Tipping points, at which complex systems can shift abruptly from one state to another, are notoriously difficult to predict. Theory proposes that early warning signals may be based on the phenomenon that recovery rates from small perturbations should tend to zero when approaching a tipping point; however, evidence that this happens in living systems is lacking. Here we test such 'critical slowing down' using a microcosm in which photo-inhibition drives a cyanobacterial population to a classical tipping point when a critical light level is exceeded. We show that over a large range of conditions, recovery from small perturbations becomes slower as the system comes closer to the critical point. In addition, autocorrelation in the subtle fluctuations of the system's state rose towards the tipping point, supporting the idea that this metric can be used as an indirect indicator of slowing down. Although stochasticity prohibits prediction of the timing of critical transitions, our results suggest that indicators of slowing down may be used to rank complex systems on a broad scale from resilient to fragile.

Summary 1. The hypothesis that cyanobacteria have higher optimum growth temperatures and higher growth rates at the optimum as compared to chlorophytes was tested by running a controlled experiment with eight cyanobacteria species and eight chlorophyte species at six different temperatures (20–35 °C) and by performing a literature survey. 2. In the experiment, all organisms except the chlorophyte Monoraphidium minutum grew well up to 35 °C. The chlorophyte Chlamydomonas reinhardtii was the fastest‐growing organism over the entire temperature range (20–35 °C). 3. Mean optimum growth temperatures were similar for cyanobacteria (29.2 °C) and chlorophytes (29.2 °C). These results are concordant with published data, yielding slightly higher mean optimum growth temperatures for cyanobacteria (27.2 °C) than for chlorophytes (26.3 °C). 4. Mean growth rates of cyanobacteria at 20 °C (0.42 day−1) were significantly lower than those of chlorophytes at 20 °C (0.62 day−1). However, at all other temperatures, there were no differences between mean growth rates of cyanobacteria and chlorophytes. 5. Mean growth rates at the optimum temperature were similar for cyanobacteria (0.92 day−1) and chlorophytes (0.96 day−1). However, analysis of published data revealed that growth rates of cyanobacteria (0.65 day−1) were significantly lower than those of chlorophytes (0.93 day−1) at their optimum temperatures. 6. Although climate warming will probably lead to an intensification of cyanobacterial blooms, our results indicate that this might not be as a result of higher growth rates of cyanobacteria compared with their chlorophyte competitors. The competitive advantage of cyanobacteria can more likely be attributed to their ability to migrate vertically and prevent sedimentation in warmer and more strongly stratified waters and to their resistance to grazing, especially when warming reduces zooplankton body size.

The cyanobacterial neurotoxin β-N-methylamino-L-alanine (BMAA) has been considered a serious health threat because of its putative role in multiple neurodegenerative diseases. First reports on BMAA concentrations in cyanobacteria were alarming: nearly all cyanobacteria were assumed to contain high BMAA concentrations, implying ubiquitous exposure. Recent studies however question this presence of high BMAA concentrations in cyanobacteria. To assess the real risk of BMAA to human health, this discrepancy must be resolved. We therefore tested whether the differences found could be caused by the analytical methods used in different studies. Eight cyanobacterial samples and two control samples were analyzed by three commonly used methods: HPLC-FLD analysis and LC-MS/MS analysis of both derivatized and underivatized samples. In line with published results, HPLC-FLD detected relatively high BMAA concentrations in some cyanobacterial samples, while both LC-MS/MS methods only detected BMAA in the positive control (cycad seed sarcotesta). Because we could eliminate the use of different samples and treatments as causal factors, we demonstrate that the observed differences were caused by the analytical methods. We conclude that HPLC-FLD overestimated BMAA concentrations in some cyanobacterial samples due to its low selectivity and propose that BMAA might be present in (some) cyanobacteria, but in the low µg/g or ng/g range instead of the high µg/g range as sometimes reported before. We therefore recommend to use only selective and sensitive analytical methods like LC-MS/MS for BMAA analysis. Although possibly present in low concentrations in cyanobacteria, BMAA can still form a health risk. Recent evidence on BMAA accumulation in aquatic food chains suggests human exposure through consumption of fish and shellfish which expectedly exceeds exposure through cyanobacteria.

The neurotoxin β-N-methylamino-l-alanine (BMAA) is suspected to play a role in the neurological diseases amyotrophic lateral sclerosis, Alzheimer’s disease, and Parkinson’s disease. BMAA production by cyanobacteria has been reported and contact with cyanobacteria infested waters or consumption of aquatic organisms are possible pathways to human exposure. However, there is little consensus regarding whether BMAA is present in cyanobacteria or not, and if so, at what concentrations. The aim of this review is to indicate the current state of knowledge on the presence of BMAA in aquatic ecosystems. Some studies have convincingly shown that BMAA can be present in aquatic samples at the µg/g dry weight level, which is around the detection limit of some equally credible studies in which no BMAA was detected. However, for the majority of the reviewed articles, it was unclear whether BMAA was correctly identified, either because inadequate analytical methods were used, or because poor reporting of analyses made it impossible to verify the results. Poor analysis, reporting and prolific errors have shaken the foundations of BMAA research. First steps towards estimation of human BMAA exposure are to develop and use selective, inter-laboratory validated methods and to correctly report the analytical work.

We aimed to determine concentrations of the neurotoxic amino acids beta-N-methylamino-L-alanine (BMAA) and alpha-,gamma-diaminobutyric acid (DAB) in mixed species scum material from Dutch urban waters that suffer from cyanobacterial blooms. BMAA and DAB were analysed in scum material without derivatization by LC-MSMS (liquid chromatography coupled to tandem mass spectrometry) using hydrophilic interaction chromatography (HILIC). Our method showed high selectivity, good recovery of added compounds after sample extraction (86% for BMAA and 85% for DAB), acceptable recovery after sample hydrolysation (70% for BMAA and 56% for DAB) and acceptable precision. BMAA and DAB could be detected at an injected amount of 0.34 pmol. Free BMAA was detected in nine of the 21 sampled locations with a maximum concentration of 42 microg/g DW. Free DAB was detected in two locations with a maximum concentration of 4 microg/g DW. No protein-associated forms were detected. This study is the first to detect underivatized BMAA in cyanobacterial scum material using LC-MSMS. Ubiquity of BMAA in cyanobacteria scums of Dutch urban waters could not be confirmed, where BMAA and DAB concentrations were relatively low; however, co-occurrence with other cyanobacterial neurotoxins might pose a serious health risk including chronic effects from low-level doses.