6]. Microcystin is produced non-ribosomally by a number of cyanobacterial genera including Microcystis, Anabaena, and Oscillatoria [7,8]. Nodularin contains a dehydroamino acid, N-methyldehydrobutyrine (Mdhb), two d-amino acids, d-glutamic (d-Glu) and d-erythro-B-methylaspartic acids (d-MeAsp) and the more common l-arginine (l-Arg) (Fig. 1). Nodularin also contains the fatty acid C20 amino acid, 3-amino-9-methoxy-2,6,8-trimethyl-10-phenyl-4,6-decadienoic acid (Adda). Both nodularin and microcystin share in common the amino acids Adda, d-Glu, d-MeAsp and l-Arg. The Mdhb residue found in nodularin is usually replaced by an N-methyldehydroalanine in microcystin [6].

1

Structure of the pentapeptide toxin nodularin.

A number of secondary metabolites from lower eukaryotes and prokaryotes, including cyanobacteria, are produced non-ribosomally via large modular enzyme complexes [911]. Non-ribosomal peptide synthesis occurs via large modular multifunctional enzyme complexes, responsible for the activation of each amino acid by acyladenylation and thioesterification. Peptide elongation is mediated via a 4′-phosphopantetheine cofactor and the resulting peptide is released via the action of a thioesterase. In some peptides, the amino acids are epimerized to the d-isomer or N-methylated by enzymes within the peptide synthetase complex [9,10]. Polyketide synthase complexes are responsible for the biosynthesis of the fatty acid-like structures called polyketides. Polyketide synthases are grouped into two main types. The non-aromatic type I polyketide synthase enzyme complexes are modular. Each module is responsible for one fatty acid-like elongation step [1214]. The presence of a polyketide synthase-like gene in Nodularia has not been shown previously. In addition, the distribution of a Nodularia specific peptide synthetase has not been examined in toxic and non-toxic strains of this genus.

Due to structural similarity, we propose that a similar mechanism of toxin biosynthesis would be present in both microcystin- and nodularin-producing cyanobacteria. Cyclic structure and conserved d-amino acids indicated the involvement of a peptide synthetase while the presence of the novel amino acid Adda supports similar polyketide synthesis. Using degenerate oligonucleotide primers previously described for the detection of peptide synthetase genes in microcystin-producing species [8], a putative peptide synthetase region was sequenced in toxic Nodularia. Degenerate polyketide synthase oligonucleotide primers were also designed during this study to sequence a putative polyketide synthase region specifically from Nodularia. We report, for the first time, that polyketide synthase-like genes have been detected in Nodularia. Specific primers were designed to determine the distribution of these putative peptide synthetase and polyketide synthase genes in toxic and non-toxic Nodularia strains. In this study, a protein phosphatase inhibition assay was used to determine the toxicity of Nodularia strains from around the world. Results were then utilized for the comparative study of possible toxin biosynthesis genes. The distribution of the genetic basis for these secondary metabolite biosynthetic pathways and toxin production is discussed with regards to the evolution of this cyanobacterial genus.

2 Materials and methods

2.1 Cyanobacterial strains and culturing

Cyanobacterial strains (Table 1) were supplied by the Commonwealth Science and Industry Research Organization (CSIRO) Marine Laboratories, Hobart, Australia, and the Department of Biological Sciences, Wright State University, Dayton, USA. Strains were also obtained from the Pasteur Culture Collection (PCC), the Culture Collection of Algae at the University of Texas (UTEX) and the Czechoslovak Database of Algae and Cyanobacteria (CDAC). Environmental samples were collected from Lake Alexandrina and stored lyophilized prior to DNA extraction. All strains were maintained in sterile BG-11 or JM media with no additional salt, at a constant growth temperature of 25°C and a light intensity of approximately 1500 lx.

View this table:
1

Cyanobacterial strains used in this study and their ability to produce nodularin as determined by protein phosphatase inhibition assay

2.2 DNA extraction, amplification and sequencing

Total genomic DNA was extracted from exponential phase cultures, after 3 weeks growth, by an SDS/lysozyme-based method used previously for cyanobacteria and plants [15].

