• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of bmcbiocBioMed Centralsearchsubmit a manuscriptregisterthis articleBMC Biochemistry
BMC Biochem. 2009; 10: 8.
Published online Mar 30, 2009. doi:  10.1186/1471-2091-10-8
PMCID: PMC2679770

Characterisation of the paralytic shellfish toxin biosynthesis gene clusters in Anabaena circinalis AWQC131C and Aphanizomenon sp. NH-5

Abstract

Background

Saxitoxin and its analogues collectively known as the paralytic shellfish toxins (PSTs) are neurotoxic alkaloids and are the cause of the syndrome named paralytic shellfish poisoning. PSTs are produced by a unique biosynthetic pathway, which involves reactions that are rare in microbial metabolic pathways. Nevertheless, distantly related organisms such as dinoflagellates and cyanobacteria appear to produce these toxins using the same pathway. Hypothesised explanations for such an unusual phylogenetic distribution of this shared uncommon metabolic pathway, include a polyphyletic origin, an involvement of symbiotic bacteria, and horizontal gene transfer.

Results

We describe the identification, annotation and bioinformatic characterisation of the putative paralytic shellfish toxin biosynthesis clusters in an Australian isolate of Anabaena circinalis and an American isolate of Aphanizomenon sp., both members of the Nostocales. These putative PST gene clusters span approximately 28 kb and contain genes coding for the biosynthesis and export of the toxin. A putative insertion/excision site in the Australian Anabaena circinalis AWQC131C was identified, and the organization and evolution of the gene clusters are discussed. A biosynthetic pathway leading to the formation of saxitoxin and its analogues in these organisms is proposed.

Conclusion

The PST biosynthesis gene cluster presents a mosaic structure, whereby genes have apparently transposed in segments of varying size, resulting in different gene arrangements in all three sxt clusters sequenced so far. The gene cluster organizational structure and sequence similarity seems to reflect the phylogeny of the producer organisms, indicating that the gene clusters have an ancient origin, or that their lateral transfer was also an ancient event. The knowledge we gain from the characterisation of the PST biosynthesis gene clusters, including the identity and sequence of the genes involved in the biosynthesis, may also afford the identification of these gene clusters in dinoflagellates, the cause of human mortalities and significant financial loss to the tourism and shellfish industries.

Background

Paralytic shellfish poisoning (PSP) is a syndrome acquired through the consumption of contaminated shellfish or drinking water. Its symptoms include numbness and ascending paralysis followed by respiratory arrest [1]. Toxicity is mediated by a group of toxins collectively referred to as paralytic shellfish toxins (PSTs) or saxitoxins (STX).

The global occurrence of PSTs coupled with their chemical stability and high toxicity, presents a formidable problem for marine and freshwater regulating bodies, while detrimentally affecting the health of humans and animal worldwide [2,3]. STX and its analogues are potent neurotoxic alkaloids. PSTs have been shown to specifically block voltage-gated sodium and calcium channels [4,5], and prolong the gating of potassium channels in heart cells [6]. This mechanism prevents the conduction of a neural action potentials, paralysing the victim, and has also been shown to exhibit a cardio-depressory effect [1]. Additionally, the globally observed abundance of PST producing microorganisms, specifically dinoflagellates, cause substantial economic damage to the fishing industry, mainly due to closure of fisheries affected by PST producing blooms, as well as the regulatory requirement for expensive toxin monitoring programs [7].

The parent compound of PSTs, STX, is a tricyclic perhydropurine alkaloid, which can be substituted at various positions, leading to more than 30 naturally occurring STX analogues [8-13] (Figure (Figure1).1). The synthesis of PSTs has been reported in freshwater and marine organisms alike, spaning two kingdoms of life. The marine microorganisms, dinoflagellates, belonging to the genera Alexandrium, Pyrodinium and Gymnodinium have been reported to produce PSTs [14-16]. In freshwater systems several filamentous species of cyanobacteria, such as Anabaena, Aphanizomenon, Cylindrospermopsis and Lyngbya are also known to produce PSTs [17-20].

Figure 1
Chemical structure of the major paralytic shellfish toxins (PST). a PSTs identified in Anabaena circinalis AWQC131C, b PSTs identified in Aphanizomenon. sp. NH-5. STX, saxitoxin; GTX, gonyautoxin; dc, decarbamoyl.

The occurrence of a neurotoxin, originally termed aphatoxin, from Aphanizomenon flos-aquae was first demonstrated by Sawyer et al. [21], where a copper sulfate treatment of a cyanobacterial bloom led to the death of six tonnes of fish in Kezar Lake, New Hampshire. PST production was later confirmed in the cultured strain Aphanizomenon flos-aquae NH-5, isolated from a small pond near Durham, New Hampshire. This study identified the presence of saxitoxin (STX) and neosaxitoxin (neoSTX), together with further unidentified toxic fractions [17]. Li et al. [22] re-evaluated the morphological based taxonomy of the toxic Aphanizomenon flos-aquae NH-5 strain. They compared the 16S rRNA gene sequences and morphology of this strain with six other strains of Aphanizomenon flos-aquae and reclassified it as Aphanizomenon sp. NH-5. According to morphological and gene sequence data, four of the investigated strains of Aphanizomenon flos-aquae, that have not been shown to produce toxins, are grouped together in the phylogenetic tree and were delineated from the branch represented by the two known toxin-producing Aphanizomenon isolates. Furthermore, Li et al. [23] reclassified the paralytic shellfish toxin-producing Aphanizomenon flos-aquae LMECYA 31 as Aphanizomenon issatschenkoi based on morphological and 16S rRNA gene sequence characteristics. The Aphanizomenon strains in this study formed a monophyletic cluster with three other species designations; the fascicle-forming Aphanizomenon flos-aquae, Aphanizomenon gracile strains having solitary trichomes, and Aphanizomenon issatschenkoi strains characterized by solitary trichomes that have tapered ends. More recently, a different Aph. flos-aquae strain from Portugal has been isolated and shown to produce the PSTs, STX, neosaxitoxin (neoSTX), gonyautoxin 5 (GTX5), gonyautoxin 6 (GTX6) and decarbamoyl saxitoxin (dcSTX) [3].

