In this study, we identified and sequenced the anatoxin-a synthetase gene cluster of a planktonic cyanobacterial strain, Anabaena sp. 37. The genes were closely related to anatoxin-a synthetase genes previously recognized in Oscillatoria sp. PCC 6506, although the organization of the gene clusters differed. The Anabaena gene cluster sequence contributes to knowledge of the biosynthesis of this important neurotoxin and opens possibilities for future research avenues, e.g., to reveal the evolutionary history of anatoxin-a production. This sequence and the additional sequences of the anaC genes of several strains of the cyanobacteria Anabaena, Oscillatoria, and Aphanizomenon were used to design methods for the detection (PCR with general primers) and identification (PCR with genus-specific primers, RFLP) of the anaC gene, and thus of potential anatoxin-a and homoanatoxin-a producers, in environmental samples. The molecular tools developed here can be used to monitor the potential anatoxin producers of the genera Anabaena, Aphanizomenon, and Oscillatoria in ecosystems and can help to rapidly estimate the risks of their occurrence for water users.
The molecular methods developed in this study to detect potential anatoxin-a or homoanatoxin-a producers were based on recognition of the anaC gene, which encodes AnaC protein, thought to be responsible for the first step in anatoxin-a synthesis, proline adenylation (). Sequencing of the 860-bp anaC gene region revealed unexpectedly that Oscillatoria strains were divided into two groups—one with sequences identical to that of Oscillatoria sp. PCC 6506 and the other with approximately 92% sequence identity to Oscillatoria sp. PCC 6506—while variation between Anabaena strains was almost nonexistent. Sequence variation, even when it does not affect the protein function, challenges the design and use of molecular detection methods, which rely on the tight connection between a certain sequence variant and the identity of the producer organism. In our study, the genus-specific PCR assays with both Anabaena anaC- and Oscillatoria anaC-specific primers performed highly specifically and detected the target gene only in Anabaena or Oscillatoria strains, respectively, except for Oscillatoria strains PCC 10608 and PCC 10702. These strains were previously reported to produce homoanatoxin-a (), but toxin production could not be detected in this study. The lack of amplification with any or most of the anaC-targeted or anaF-targeted (data not shown) primers suggests that the strains have undergone extensive genetic rearrangements that have caused deletion of some or all of the biosynthetic genes. Deletions and insertions in the microcystin (, , , ) and cylindrospermopsin () biosynthetic gene clusters have been determined to cause a lack of toxin production. Smaller changes, e.g., point mutations in gene or regulatory regions (, , ), have been suspected to cause a lack of toxin production in cases where all the biosynthetic genes were shown to be intact. This could be the case with Anabaena sp. strain 86 and Aphanizomenon sp. strain 3, which had both the anaC and anaF (data not shown) genes but produced no anatoxin-a according to LC-MS results. These strains have been kept in culture continuously since 1986 and were shown to produce anatoxin-a at that time (, ). It is hypothesized that at some point during the 25 years of culture, they underwent spontaneous mutations inactivating the gene clusters. However, a more detailed investigation of the gene clusters or their remnants is needed in order to elucidate what has happened in the ana gene clusters of these strains.
Cyanobacterial mass occurrences are common in fresh and brackish waters. They pose a threat to water users due to toxins frequently produced by the cyanobacterial species present. Anatoxin-a and homoanatoxin-a are neurotoxins synthesized by various cyanobacteria, e.g., Anabaena, Oscillatoria, and Aphanizomenon. The biosynthesis of these toxins and the genes involved in anatoxin production were recently described for Oscillatoria sp. strain PCC 6506 (A. Méjean et al., J. Am. Chem. Soc. 131:7512-7513, 2009). In this study, we identified the anatoxin synthetase gene cluster (anaA to anaG and orf1; 29 kb) in Anabaena sp. strain 37. The gene (81.6% to 89.2%) and amino acid (78.8% to 86.9%) sequences were highly similar to those of Oscillatoria sp. PCC 6506, while the organization of the genes differed. Molecular detection methods for potential anatoxin-a and homoanatoxin-a producers of the genera Anabaena, Aphanizomenon, and Oscillatoria were developed by designing primers to recognize the anaC gene. Anabaena and Oscillatoria anaC genes were specifically identified in several cyanobacterial strains by PCR. Restriction fragment length polymorphism (RFLP) analysis of the anaC amplicons enabled simultaneous identification of three producer genera: Anabaena, Oscillatoria, and Aphanizomenon. The molecular methods developed in this study revealed the presence of both Anabaena and Oscillatoria as potential anatoxin producers in Finnish fresh waters and the Baltic Sea; they could be applied for surveys of these neurotoxin producers in other aquatic environments.
Not all cyanobacterial strains produce toxins. However, the toxin-producing strains cannot be distinguished from the non-toxin-producing strains by traditional light microscopy, commonly used to monitor water bodies. An alternative for the differentiation of potentially toxic strains from nontoxic strains is to use molecular methods to detect the presence of toxin biosynthetic genes (, , ). Such methods are already available and could be used for the detection and identification of potential microcystin and nodularin producers present in environmental samples, e.g., blooms (, ). The elucidation of the biosynthetic gene clusters for cylindrospermopsin (, , ) and saxitoxins (, ) has enabled the development of molecular detection methods for the producers of these toxins (, ).
Once in the nucleus, the steroid appears to enhance transcription of specific genes. The resulting mRNA is processed and sent out of the nucleus, resulting in increased protein synthesis.