Amplification and sequencing of putative peptide synthetase regions from toxic Nodularia strains using oligonucleotide primers FAA and RAA were performed as previously described [8]. Amplification of polyketide synthase regions was performed using degenerate oligonucleotide primers DKF and DKR (Table 2). Nodularia specific oligonucleotide primers (Table 2) were designed from consensus sequences of the peptide synthetase and polyketide synthase PCR products. Amplification of 16S rRNA fragments was performed as previously described [5,16]. Thermal cycling was performed in a PCR Sprint Temperature Cycling System machine (Hybaid Limited, Middlesex, UK) or a GeneAmp PCR System 2400 Thermocycler (Perkin Elmer Corporation, Norwalk, USA). The initial denaturation step at 94°C for 2 min was followed by 30 cycles of DNA denaturation at 94°C for 5 s, primer annealing for 10 s at the corresponding annealing temperature (Table 2) and DNA strand extension at 72°C for 20 s for specific PCR and 1 min for degenerate PCR, and a final extension step at 72°C for 7 min.

View this table:
2

Peptide synthetase and polyketide synthase degenerate and specific oligonucleotide primer sequences and PCR annealing temperatures

Automated sequencing was performed using the Prism Big Dye cycle sequencing system and a model 373 sequencer (Applied Biosystems Inc., Foster City, CA, USA). Sequence data were analyzed using the ABI Prism Auto-Assembler computer program and percentage similarity and identity to other translated sequences determined using BLAST in conjunction with the National Center for Biotechnology Information (NIH, MD, USA).

2.3 Phylogenetic analysis

Peptide synthetase and polyketide synthase gene sequences were aligned using the program Pileup from GCG and the multiple-sequence alignment tool from Clustal W [17]. The input order of taxa was randomized for all sequence alignments and phylogenetic inference programs. Genetic distances (D) between strains were calculated using the formula described by Jukes and Cantor [18], D=−3/4 ln(1−4/3d), where d is the level of sequence dissimilarity. The phylogenetic inference protocols DNADIST, NEIGHBOR, SEQBOOT, and CONSENSE were supplied by the PHYLIP package (version 3.57c) [19]. Sequence manipulation and phylogeny programs were accessed through the Australian National Genome Information Service (Sydney, Australia). DNA sequences were used in the phylogenetic inferences since only sequences from a single species were determined and codon usage was assumed to be evolutionarily negligible.

2.4 Toxicity assays

Presence of nodularin was determined using a protein phosphatase 2A (PP-2A) inhibition assay as previously described [20]. The toxin was liberated from cells by multiple freezing and thawing. Samples were then incubated with PP-2A enzyme (Promega, Madison, WI, USA) at 37°C for 80 min. The toxicity of the sample was determined by assaying the amount of substrate p-nitrophenol phosphate converted to nitrophenol phosphate. The relative concentration of nodularin was determined from a standard curve of inhibition of PP-2A by various concentrations of microcystin-LR. Levels of nodularin were expressed as pmol of microcystin equivalents per mg of total protein.

3 Results and discussion

3.1 Identification of peptide synthetase and polyketide synthase homologues in Nodularia

Putative peptide synthetase gene sequence was obtained from the toxic N. spumigena strains NSOR10, BY1 and HEM using the previously described microcystin synthetase specific primer set FAA and RAA in a PCR with a less stringent annealing temperature of 45°C. This primer set was designed from non-conserved regions within the acyladenylate-forming domain for the amplification of peptide synthetase modules from microcystin-producing cyanobacteria [8]. BLAST X sequence analysis showed that the putative peptide synthetase fragment from NSOR10 and BY1 using FAA/RAA primers gave greatest identity (69% and 64%, respectively) and similarity (77% and 78%, respectively) to the peptide synthetase open reading frame (ORF) mcyC of the microcystin synthetase gene cluster isolated from Microcystis aeruginosa PCC7806 (GenBank accession number AAF00962) [21]. Peptide synthetase sequence amplified from the strain HEM had greatest identity (63%) and similarity (74%) to the peptide synthetase nosD from the nostopeptolide synthetase gene cluster isolated from Nostoc sp. (GenBank accession number AAF17281). The conserved adenylation domain regions A4–A7 [10] were identified in all three N. spumigena peptide synthetase sequences.

