Analysis of microcystins and microcystin genes in 60–170-year-old dried herbarium specimens of cyanobacteria

Abstract

Subsamples of cyanobacteria which had been stored as herbarium specimens at the Natural History Museum, London (BM), were analysed for the presence of microcystins and for genes involved in their biosynthesis using PCR. The samples had been collected worldwide between 1839 and 1950 and stored as dried specimens on paper, or between thin sheets of mica, under ambient conditions in the dark. Specimens for analysis were selected on the assumption that the chosen genera or species identified at the time of collection would have had a high potential for microcystin production based on current knowledge of the phylogeny of microcystin biosynthesis among the cyanobacteria. Of the 30 specimens analysed, 46% were positive for microcystins by high performance liquid chromatography with photodiode array detection and 83% were positive for the toxins according to microcystin immunoassay. Ninety seven percent of the specimens had ions which corresponded with known microcystins according to matrix-assisted laser desorption ionization time of flight mass spectrometry and 17% of the samples showed positive PCR bands for the mcyD gene for microcystin synthetase. These results demonstrate the potential for long-term survival of microcystins, and to a lesser extent of microcystin synthetase genes in herbarium specimens of dried cyanobacteria. They also offer the possibility for comparative studies on cyanotoxin occurrence at identifiable sites then and now, and of the use of archived cyanobacterial specimens in retrospective studies in the case of ecotoxicological investigations at the sampling locations.

Introduction

The association of cyanobacterial blooms with animal and bird mortalities, for example in Denmark and North-Western Poland (then West Prussia), was reported in the scientific literature in the 19th century (Codd et al., 2005a). The causative nature of this association was confirmed by the experimental dosing of healthy sheep with hepatotoxic Nodularia spumigena scum after mass mortalities of sheep, horses and cattle along the shores of Lake Alexandrina in South Australia (Francis, 1878). Since that time, animal deaths due to the ingestion of planktonic and benthic masses of cyanobacteria have continued and cyanotoxin-associated human illnesses and deaths have been reported (Codd et al., 1999). It is increasingly apparent that cyanotoxin production is a common characteristic of extant cyanobacteria throughout the world (Codd et al., 2005b, Pearson et al., 2010).

Phylogenetic analysis of genes for the biosynthesis of the heptapeptide microcystins and the closely related pentapeptide nodularins has clearly indicated that the capacity for production of these cyanotoxins arose early in cyanobacterial evolution (Rantala et al., 2004). Genes for microcystin biosynthesis (by microcystin synthetase) are inferred to have existed in the last common cyanobacterial ancestor of a diverse range of the cyanobacteria which produce the toxins today, e.g. members of the genera Microcystis, Anabaena and Planktothrix. Nodularin synthetase, responsible for nodularin biosynthesis by Nodularia strains, appears to have arisen later in evolution by modification of microcystin synthetase. The processes of horizontal gene transfer and cyanotoxin gene loss appear to have resulted in the patchy occurrence of microcystin and nodularin biosynthesis in extant cyanobacteria (Rantala et al., 2004).

Collections of cyanobacteria, stored as dried herbarium specimens, present a valuable, but largely overlooked, potential resource to investigate the occurrence of cyanotoxin genes in materials of historic and geographic interest. Palinska et al. (2006) have demonstrated the feasibility of this approach by their detection and characterisation of cyanobacterial 16S rRNA gene fragments in herbarium collections that were deposited in European museums over 100 years ago, although extending this research to cyanotoxin genes does not appear to have been attempted so far.

Herbarium specimens also present a potentially valuable resource for direct analysis of the cyanotoxins. The feasibility of this approach is supported by the low minimum detection levels achievable in cyanotoxin analysis (sub-nanogram), the high degrees of specificity increasingly available via multiple cyanotoxin analysis methods, and the relatively stable nature of microcystins (Codd et al., 2001, Metcalf and Codd, 2003, Spoof, 2005). Here we report on the analysis of 30 samples taken from material that had been held in the herbarium at BM for periods of between 60 and about 170 years. The samples were selected for analysis as having a very high probability of microcystin content at the time of collection, based on their taxonomic identification. Selecting cyanobacteria likely to be microcystin-producers afforded the greatest chance of providing an indication as to whether microcystins, and the genes associated with their synthesis, can survive 100+ years in herbaria. This would allow future studies with herbarium specimens to be undertaken with the possibility of retrospective studies at sites with cyanobacterial blooms.