Isolation and characterisation of antimicrobial compounds synthesised by Microcystis sp.

Abstract

Cyanobacterial secondary metabolites, often identified as toxins such as microcystin, have also demonstrated biological functions including inhibition of bacterial and viral growth. In this study, 10 cyanobacterial strains were isolated from field sites around Adelaide and laboratory cultures and assessed for bioactivity against bacterial, viral and fungal pathogens. A comprehensive literature search identified a number of screening assays employed by research groups to identify cyanobacterial strains with biological activity. Within the review, methods to optimise extraction of the compounds were also noted. Combinations of extraction methods, solvents and assay procedures were investigated to optimise the success of this phase of the study. Bioactivity was confirmed by development of agar disc diffusion and microtitre plate assays to analyse cyanobacterial biomass extracts. Result of the assays indicated a methanolic extract of one species, Microcystis flos-aquae (Wittr Elenkin), inhibited growth of bacterial cells and viral infectivity and was selected for further analysis. The bioactive compound was isolated by HPLC and mass spectrometric analysis. Separation of the bioactive extract into component peaks indicated only one that was likely to represent the metabolite of interest, at a retention time of approximately 18 min. A second profile was constructed of a methanolic extract of the same species in a later growth stage that did not inhibit growth of either the bacterial or viral test organisms. Comparison of the profiles exposed the absence of the peak at 18 min retention time in the second profile. Accumulation of the fraction was conducted using a semi-preparative HPLC column for analysis by mass spectrometry. A sample of the isolated peptide was submitted to Proteomics International, a subsidiary of Murdoch University, WA, for identification and structural characterisation. Proteomics International analysed the data by electronspray ionisation time of flight mass spectrometry (LC/MS/TOF) followed by LC. De novo sequence analysis of the data was carried out using Analyst QS software; however, PI was unable to provide a readily interpretable, continuous amino acid sequence, despite their admission that some gaps in the fragmentation ladder corresponded to known amino acids. Interpretation of the data generated by Proteomics International by a research chemist within the University of Adelaide proposed the following amino acid sequence and subsequent structure for the compound: [Figures omitted] Proposed (a) amino acid sequence and (b) structure for the bioactive compound isolated from non-toxic M. flos-aquae. Comparison of the proposed sequence with those contained in peptide databases was unable to classify the compound (B Neilan, personal communication, April 2008), suggesting the bioactive metabolite is perhaps previously undetected and therefore may be considered a novel compound, or has undergone a modification and is thus a variant of a known compound. Taxonomic classification of the strain used during this study was completed by PCR amplification of 16S ribosomal RNA, using primers from alternative cyanobacterial sources. The sequence was classified in the following taxonomic hierarchy (with 100% assignment detail, for a confidence threshold of 95%): Domain: Bacteria Phylum Cyanobacteria Class Cyanobacteria Family Family 1.1 Genus Microcystis This classification confirms that the species investigated during this research is of the genus Microcystis. Synthesis of cyanobacterial metabolites is generally accepted to be a result of nonribosomal synthetic pathways. The presence of non-ribosomal peptide synthetase and polyketide synthetase genes in Microcystis flos-aquae was confirmed by PCR amplification using degenerate primers from other cyanobacterial sources. Analysis of sequence data identified the presence of an NRPS gene demonstrating significant similarity (98%) to the NRPS cyanopeptolin gene of Microcystis sp. However, the PKS (polyketide) gene identified verified only a 63% similarity to a known sequence, that of the PKS (mcyG) gene of M. aeruginosa PCC 7806 (Koch). Results of the molecular investigation imply this compound may belong within the cyanopeptolin family. Researchers have speculated that the majority of cyanobacteria possess genes for production of toxins, though in many instances the gene cluster may be incomplete or one or more genes may be absent or mutated. The presence of microcystin genes was confirmed by PCR amplification using primers from previously characterised cyanobacterial genes. Analysis of the sequence data identified the presence of several mcy genes generally found in toxic strains of cyanobacteria noted for synthesis of the toxin microcystin. The DNA sequences show significant similarity to the mcyA, mcyC, mcyD and mcyE genes described for Microcystis sp. and Microcystis aeruginosa PCC 7806. However, analysis of the sequence data for the mcyB gene revealed that this gene was not present. Further PCR amplification of the region between mcyA and mcyC using the reverse complements of the original primers indicated that a sequence was present that may have been a truncated variant of mcyB or another gene entirely. Time constraints prohibited submission of this region for sequence analysis. The primary objective of this research project was to screen a field strain of cyanobacteria for synthesis of biologically active secondary metabolites, and to isolate those compounds using a combination of analytical chemistry and molecular biotechnology. This study forms part of a collaborative project between the University of Adelaide, South Australian Research and Development Institute (Aquatic Sciences) and the Environmental Biotechnology Cooperative Research Centre, entitled “P6: Commercial scale integrated biosystems for organic waste and wastewater treatment for the livestock and food processing industries”.