In order to amplify putative polyketide synthase gene sequences from Nodularia, the degenerate primers DKF/DKR were designed from conserved regions within the alignment of the polyketide gene region from microcystin synthetase [21] against polyketide synthase genes isolated from Mycobacteria (GenBank accession number U00023). This primer set was used for the identification of polyketide synthase orthologues in toxic N. spumigena strains NSOR10, HEM and BY1. All three sequences were highly similar and analysis by BLAST X found sequence identity (82%, 80% and 81%, respectively) and similarity (90%, 87% and 88%, respectively) to the polyketide synthase ORF mcyD from M. aeruginosa PCC7806 (GenBank accession number AAF00959) [15]. Sequence analysis indicated that a type I ketosynthase region had been amplified from strains of N. spumigena.

3.2 Characterization of putative nodularin biosynthesis genes in toxic and non-toxic Nodularia

Specific primers NPF/NPR were created to screen for a putative peptide synthetase region in toxic and non-toxic geographically diverse strains of Nodularia. As previously shown, sequences from BY1 and HEM using FAA/RAA primers were not highly related [8]. We were, however, able to align the putative peptide synthetase regions from the toxic strains BY1 and NSOR10. Hence the Nodularia specific peptide synthetase primer set, NPF/NPR (Table 2), was designed from the alignment of FAA/RAA PCR product sequences from NSOR10 and BY1 against that of M. aeruginosa PCC7806. Primers were designed in regions conserved between the nodularin-producing strains but not present within the microcystin synthetase gene of M. aeruginosa PCC7806 such that a 150-bp putative peptide synthetase gene fragment was only amplified from nodularin-producing strains.

Screening for the putative polyketide synthase region from nodularin-producing strains was performed using the primer set MNKF/MNKR (Table 2). This primer set was designed from the alignment of all three polyketide synthase-like sequences from strains NSOR10, BY1 and HEM with the polyketide synthase region from M. aeruginosa PCC7806. This primer set was designed to detect a 210-bp putative polyketide synthase region from toxic strains of Nodularia and M. aeruginosa.

Amplification using the NPF/NPR and MNKF/MNKR primers was able to detect putative peptide synthetase and polyketide synthase regions, respectively, from all toxic N. spumigena strains isolated from Australia, the Baltic Sea and New Zealand (Fig. 2). PCR amplifications were also possible from toxic N. harveyana PCC7804. There were no toxin-associated putative peptide synthetase or polyketide synthase gene homologues amplified from Nodularia strains found to be non-toxic by the protein phosphatase inhibition assay, with the exception of N. spumigena NSBL05. No amplification was observed from the cyanobacterial strains Cylindrospermopsis raciborskii AWT205, Anabaena circinalis AWQC118a and Synechocystis PCC6803 (Fig. 2). A polyketide synthase region, only, was amplified from M. aeruginosa PCC7806 using Nodularia specific primer pairs. As PCR controls, the amplification of a cyanobacterial 16S rRNA gene fragment is also shown in all cyanobacterial isolates (Fig. 2).

2

Analysis of peptide synthetase and polyketide synthase regions from toxic and non-toxic Nodularia. Lanes 1–26 correspond to PCR fragments from the strains N. spumigena NSBL05, NSBR01, NSGL02A10, NSKR07, NSLA01, NSLA02A4, NSOR04, NSOR10, NSOR12, NSPH02, NSPH05A14, L575, BY1, and HEM, N. harveyana PCC7804, N. sphaerocarpa 1966/93-1, N. spumigena HKVV and PCC73104, Escherichia coli, N. harveyana UTEX-B2092, UTEX-B2093 and 1983/300, Cylindrospermopsis, Anabaena, and Synechocystis and a no DNA control. The marker lanes (M) contain PhiX174/HaeIII. Cyanobacterial specific 16S rRNA PCR [16] amplified a 400-bp fragment from all cyanobacterial strains analyzed (CYAN). Polyketide synthase specific primer set amplified a 300-bp fragment (PKS) from all toxic Nodularia strains and from non-toxic N. sphaerocarpa 1966/93-1. Peptide synthetase specific primer set amplified a 150-bp fragment (PS) from all toxic Nodularia strains.