Similarly, Anabaena circinalis is a common toxic bloom-forming planktonic freshwater cyanobacterium with a global distribution and an unusual geographical segregation of toxin production [24]. Australian reports of neurotoxic cyanobacterial blooms have principally implicated A. circinalis as the toxin producing species. In 1991 a neurotoxic bloom of A. circinalis covering over 1000 km of the Darling River in Australia, reportedly with concentrations of up to 500,000 cells per millilitre, was the cause of cattle mortality [25,26]. A following comprehensive study of the Murray-Darling basin in 1994 implicated A. circinalis in all neurotoxic blooms in Australia, and also showed A. circinalis to produce PSTs, mainly STX, GTX2/3 and dcGTX2/3 [18,27]. A detailed chemical analysis of the cultured strain A. circinalis AWQC131C revealed the presence of STX, GTX2/3, C1/2, dcSTX and dcGTX2/3 [28].

We have recently proposed a putative STX biosynthesis gene cluster in Cylindrospermopsis raciborskii T3 [29]. The putative STX biosynthesis gene cluster (sxt) in C. raciborskii T3 spans approximately 36 kb, and encodes genes involved in the biosynthesis, regulation and export of PSTs (Figure (Figure2).2). Some of the genes identified in the C. raciborskii T3 gene cluster have not have been assigned a function due to their low level of structural homology to proteins in available databases. These genes may therefore represent novel enzyme families. Since C. raciborskii T3 is not genetically transformable, mutagenesis of the sxt cluster was not possible. The choice of identifying and characterizing the two novel putative PST biosynthesis gene clusters in Anabaena and Aphanizomenon was principally motivated by the detrimental effect the producer organisms have on water quality and public health. Furthermore, comparative analysis of the differences identified in toxin production, that is, the lack of neoSTX and all other N1-hydroxylated PSTs in A. circinalis AWQC131C, may provide further insights into the PST biosynthetic machinery [17,28]. The close taxonomic and phylogenetic relatedness of these two PST-producing organisms, also provides insights into the evolutionary history of the putative PST biosynthesis gene clusters, revealing whether they are of a common ancestor or laterally derived via horizontal gene transfer.

Figure 2
Structure of the paralytic shellfish toxin biosynthesis cluster identified in A. Aphanizomenon. sp. NH-5, B. Anabaena circinalis AWQC131C, C. Cylindrospermopsis raciborskii T3. The gene cluster schematic for C. raciborskii T3 has been adapted from kellmann ...

Results and discussion

Identification of PST gene clusters in A. circinalis AWQC131C and Aph. sp. NH-5

The gene cluster putatively responsible for the biosynthesis of PSTs in the cyanobacterium C. raciborskii T3 has recently been characterized by our group [29]. In an attempt to identify the putative PST biosynthesis gene cluster analogs, from the PST producing organisms A. circinalis AWQC131C and Aph. sp. NH-5, a reverse genetic approach was employed. The primer pair nodF and nodR (Table (Table1)1) was used in a degenerate PCR, targeting conserved regions in carbamoyltransferases (sxtI), which are putatively involved in the transfer of a carbamoyl group from carbamoyl phosphate onto the free hydroxymethyl side chain of the saxitoxin precursor [29,30]. A single amplicon of about 900 bp in size was amplified using genomic DNA isolated from both A. circinalis AWQC131C and Aph. sp. NH-5. The degenerate PCR products were then cloned and clone libraries were constructed, in an attempt to identify all the carbamoyltrasferases in these PST producing organisms. Screening of these clone libraries revealed only one gene fragment in each of the libraries. BLAST similarity searches showed homology between these gene fragments and other carbamoyltransferases. The sxtI gene homologs identified in A. circinalis AWQC131C and Aph. sp. NH-5 were 90% and 91% identical in sequence to the sxtI gene from C. raciborskii T3, respectively, and were 97% identical among themselves (Table (Table2)2) [29]. These gene fragments were consequently good candidates for sxtI homologs, and therefore also PST biosynthesis genes in these cyanobacteria.

Table 1
PCR primer sequences used in this study
Table 2
Similarity and predicted function of PST biosynthesis genes

In order to obtain the DNA sequences of the entire putative PST biosynthesis gene clusters in both A. circinalis AWQC131C and Aph. sp. NH-5, a gene walking technique (known as pan-handle PCR) was employed [31]. Following numerous rounds of sequencing-out and gene-walking reactions from the putative sxtI homologues, the entire gene cluster in both organisms was identified, sequenced and characterized (Figure (Figure2).2). The putative PST gene cluster in both organisms is of comparable size. The A. circinalis AWQC131C sxt gene cluster spans 29 kb, flanked by a β-lactamase, a gene involved in antibiotic resistance, at the 5' end and a smf gene homolog, believed to be involved in DNA uptake at the 3' end (Figure (Figure2).2). The Aph. sp. NH-5 sxt gene cluster is slightly smaller and covers 27.5 kb. It is flanked on both sides by genes coding for photosynthesis machinery, at the 5' end by psbH, a photosystem II reaction centre gene and at the 3' end by a prenyltransferase involved in the synthesis of ubiquinone (Figure (Figure22).

Characterisation of the PST biosynthesis gene clusters

The major PSTs identified in Aph. sp. NH-5 are neoSTX and STX [17] and other toxic fractions that were not identified, although the methods used were not highly sensitive. While the major PSTs identified in A. circinalis AWQC131C were STX, GTX2/3, C1/2, dcSTX and dcGTX2/3 [28]. This observed difference in the toxin profiles is most probably the result of the different genetic backgrounds in these producer organisms, as modification reactions to the STX parent molecule, carried out by the sxt gene cluster tailoring enzymes, putatively result in the formation of the various PSTs.