We identified anatoxin-a synthetase (ana) genes required for anatoxin-a production in the genomic sequence of Anabaena sp. 37. The genes resembled those identified in Oscillatoria sp. strain PCC 6506 (); the ana gene content was the same, with high sequence identity. In addition, the protein functions and domains predicted agreed with those of Oscillatoria sp. PCC 6506 proteins, further confirming the identification of the sequences as anatoxin-a synthetase genes. The most prominent difference between the gene clusters was the organization of the genes. In Anabaena sp. 37, orf1 and anaA are located on the opposite strand downstream from the other biosynthetic genes (anaB to anaG), while in Oscillatoria sp. PCC 6506, orf1 and anaA lie upstream and are transcribed in the same direction as the other genes. Similar differences in gene order have been recognized in biosynthetic gene clusters of other cyanobacterial toxins: microcystins (), cylindrospermopsin (, ), and saxitoxins (). The DNA (81.6 to 89.2%) and encoded amino acid (78.8 to 86.9%) sequence similarities between the Anabaena sp. 37 and Oscillatoria sp. PCC 6506 clusters were at the same level as those between the biosynthetic gene clusters of different microcystin () and saxitoxin producers ().
The ongoing Anabaena sp. 37 genome sequencing project was started by use of the 454 method at the Institute for Biomedical Technologies (National Research Council, Milan, Italy). A single-stranded DNA (ssDNA) library and a paired-end library were prepared according to the Roche-454 library preparation manual. Shotgun and paired-end sequencing were performed using the Titanium version of the Roche-454 GS FLX system. From the sequences produced, an initial set of open reading frames (ORF) for anatoxin-a synthetase genes was predicted by Glimmer, version 3.02 () (), on the basis of similarity with the corresponding genes of Oscillatoria sp. PCC 6506 (). The Needle program of the EMBOSS package (the European Molecular Biology Open Software Suite) () was used for comparison of Anabaena sp. 37 and Oscillatoria sp. PCC 6506 gene clusters with default parameters. Sequence similarity searches in databases were carried out with BLAST () (). The InterProScan Sequence Search () and the PKS/NRPS (polyketide synthase/nonribosomal peptide synthetase) Analysis Web-site () were used to predict the protein functions and identify the domain structures.
Recently, biosynthetic genes responsible for anatoxin-a production were reported in a benthic Oscillatoria strain, PCC 6506 (, , ). Subsequently, methods for the detection of the anaF genes of the anatoxin-producing Oscillatoria (), Phormidium (), and Aphanizomenon () strains were designed. Our aim was to identify the anatoxin-a synthetase (ana) gene cluster in the genus Anabaena by sequencing the entire cluster in Anabaena sp. strain 37, a planktonic strain originally isolated from a cyanobacterial bloom in Lake Sääskjärvi that caused cattle deaths in the summer of 1985 (). Comparison of the ana gene clusters of Anabaena sp. strain 37 and Oscillatoria sp. strain PCC 6506 and the anaC gene sequences from additional anatoxin-a-producing strains of the genera Anabaena, Aphanizomenon, and Oscillatoria enabled us to design the primers needed for molecular detection methods. The primers were applied in PCR to detect these three producer genera at the same time (general primers) or to identify either Anabaena or Oscillatoria anaC gene variants (genus-specific primers). In addition, restriction fragment length polymorphism (RFLP) analysis of the general PCR products allowed simultaneous identification of the three producer genera studied. The methods were applied to cyanobacterial strains as well as to environmental DNA samples so as to establish molecular tools for the detection of these potential anatoxin producers.
Anabolic steroids work by binding with the cytoplasmic (free within the cell) androgen receptor. Like all steroids, the steroid-receptor complex has a strong affinity for the nucleus. The complex is translocated into the nucleus and binds to DNA. It is also possible that the steroid and receptor dissociate in the nucleus and act on DNA separately. The rate limiting factor in this process appears to be the cytoplasmic concentration of the receptor, rather than the concentration of the steroid or translocation of the complex. Testosterone also appears to inhibit the catabolic (protein degrading) pathway associated with glucocorticoids, but it is unclear whether this is related to an interaction between testosterone and glucocorticoid or the nuclear behavior of the steroid-receptor complex.
The anatoxin-a biosynthesis genes (ana) of Anabaena sp. 37 (GenBank accession number ) were identified on the basis of comparison to the ana genes of Oscillatoria sp. PCC 6506 (GenBank accession number ) (). The ana genes and proteins of Anabaena sp. 37 were very similar, with ≥81.6% and 78.8% identity, respectively, to those of Oscillatoria sp. PCC 6506 (). Accordingly, the protein functions predicted and the functional domains identified in the polyketide synthases () were the same as those of Oscillatoria sp. PCC 6506. The main difference between the two ana gene clusters was seen in the organization of the genes (). In Anabaena sp. 37, the biosynthetic genes were found in two clusters spanning approximately 29 kb. The anaB to anaG genes formed one cluster of 20.3 kb, while anaA and the putative cyclase gene, orf1 (1.7 kb), were separated from the main cluster by a 6.9-kb section of DNA and were transcribed in the opposite direction (). The biosynthetic genes were putatively organized in four or five operons. The −10 (Pribnow box) and −35 sequences were recognized upstream of the gene regions anaB to anaD, anaEF, orf1, and anaA, but not before anaG. Although RNA polymerase recognition sequences were not identified, a long gap (288 bp) () between anaF and anaG suggested that anaG forms its own operon. The 6.9-kb insert contained genes coding for four hypothetical proteins, an N-acetyltransferase, a Rieske domain-containing protein, an oxidoreductase, and a putative multidrug exporter, which were absent from the vicinity of the anatoxin-a synthetase gene cluster of Oscillatoria sp. PCC 6506.