Unusually, the non-toxic strain N. spumigena NSBL05 isolated from Australia has shown the presence of both the putative peptide synthetase and polyketide synthase regions. Sequencing of these regions also gave strong similarity to other peptide synthetase and polyketide synthase regions isolated from toxic Nodularia in this study. Recent results show that this strain is closely related to other toxic strains of N. spumigena isolated from Australia, New Zealand and the Baltic Sea by phylogenetic analysis of 16S rDNA sequences [5]. The non-toxic nature of NSBL05 may be explained by a mutation event, in either the regulon or in another vital region of the gene cluster resulting in a non-functional enzyme complex.

The specific primers were tested for their ability to detect nodularin synthetase genes in environmental samples. A bloom sample of Nodularia taken from Lake Alexandrina (1995) [2] was tested by PCR using specific peptide synthetase and polyketide synthase primers as well as 16S rRNA primers specific for members of the Nodularia genus, N. spumigena [5] or, more generally, of the cyanobacteria [16] (Fig. 3). Both putative nodularin synthetase gene regions (peptide and polyketide) were detected along with the cyanobacterial, Nodularia and N. spumigena specific 16S rRNA amplification products from the bloom sample. Mixed cyanobacterial blooms were also simulated by combining a toxic or a non-toxic Nodularia strain with toxic Microcystis and Anabaena strains. The specific toxin primers and the specific 16S rRNA primers correctly identified the presence of toxic Nodularia in these natural and simulated cyanobacterial communities (Fig. 3).

3

Analysis of the presence of peptide synthetase and polyketide synthase regions from an environmental sample containing a toxic Nodularia bloom and mixed cyanobacterial communities simulated to consist of Nodularia, toxic Microcystis and toxic Anabaena. Lanes 1–6 correspond to Lake Alexandrina bloom sample, mixed cyanobacterial community containing toxic Nodularia, mixed cyanobacterial community containing non-toxic Nodularia, toxic N. spumigena NSOR10, C. raciborskii A205 and a negative control, respectively. The marker lanes (M) contain PhiX174/HaeIII. Cyanobacterial specific 16S rRNA PCR [16] amplified a 400-bp fragment in all samples analyzed (CYAN). Nodularia genus specific 16S rDNA PCR [5] amplified a 780-bp fragment from the bloom, mixed communities containing Nodularia and from NSOR10 (NOD). Toxic N. spumigena specific 16S rDNA PCR [11] amplified a 200-bp fragment from the bloom sample, the mixed community containing toxic Nodularia and from NSOR10 (TOX). Polyketide synthase specific primer set amplified a 300-bp fragment (PKS) from the bloom sample, the mixed community containing toxic Nodularia, the mixed community containing toxic Microcystis and non-toxic Nodularia and from N. spumigena NSOR10. Peptide synthetase specific primer set amplified a 150-bp fragment (PS) from the bloom sample, the mixed community containing toxic Nodularia and from NSOR10.

3.3 Phylogenetic analysis of Nodularin biosynthesis genes

The amplified peptide synthetase and polyketide synthase regions were then sequenced in one strain from each phylogenetic cluster as determined by 16S rRNA gene phylogeny [5]. These included the Baltic Sea strains HEM and BY1, Australian strains NSOR10 and NSPH02, New Zealand strain L575, N. harveyana PCC7804 and non-toxic Australian strain NSBL05. Sequencing confirmed the amplified region was a putative peptide synthetase or polyketide synthase in all strains.