Both gene clusters characterized in this study contain the same set of genes identified in the putative PST biosynthesis gene cluster (sxt) of C. raciborskii T3 with some exceptions (Table (Table2)2) [29]. Specifically, they do not contain the genes sxtY, sxtZ and ompR, believed to be involved in signal transduction and the transcriptional regulation of PST production in C. raciborskii T3, and therefore might be regulated in a different manner, however it can not be completely excluded that these genes have diverged and transposed to a different locus in the genome. Furthermore, they do not contain the gene sxtF putatively involved in PST transport. Both the A. circinalis AWQC131C and the Aph. sp. NH-5 PST gene clusters contain a gene, denoted sxtPER. Curiously, sxtPER, a gene most similar to members of the drug and metabolite transport family, has not been identified in the C. raciborskii T3 PST gene cluster.sxtPER might therefore fulfil a similar role to sxtF in C. raciborskii T3 and is postulated to be involved in the transport of PSTs in Anabaena and Aphanizomenon. Interestingly, sxtPER has an evolutionary history apparently distinct from sxtM and sxtF and represents a different lineage, while sxtF seems to have arisen by a gene duplication event of sxtM. The phylogeny, function and distribution of cyanobacterial MATE genes are addressed in a separate publication (Pengelly JJL, Mihali TK, Neilan BA: Identification, phylogeny and expression of novel members of the multi-drug and toxic compound extrusion (MATE) family from cyanobacteria, submitted).

Furthermore, the Aph. sp. NH-5 sxt gene cluster is missing a sxtO homolog, an adenylylsulfate kinase. sxtO is present in the A. circinalis AWQC131C sxt gene cluster and the C. raciborskii T3 sxt gene cluster. sxtO is most similar to adenylylsulfate kinases that are involved in the formation of 5'-phosphoadenosine 3'-phosphosulfate (PAPS), which is the sulfate donor for PAPS dependant sulfotransferases (Table (Table2)2) [29]. The toxin profile of Aph. sp. NH-5 has not been fully characterized, however, it has been shown to produce STX and neoSTX and possibly other unidentified toxic fractions [17]. It is therefore not possible to asses whether this strain actually produces sulfated PSTs. As adenylylsulfate kinases are not unique to the PST gene clusters, and are ubiquitous enzymes needed for all PAPS dependant sulfotransferases, it seems plausible that this genetic mutation could be complemented by an additional adenylylsulfate kinase elsewhere in the genome of Aph. sp. NH-5. Curiously, the sxtO homolog identified in A. circinalis AWQC131C sxt gene cluster has very low similarity to sxtO identified in C. raciborskii T3 (Table (Table2),2), and is more similar to other cyanobacterial derived adenylylsulfate kinases. This observation further supports the idea that PST biosynthesis accessory genes may be complimented by gene homologs in the genome of the producer organism.

On the other hand, the A. circinalis AWQC131C putative PST biosynthesis gene cluster does not contain the tailoring gene sxtX, which is most similar to a cephalosporine hydroxylase, and presumably responsible for the hydroxylation of N-1 in STX, thereby converting STX to neoSTX (Figure (Figure3).3). Hence, sxtX is putatively involved in the formation of all the analogs containing a hydroxyl at N-1 (Figure (Figure1).1). In accordance with this finding, A. circinalis AWQC131C does not produce any STX congeners that contain a hydroxyl at N-1. In contrast the Aph. sp. NH-5 sxt gene cluster does contain the tailoring enzyme gene sxtX, and correspondingly been shown to produce the N-1 hydroxylated analog neoSTX [17], further affirming its putative role in the PST biosynthesis gene cluster.

Figure 3
Proposed saxitoxin biosynthetic pathway. Dotted lines indicate additional but not essential tailoring reactions, dotted box indicates reaction only present in Aph. sp. NH-5. See text for detail. Adopted and modified from [29].

The A. circinalis AWQC131C sxt gene cluster does not contain the gene sxtW, which is present in the Aph. sp. NH-5 and the C. raciborskii T3 sxt gene clusters. sxtW is most similar to a ferredoxin and is believed to be involved in electron transport, required for the hydroxylation of C-12 by the two ring-hydroxylating dioxygenases, encoded by sxtT and sxtH. Curiously, the A. circinalis AWQC131C sxt gene cluster contains a small (98 bp), disrupted ORF denoted orf8, with similarity to sxtW and other ferredoxin genes (Table (Table2).2). Orf8 might therefore represent yet another fragment of a PST biosynthesis accessory gene that has been inactivated. It is plausible that an endogenous ferredoxin complements for its loss in A. circinalis AWQC131C, since homologs of ferredoxin (Accession no. Npun_R5178, Ava_5008) have been detected in several recently released cyanobacterial genomes, as evident from BLAST homology searches.

sxtV, a gene most similar to succinate dehydrogenases, originally identified in the C. raciborskii T3 PST gene cluster, is also believed to be involved in this electron transport mechanism. Only a fragmented part of this gene is present in the A. circinalis AWQC131C sxt gene cluster, while sxtV in the Aph. sp. NH-5 sxt gene cluster contains a stop codon interrupting the ORF. It is therefore postulated that this gene is also complemented by another locus in the genome of these producer organisms or is not essential for PST biosynthesis.

Interestingly, the A. circinalis AWQC131C sxt gene cluster also contains partial transposase-like sequences, which interrupt a putative ORF and split it into 3 different fragmented partial ORFs, denoted orf3, orf5 and orf7 (Figure (Figure2).2). This fragmented gene is partially similar to sxtN, homologs of which are present in the A. circinalis AWQC131C and Aph. sp. NH-5 sxt gene clusters, and putatively involved in the sulfation of STX and its derivative [29]. Due to its fragmentation and short coding sequence we were unable to assign a putative role for this gene, but it is assumed to be inactive.

Structural organization

The organization of genes in the, A. circinalis AWQC131C and Aph. sp. NH-5 sxt gene clusters, is very conserved, and has less common features with the organization of the C. raciborskii T3 sxt cluster (Figure (Figure2).2). As depicted in Figure Figure2,2, the putative PST biosynthesis gene cluster can be divided into five segments of coding DNA, denoted as A-E. The A. circinalis AWQC131C sxt gene cluster contains three truncated ORFs (orf 3, 5 and 7) intersected by transposases, however, these are absent from the Aph. sp. NH-5 sxt gene cluster. In the Aph. sp. NH-5 sxt gene cluster, segment C contains the genes sxtX, sxtV and sxtW, whereas in the A. circinalis AWQC131C sxt gene cluster most of segment C is missing, only a fragmented section of sxtV is retained. Furthermore, in the A. circinalis AWQC131C putative PST biosynthesis gene cluster, segment E is adjacent to segment A, whereas in Aph. sp. NH-5 segment E is connected to segment D (Figure (Figure2).2). Therefore, it seems plausible to assume that the A. circinalis AWQC131C sxt gene cluster has had at least two DNA recombination events as compared to the Aph. sp. NH-5 sxt gene cluster, with one event causing the truncation of segment C, thereby losing the genes sxtX, sxtV and sxtW. The other recombination event would have involved segment E, which contains the gene denoted sxtPER, to relocate from its 5' position in Aph. sp. NH-5 to the central region of the gene cluster in A. circinalis AWQC131C (Figure (Figure22).