Phylogenetic analysis was performed on the polyketide synthase and peptide synthetase DNA sequences from the above strains (Fig. 4). Since amino acid alignments from strains of the genus Nodularia showed high protein sequence similarities (Fig. 5), DNA sequence analysis was used to delineate between the strains. DNA sequence alignments of the polyketide synthase regions (220 bp) found that all sequences were highly related and clustered separately from the outgroup, M. aeruginosa PCC7806. Similarly, phylogenetic analysis was performed on DNA sequences of the peptide synthetase regions (152 bp). Strains appeared to separate into two closely related clusters which were distinct from the microcystin synthetase sequence. Sequences indicated that the polyketide synthase is more highly conserved between Nodularia strains than that of the putative peptide synthetase gene. In both cases, the clustering does not, however, relate to 16S rRNA phylogenetic clusters [5] with toxic N. harveyana clustering closely with all other Nodularia strains. Evolutionary analysis was also performed on protein sequences, revealing similar evolutionary affiliations (data not shown). These results may indicate the horizontal acquisition of toxigenicity in Nodularia.

4

Phylogenetic analysis of (a) peptide synthetase and (b) polyketide synthase regions. Genetic distances (D) between strains were calculated using the formula described by Jukes and Cantor [18], D=−3/4 ln(1−4/3d), where d is the level of sequence dissimilarity. The phylogenetic inference protocols DNADIST, NEIGHBOR, SEQBOOT, and CONSENSE were supplied by the PHYLIP package (version 3.57c) [19]. The scale bar represents two base substitution per 100 nucleotides. Significant bootstrap values [19] calculated from 1000 bootstrap trees are indicated at the nodes.

5

Amino acid sequence alignments of (a) peptide synthetase region and (b) polyketide synthase regions from N. spumigena strains NSOR10, HEM, L575, NSBL05, NSPH05A14, N. harveyana PCC7804 and M. aeruginosa PCC7806.

3.4 Toxicity of Nodularia strains

Levels of toxicity for strains of Nodularia were greater than 8000 pmol mg−1 protein and were as high as 62 000 pmol mg−1 protein for strain NSKR07 (Table 1). Since nodularin inhibits PP-2A with approximately the same potency as microcystin-LR [22,23], the apparent concentration of nodularin was determined using a standard curve of PP-2A inhibition by known concentrations of microcystin. Toxicity of M. aeruginosa PCC7806 was found to be 3000 pmol mg−1 of total cellular protein. All N. spumigena strains isolated from Australia and the Baltic Sea tested exhibited higher levels of PP-2A inhibition than M. aeruginosa PCC7806 except N. spumigena NSBL05 and N. spumigena/sphaerocarpa HKVV which appeared to produce no toxin as determined by the biochemical assay. N. spumigena strain PCC73104, isolated from alkaline soil in Canada, also produced no toxin. All strains of N. harveyana and N. sphaerocarpa analyzed were not toxic by protein phosphatase assay, with the exception of N. harveyana PCC7804 which appeared to be highly toxic by PP-2A inhibition assay. This strain is believed to produce a nodularin derivative, having a longer high performance liquid chromatography (HPLC) retention time [24], with an m/z 839 by MALDI-TOF mass spectrometry, possibly due to an additional methylation or unusual amino acid content (M. Erhard, personal communication, 2000).

4 Conclusions

We have identified peptide synthetase-like and polyketide synthase-like gene regions in Nodularia, the presence of which reflects the toxicity of the strain. The PP-2A inhibition assay was used to quantify the nodularin production by Nodularia strains using microcystins-LR as a standard. The increasing problem of Nodularia blooms in Australia and the Baltic Sea has led to the need for rapid detection of toxic Nodularia blooms in conjunction with current tests such as HPLC and organism bioassays. This paper reports the design of specific oligonucleotide primers for the PCR detection of peptide synthetase and polyketide synthase genes in toxic Nodularia laboratory strains as well as in toxic environmental bloom samples. This specific PCR, when used in conjunction with previously described Nodularia specific 16S rRNA PCR [5], is able to distinguish toxic Nodularia from other cyanobacteria. In addition, the results of this study show that the presence of both the peptide synthetase and polyketide synthase regions analyzed correlates directly with the taxonomic group, N. spumigena, and associated nodularin-based toxicity of the sample.

Acknowledgements

This work was financially supported by the Australian Research Council (ARC). M.C.M. is jointly funded by the ARC and the CRC for Water Quality and Treatment. B.A.N. is a fellow of the ARC. Thank you to Dr. Brenton Nicholson for supplying us with a sample of the Lake Alexandrina Nodularia bloom.

References

  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
View Abstract