Biosynthesis of PST congeners

A detailed description of the proposed biosynthesis of the STX parent molecule has recently been published by our group [29]. Briefly, biosynthesis is initiated with SxtA, which contains 4 catalytic domains (Table (Table2,2, Figure Figure33 step 1–2). A methyltransferase domain (sxtA1), a GNAT domain (sxtA2) (loading of acyl carrier protein), an acyl carrier protein (ACP) domain (sxtA3) and an AONS domain (sxtA4) (which acts as a condensation domain). The predicted reaction sequence of SxtA, is the loading of the ACP (SxtA3) with acetate from acetyl-CoA by sxtA2, followed by the SxtA1-catalysed methylation of acetyl-ACP, converting it to propionyl-ACP. The class II aminotransferase domain (AONS), SxtA4, then performs a Claisen-condensation between propionyl-ACP and arginine. The product of SxtA is thus 4-amino-3-oxo-guanidinoheptane. The next step (Figure (Figure33 step 3) is carried out by SxtG, an amidinotransferase that transfers an amidino group from arginine onto the product of SxtA, producing 4,7-diguanidino-3-oxoheptane. This compound is then condensed via SxtB, which is most similar to a cytidine deaminase, in a retro-aldol-like condensation (Figure (Figure33 step 4) and the first heterocycle is formed. The following step (Figure (Figure33 step 5) involves SxtD a desaturase, which introduces a double bond between C-1 and C-5 resulting in the 1,2-H shift between C-5 and C-6. SxtS a 2-oxoglutarate-dependent (2OG) dioxygenase, then performs the consecutive epoxidation of the new double bond, and opening of the epoxide to an aldehyde with concomitant bicyclisation (Figure (Figure33 steps 6–7). The dehydrogenase SxtU then reduces the terminal aldehyde group of the STX precursor (Figure (Figure33 step 8). Thereafter (Figure (Figure33 step 9), SxtH and SxtT, each coding a terminal oxygenase subunit of bacterial phenyl-propionate and related ring-hydroxylating dioxygenases, catalyse the consecutive hydroxylation of C-12, forming dcSTX. Thereafter, SxtI in conjunction with SxtJ and SxtK catalyse a carbamoyltrasfer from carbamoylphosphate onto the free hydroxyl at C-13, forming saxitoxin (STX).

Subsequently, tailoring reactions catalyse the formation of the PST analogues, which are derivatives of the parent molecule STX. One of the most potent STX derivatives is the N-1 hydroxylated saxitoxin analogue neoSTX. As previously mentioned, SxtX has been predicted to be involved in the N-1 hydroxylation of saxitoxin [29]. Therefore we postulate that SxtX, which is not encoded in the A. circinalis AWQC131C PST biosynthesis gene cluster, carries out the conversion of STX to neoSTX (Figure (Figure3).3). A. circinalis AWQC131C is known to produce N-21 and O-22 sulfated STX analogues (GTX2/3, C1/2, dcGTX2/3). The activity of two PAPS dependent sulfotransferases, which were specific for the N-21 of STX and GTX-3/2, and O-22 of 11-hydroxy STX, respectively, have been described from the PSP toxin-producing dinoflagellate Gymnodinium catenatum, although no protein sequence information was obtained due to instability of the enzyme and low yield [32,33]. Both putative PST biosynthesis gene clusters identified in this study contain a gene denoted sxtN, which is most similar to estrogen sulfotransferase and putatively responsible for the sulfation of STX. Therefore, the protein encoded by sxtN, is postulated to be responsible for the sulfation of STX at O-22 resulting in the formation of GTX2 and GTX3 (Figure (Figure3).3). Alternatively SxtN could sulfate N-21 of STX resulting in the formation of C1 and C2. It is further possible that the enzyme catalyses both reactions, or that a sulfotransferase not encoded in the PST biosynthesis gene cluster is involved. Whether the product of this gene sulfates O-22 or N-21 of STX could not be determined at this point, and will require heterologous expression of the enzyme to determine its natural substrate specificities. dcSTX derivatives are postulated to result from the hydrolytic cleavage of carbamoylated STXs. The candidate enzyme encoded by sxtL, harbouring homology to GDSL-lipases, has been proposed to catalyse this cleavage [29]. SxtL would therefore catalyse the formation of dcGTX2/3 from GTX2/3, and putatively convert STX to dcSTX (Figure (Figure3).3). The exact sequence of reactions is not clear at this point, as there is need for heterologous expression of each of these tailoring enzymes and their substrate specificities have to be confirmed experimentally.

Phylogeny of the PST biosynthesis genes

It is interesting to note that the organization of the putative PST gene clusters in A. circinalis AWQC131C and Aph. sp. NH-5 are more similar to each other, than to the organization in the recently identified putative PST gene cluster in C. raciborskii T3 (Figure (Figure2).2). Furthermore, the A. circinalis AWQC131C and Aph. sp. NH-5 sxt genes show overall greater sequence similarity to each other than to the recently identified sxt genes in C. raciborskii T3 (Table (Table2).2). In an attempt to reconstruct the taxonomic phylogeny of the PST producing cyanobacteria A. circinalis AWQC131C, Aph. sp. NH-5, C. raciborskii T3 and L. wollei, a phylogenetic approach was used. 16S rRNA gene sequences for the PST-producing and non-producing cyanobacteria were retrieved from GenBank (Table (Table3).3). An alignment of these sequences was created and a phylogenetic tree based on a 685 bp segment of the 16S rRNA gene was reconstructed. As evident from the phylogenetic tree (Figure (Figure4),4), A. circinalis AWQC131C and Aph. sp. NH-5 form a supported clade together with the (non-toxic) A. circinalis AWQC310F (64% bs) and are more closely related to each other than to C. raciborskii T3. Consequently, the topological organization of the PST biosynthesis gene cluster is reflected by the phylogeny of the producer organism, and they therefore most likely have a paralleled evolution.

Table 3
16S rDNA sequences
Figure 4
Phylogenetic tree of the 685 bp partial 16S rRNA gene from saxitoxin-producing cyanobateria. Anabaena circinalis AWQC131C and Aphanizomenon sp. NH-5 show highest similarity to each other compared to C. raciborskii T3 or L. wollei. Bootstrap confidence ...

Australian isolates of A. circinalis have been shown to be phylogenetically related, forming two distinct branches [24]. Furthermore, genetic screening has shown that only toxic isolates contain the putative genes for PST biosynthesis [29]. An attempt was made to identify the insertion/excision point of the putative PST gene cluster in the Australian A. circinalis isolates. PCR primers AnastartF and Ana30kbR were designed based on the sequences obtained from A. circinalis AWQC131C (Table (Table1),1), as strain representative of the PST-producing lineage. These primers were designed within the ORF of the β-lactamase at the 5' end, and smf at the 3' end of the PST gene cluster in A. circinalis AWQC131C (Figure (Figure5).5). A PCR reaction with this primer set was performed using genomic DNA isolated from the non-toxic A. circinalis 310F strain belonging to the non-PST producing clade, and the PST-producing A. circinalis AWQC131C. The PCR resulted in the amplification of a 1.6 kb segment of DNA from A. circinalis 310F, whereas genomic DNA from A. circinalis AWQC131C did not yield a product. The predicted distance between the primer set in A. circinalis AWQC131C genomic DNA was over 29 Kb and therefore an amplicon was not expected using the conditions described. Sequencing followed by bioinformatic analysis of the 1.6 kb amplicon from A. circinalis 310F revealed two ORFs, one at the 5' end that was similar to a β-lactamase, and another at the 3' end similar to a smf gene, both homologous to the genes identified on either ends of the putative PST biosynthesis gene cluster in A. circinalis AWQC131C. A closer analysis of the A. circinalis 310F amplified region revealed the presence of four direct repeats of the sequence TTCCCTG within the intragenic spacer identified between the two flanking ORFs (Figure (Figure5).5). Insertion sites which are targeted by a transposon are usually innocuous and do not contain repeats, however, the presence of direct repeats is indicative of an excision event that has followed the insertion of a transposon [34]. Further examination of the non-coding regions flanking the PST biosynthesis gene cluster in A. circinalis AWQC131C revealed a relic inverted repeat of the sequence GGCTATCAATTCAGAT on either side of the PST biosynthesis gene cluster that could also indicate a past transposition event. The combination of these findings supports the theory that the non-toxic A. circinalis strains have lost the putative PST biosynthesis gene cluster during an excision event. Together with the observation that the organization of the putative PST biosynthesis gene clusters resembles the phylogeny of the producer organisms, as well as the fact that A. circinalis AWQC131C and Aph. sp. NH-5 PST genes are overall more similar to each other than to the C. raciborskii T3 PST genes (Table (Table2),2), might imply that the sxt gene cluster is of ancient origin. Therefore, the sporadic distribution of PST biosynthesis throughout phylogeny could be explained by the loss of function in non-toxic strains, rather than the acquisition of one or more genes during more recent horizontal transfer in PST producing toxic strains. A similar evolutionary history has recently been proposed for the sporadic distribution of another cyanobacterial toxin biosynthesis gene cluster encoding microcystin synthetase [35].

Figure 5
Putative insertion/excision site of the paralytic shellfish toxin gene cluster, in Australian Anabaena circinalis strains. Diagram not to scale. See text for details.

Conclusion

Here we have described the identification, annotation and characterisation of the putative PST biosynthesis gene clusters in A. circinalis AWQC131C and Aph. sp. NH-5. The putative PST biosynthesis gene cluster presents a mosaic structure, whereby genes have apparently transposed in segments of varying size, resulting in different gene arrangements in all three sxt clusters sequenced so far. This gene cluster arrangement is in agreement with a mobile genetic element, further supported by the identification of multiple transposase like sequences in other recently characterized cyanobacterial toxin and secondary metabolite gene clusters [31,36,37]. The putative point of insertion/excision of the sxt gene cluster in Australian isolates of A. circinalis was also identified. The gene cluster organizational structure and gene sequence identity seem to reflect the phylogeny of the producer organisms, indicating that the gene clusters might have an ancient origin, or that their lateral transfer was also an ancient event. The further sequencing and characterisation of cyanobacterial PST biosynthesis gene clusters in different producer organisms, with different toxin profiles, may provide new insights into the biosynthetic machinery and the enzymes involved at each step. The availability of more putative PST biosynthesis gene sequences from different producer organisms will also enable the better monitoring of algal blooms by water authorities, including the use of PCR-based early warning systems. Future studies into the transcriptional control and physiological roles of the PSTs, furthering our ability to predict and prevent the formation of harmful algal blooms, is also possible given these gene sequences and associated regulatory regions. The knowledge we gain from the characterisation of the PST biosynthesis gene clusters, including the identity and sequence of the genes involved in the biosynthesis, may also afford the identification of these gene cluster in dinoflagellates, the cause of human mortalities and significant financial loss to the tourism and shellfish industries.

Methods

Cyanobacterial strains

A. circinalis AWQC131C and Anabaena circinalis 310F [24] were grown in Jaworski medium [38] in static batch culture at 26°C under continuous illumination (10 μmol m-2s-1), Aph. sp. NH-5 and L. wollei freeze-dried cultures were kindly supplied by W. W. Carmichael [17,19].

DNA extraction

Total genomic DNA was extracted from cyanobacterial cells using the Mo Bio PowerPlant DNA isolation kit (Carlsbad, CA) in accordance with the manufacturers instructions. Genomic DNA was stored at -20°C.

PCR

Insertion/excision point PCR was carried out using the primer set AnastartF and Ana30kbR (Table (Table1),1), and performed in 20 μl reaction volumes containing 1 × Taq polymerase buffer, 2.5 mM MgCl2, 0.2 mM deoxynucleotide triphosphates, 10 pmol each of the forward and reverse primers, 50 ng genomic DNA and 0.2 U of Taq polymerase (Fisher Biotech, Australia). Thermal cycling was performed in a GeneAmp PCR System 2400 Thermal cycler (Perkin Elmer Corporation, Norwalk, CT). Cycling began with a denaturing step at 94°C for 4 min followed by 30 cycles of denaturation at 94°C for 10 s, primer annealing at 55°C for 25 s and a DNA strand extension at 72°C for 2 min. Amplification was completed by a final extension step at 72°C for 7 min.

Degenerate PCR

Degenerate PCR primers targeting the carbamoyltransferase sxtI used in this study, nodF and nodR (Table (Table1),1), were designed ex-silico from sequence alignments using ClustalX [39]. Degenerate PCR was performed in 20 μl reaction volumes containing 1 × Taq polymerase buffer, 2.5 mM MgCl2, 0.2 mM deoxynucleotide triphosphates, 25 pmol each of the forward and reverse primers, 50 ng of genomic DNA and 0.2 U of Taq polymerase (Fisher Biotech, Australia). Thermal cycling was performed in a GeneAmp PCR System 2400 Thermal cycler (Perkin Elmer Corporation, Norwalk, CT). Cycling began with a denaturing step at 94°C for 4 min followed by 35 cycles of denaturation at 94°C for 10 s, primer annealing at 50°C for 30 s and a DNA strand extension at 72°C for 45 s. Amplification was completed by a final extension step at 72°C for 5 min.

Gene walking

The sequence of unknown regions that were flanking candidate genes was determined by an adaptor-mediated (Pan-handle) PCR method [40] that was modified as previously described [31]. Short adaptor DNA was ligated to digested genomic DNA, and a specific genomic outward-facing primer in combination with an adaptor primer was then used to amplify the unknown regions of the genome. Twenty picomoles of T7 adaptor were added to each reaction mixture, containing 1 μg of genomic DNA, 10 U of blunt-ended restriction enzyme, and 5 U of T4 ligase (Promega) in 1 × One Phor All buffer (Amersham/Pharmacia). The one-step digestion and ligation reaction mixture was incubated at room temperature overnight. The single-stranded end of the adaptor was blocked in a solution containing 1 × PCR buffer (Fischer Biotech), 4 mM MgCl2, and 12.5 μM ddNTP with 1 U of Taq DNA polymerase (Fischer Biotech). Thermal cycling was performed in a PCR Sprint temperature cycling system machine (Hybaid Ltd) with an initial step at 70°C for 15 min followed by 10 cycles of DNA denaturation at 95°C for 10 s, DNA reannealing at 40°C for 1 min, and extension of the strand with ddNTP at 70°C for 1 min. Following the PCR cycles, the reaction mixture was incubated with 1 U of shrimp alkaline phosphatase (Boehringer Mannheim, Göttingen, Germany) at 37°C for 20 min, and the enzyme was heat inactivated at 85°C for 5 min.

The flanking region PCR mixture contained 1 to 2 μl of adaptor-ligated DNA, 10 pmol of adaptor primer, 10 pmol of a genome-specific oligonucleotide primer and 0.5 U of a mixture of 10:1 Taq polymerase/PFU (Fischer Biotech, Australia). PCR cycling was performed as described above, with DNA strand extension at 72°C for 3 min. In addition, the primer annealing temperature was decreased for the first 10 cycles, by 1°C at each cycle, from 65 to 55°C, followed by primer annealing at 55°C for a further 25 cycles.

Agarose gel electrophoresis

Amplified DNA was separated by gel electrophoresis with 1–2% agarose in TAE buffer (40 mM Tris-acetate, 1 mM EDTA, pH 7.8), and visualized by UV transillumination after staining with ethidium bromide (0.5 μg/ml). Where multiple amplicons were detected during the gel electrophoresis, single amplicons were excised from the gels and purified using the Promega Wizard® SV Gel and PCR Clean-Up (Wisconsin, USA), prior to sequencing.

Gene cloning

Clone libraries were created using the pGEM®-T Easy cloning kit from Promega (Wisconsin, USA) in accordance with the manufacturers instructions.

DNA sequencing

Automated DNA sequencing was performed using the PRISM Big Dye cycle sequencing system and a model 373 sequencer (Applied Biosystems, Foster City, CA).

Bioinformatic analysis

Sequence data were analysed using ABI Prism-Autoassembler software, while identity/similarity values to other translated sequences were determined using BLAST against the non-redundant (nr) data set, in conjunction with the National Center for Biotechnology Information (NIH, Bethesda, MD). Fugue blast http://www-cryst.bioc.cam.ac.uk/fugue/ was used to identify distant homologs via sequence-structure comparisons. The gene clusters were assembled using the software package Phred, Phrap, and Consed http://www.phrap.org/phredphrapconsed.html, and open reading frames were identified manually.

Phylogenetic analysis

16S rDNA sequences of the studied cyanobacteria were retrieved from GenBank (Table (Table3).3). Sequence alignments were preformed using ClustalX [41] and were manually edited to exclude ambiguous regions. Maximum-likelihood phylogenetic trees were reconstructed using PhyML, with four gamma substitution rate categories and a 1000 bootstrap replicas [42] (available on-line at http://atgc.lirmm.fr/phyml/). The optimal DNA substitution model for the data set (GTR+G) was identified using the AiC in the program Modeltest [43].

Nucleotide sequence accession number

Nucleotide sequences were submitted to GenBank. A. circinalis AWQC131C PST biosynthesis gene cluster is available under accession number DQ787201. Aph. sp. NH-5 PST biosynthesis gene cluster is available under accession number EU603710. A. circinalis 310F insertion excision sequence is available under accession number EU603709.

Abbreviations

PKS: polyketide synthase; PST: paralytic shellfish toxin; PSP: paralytic shellfish poisoning; STX: saxitoxin; GTX: gonyautoxin; dcSTX: decarbamoyl saxitoxin; neoSTX: neosaxitoxin; ACP: acyl carrier protein; PAPS: 5'-phosphoadenosine 3'-phosphosulfate.

Authors' contributions

TKM carried out all experimental work, acquired, analysed and interpreted data and drafted the manuscript. RK participated in drafting of the manuscript and data interpretation. BAN participated in drafting of the manuscript, experimental design, data interpretation and supervised the overall progress of this project.

References

  • Llewellyn LE. Saxitoxin, a toxic marine natural product that targets a multitude of receptors. Natural Product Reports. 2006;23:200–222. doi: 10.1039/b501296c. [PubMed] [Cross Ref]
  • Kaas H, Henriksen P. Saxitoxins (PSP toxins) in Danish lakes. Water Research. 2000;34:2089–2097. doi: 10.1016/S0043-1354(99)00372-3. [Cross Ref]
  • Pereira P, Onodera H, Andrinolo D, Franca S, Araujo F, Lagos N, Oshima Y. Paralytic shellfish toxins in the freshwater cyanobacterium Aphanizomenon flos-aquae, isolated from Montargil reservoir, Portugal. Toxicon. 2000;38:1689–1702. doi: 10.1016/S0041-0101(00)00100-8. [PubMed] [Cross Ref]
  • Kao CY, Levinson SR, eds . Tetrodotoxin, saxitoxin, and the molecular biology of the sodium channel. New York: The New York Academy of Science; 1986.
  • Chang Z, Sitachitta N, Rossi JV, Roberts MA, Flatt PM, Jia J, Sherman DH, Gerwick WH. Biosynthetic pathway and gene cluster analysis of curacin A, an antitubulin natural product from the tropical marine cyanobacterium Lyngbya majuscula. Journal of Natural Products. 2004;67:1356–1367. doi: 10.1021/np0499261. [PubMed] [Cross Ref]
  • Wang J, Salata JJ, Bennett PB. Saxitoxin is a gating modifier of HERG K+ channels. The Journal of General Physiology. 2003;121:583–598. doi: 10.1085/jgp.200308812. [PMC free article] [PubMed] [Cross Ref]
  • Hallegraeff GM. In: Harmful algal blooms: a global overview. Hallegraeff GM, Anderson DM, Cembella AD, editor. Manual on Harmful Marine Microalgae, UNESCO, Paris; 2003. pp. 25–49. ISBN 923403871.
  • Arakawa O, Noguchi T, Shida Y, Onoue Y. Occurrence of carbamoyl-N-hydroxy derivatives of saxitoxin and neosaxitoxin in a xanthid crab Zosimus aeneus. Toxicon. 1994;32:175–183. doi: 10.1016/0041-0101(94)90106-6. [PubMed] [Cross Ref]
  • Arakawa O, Nishio S, Noguchi T, Shida Y, Onoue Y. A new saxitoxin analogue from a xanthid crab Atergatis floridus. Toxicon. 1995;33:1577–1584. doi: 10.1016/0041-0101(95)00106-9. [PubMed] [Cross Ref]
  • Oshima Y. Postcolumn derivatization liquid chromatographic method for paralytic shellfish toxins. Journal of AOAC International. 1995;78:528–532.
  • Onodera H, Satake M, Oshima Y, Yasumoto T, Carmichael WW. New saxitoxin analogues from the freshwater filamentous cyanobacterium Lyngbya wollei. Natural Toxins. 1997;5:146–151. [PubMed]
  • Zaman L, Arakawa O, Shimosu A, Shida Y, Onoue Y. Occurrence of a methyl derivative of saxitoxin in Bangladeshi freshwater puffers. Toxicon. 1998;36:627–630. doi: 10.1016/S0041-0101(97)00086-X. [PubMed] [Cross Ref]
  • Negri A, Stirling D, Quilliam M, Blackburn S, Bolch C, Burton I, Eaglesham G, Thomas K, Walter J, Willis R. Three novel hydroxybenzoate saxitoxin analogues isolated from the dinoflagellate Gymnodinium catenatum. Chemical Research in Toxicology. 2003;16:1029–1033. doi: 10.1021/tx034037j. [PubMed] [Cross Ref]
  • Shimizu Y. In: Chemistry and distribution of deleterious dinoflagellate toxins. Faulkner D, Fenical W, editor. New York: Plenum; 1977. pp. 261–269.
  • Harada T, Oshima Y, Yasumoto T. Structure of two paralytic shellfish toxins, gonyautoxins V and VI, isolated from a tropical dinoflagellate Pyrodinium bahamense var. compressa. Agricultural & Biological Chemistry. 1982;46:1861–1864.
  • Oshima Y, Hasegawa M, Yasumoto T, Hallegaeff G, Blackburn S. Dinoflagellate Gimnodium catenatum as the source of paralytic shellfish toxins in Tasmanian shellfish. Toxicon. 1987;25:1105–1111. doi: 10.1016/0041-0101(87)90267-4. [PubMed] [Cross Ref]
  • Mahmood NA, Carmichael WW. Paralytic shellfish poisons produced by the freshwater cyanobacterium Aphanizomenon flos-aquae NH-5. Toxicon. 1986;24:175–186. doi: 10.1016/0041-0101(86)90120-0. [PubMed] [Cross Ref]
  • Humpage AR, Rositano J, Bretag AH, Brown R, Baker PD, Nicholson BC, Steffensen DA. Paralytic shellfish poisons from Australian cyanobacterial blooms. Australian Journal of Marine & Freshwater Research. 1994;45:761–771. doi: 10.1071/MF9940761. [Cross Ref]
  • Carmichael WW, Evans WR, Yin QQ, Bell P, Moczydlowski E. Evidence for paralytic shellfish poisons in the freshwater cyanobacterium Lyngbya wollei (Farlow ex Gomont) comb. nov. Applied & Environmental Microbiology. 1997;63:3104–3110. [PMC free article] [PubMed]
  • Lagos N, Onodera H, Zagatto PA, Andrinolo D, Azevedo S, Oshima Y. The first evidence of paralytic shellfish toxins in the freshwater cyanobacterium Cylindrospermopsis raciborskii, isolated from Brazil. Toxicon. 1999;37:1359–1373. doi: 10.1016/S0041-0101(99)00080-X. [PubMed] [Cross Ref]
  • Sawyer PJ, Gentile JH, Sasner JJJ. Demonstration of a toxin from Aphanisomenon flos-aquae (L.) Ralfs. Canadian Journal of Microbiology. 1968;14:1199–1204. [PubMed]
  • Li R, Carmichael WW, Liu Y, Watanabe MM. Taxonomic re-evaluation of Aphanizomenone flos-aquae NH-5 based on morphology and 16S rRNA gene sequences. Hydrobiologia. 2000;438:99–105. doi: 10.1023/A:1004166029866. [Cross Ref]
  • Li R, Carmichael WW. Morphological and 16S rRNA gene evidence for reclassification of the paralytic shellfish toxin producing Aphanizomenon flos-aquae LMECYA 31 as Aphanizomenon issatschenkoi (Cyanophyceae) Journal of Phycology. 2003;39:814–818. doi: 10.1046/j.1529-8817.2003.02199.x. [Cross Ref]
  • Beltran EC, Neilan BA. Geographical Segregation of the Neurotoxin-Producing Cyanobacterium Anabaena circinalis. Applied & Environmental Microbiology. 2000;66:4468–4474. doi: 10.1128/AEM.66.10.4468-4474.2000. [PMC free article] [PubMed] [Cross Ref]
  • Bowling L, Baker P. Major cyanobacterial bloom in the Barwon-Darling River, Australia, in and underlying limnological conditions. Marine & Freshwater Research. 1991;47:643–657. doi: 10.1071/MF9960643. [Cross Ref]
  • Bowling L. The cyanobacterial (blue-green algae) bloom in the Darling/Barwon river system, November-December 1991. NSW Department of Water Resources, Technical Services Division, Report No 92074. 1992. p. 49. ISBN 0730578925.
  • Baker PD, Humpage AR. Toxicity associated with commonly occurring cyanobacteria in surface waters of the Murray-Darling Basin. Australian Journal of Marine & Freshwater Research. 1994;45:773–786. doi: 10.1071/MF9940773. [Cross Ref]
  • Llewellyn LE, Negri AP, Doyle J, Baker PD, Beltran EC, Neilan BA. Radioreceptor assays for sensitive detection and quantitation of saxitoxin and its analogues from strains of the freshwater cyanobacterium, Anabaena circinalis. Environmental Science & Technology. 2001;35:1445–1451. doi: 10.1021/es001575z. [PubMed] [Cross Ref]
  • Kellmann R, Mihali TK, Jeon YJ, Pickford R, Pomati F, Neilan BA. Biosynthetic Intermediate Analysis and Functional Homology Reveal a Putative Saxitoxin Gene Cluster in Cyanobacteria. Applied & Environmental Microbiology. 2008;74:4044–4053. doi: 10.1128/AEM.00353-08. [PMC free article] [PubMed] [Cross Ref]
  • Kellmann R, Mihali TK, Neilan BA. Identification of a saxitoxin biosynthesis gene with a history of frequent horizontal gene transfers. Journal of Molecular Evolution. 2008;67:526–538. doi: 10.1007/s00239-008-9169-2. [PubMed] [Cross Ref]
  • Moffitt MC, Neilan BA. Characterization of the nodularin synthetase gene cluster and proposed theory of the evolution of cyanobacterial hepatotoxins. Applied & Environmental Microbiology. 2004;70:6353–6362. doi: 10.1128/AEM.70.11.6353-6362.2004. [PMC free article] [PubMed] [Cross Ref]
  • Yoshida T, Sako Y, Uchida A, Kakutani T, Arakawa O, Noguchi T, Ishida Y. Purification and characterization of sulfotransferase specific to O-22 of 11-hydroxy saxitoxin from the toxic dinoflagellate Gymnodinium catenatum (dinophyceae) Fisheries Science. 2002;68:634–642. doi: 10.1046/j.1444-2906.2002.00471.x. [Cross Ref]
  • Sako Y, Yoshida T, Uchida A, Arakawa O, Noguchi T, Ishida Y. Purification and characterization of a sulfotransferase specific to N-21 of saxitoxin and gonyautoxin 2+3 from the toxic dinoflagellate Gymnodinium catenatum (Dinophyceae) Journal of Phycology. 2001;37:1044–1051. doi: 10.1046/j.1529-8817.2001.00119.x. [Cross Ref]
  • Mahillon J, Chandler M. Insertion sequences. Microbioogy and Molecular Biology Reviews. 1998;62:725–774. 1092-2172/98/ [PMC free article] [PubMed]
  • Rantala A, Fewer DP, Hisbergues M, Rouhiainen L, Vaitomaa J, Börner T, Sivonen K. Phylogenetic evidence for the early evolution of microcystin synthesis. Proceedings of the National Academy of Sciences. 2004;101:568–573. doi: 10.1073/pnas.0304489101. [PMC free article] [PubMed] [Cross Ref]
  • Mihali TK, Kellmann R, Muenchhoff J, Barrow KD, Neilan BA. Characterization of the gene cluster responsible for cylindrospermopsin biosynthesis. Applied & Environmental Microbiology. 2008;74:716–722. doi: 10.1128/AEM.01988-07. [PMC free article] [PubMed] [Cross Ref]
  • 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. Chemistry & Biology. 2000;7:753–764. doi: 10.1016/S1074-5521(00)00021-1. [PubMed] [Cross Ref]
  • Thompson AS, Rhodes JC, Pettman I, eds . Culture collection of algae and protozoa. catalogue of strains. 5. Natural Environment Research Council, Freshwater Biological Association; 1988. p. 164.
  • Chenna R, Sugawara H, Koike T, Lopez R, Gibson TJ, Higgins DG, Thompson JD. Multiple sequence alignment with the Clustal series of programs. Nucleic Acids Research. 2003;31:3497–3500. doi: 10.1093/nar/gkg500. [PMC free article] [PubMed] [Cross Ref]
  • Siebert PD, Chenchik A, Kellogg DE, Lukyanov SA. An improved PCR method for walking in uncloned genomic DNA. Nucleic Acids Research. 1995;23:1087–1088. doi: 10.1093/nar/23.6.1087. [PMC free article] [PubMed] [Cross Ref]
  • Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research. 1997;25:4876–4882. doi: 10.1093/nar/25.24.4876. [PMC free article] [PubMed] [Cross Ref]
  • Guindon S, Gascuel O. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology. 2003;52:696–704. doi: 10.1080/10635150390235520. [PubMed] [Cross Ref]
  • Posada D, Crandall KA. MODELTEST: testing the model of DNA substitution. Bioinformatics. 1998;14:817–818. doi: 10.1093/bioinformatics/14.9.817. [PubMed] [Cross Ref]

Articles from BMC Biochemistry are provided here courtesy of BioMed Central
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...