Characterising the benthic Phormidium autumnale-dominated biofilm community and anatoxin production throughout biofilm succession

There has been an increase in the prevalence and intensity of Phormidium autumnale-dominated benthic blooms in New Zealand over the last decade. This species produces the potent neurotoxins Anatoxin-a, Homoanatoxin-a and their derivatives, and consumption of P. autumnale biofilms has led to over 70 dog deaths since 2005. The mechanisms regulating the dominance and toxicity of P. autumnale are still unclear, as these blooms can reach high biomass in low nutrient conditions. Benthic biofilms are composed of multiple taxa and usually harbor a complex community of bacteria and other microbes, which can change over time and interact to facilitate biofilm development and metabolic processing. Prior to this thesis, the microbial composition of P. autumnale-dominated biofilms was unknown. This study provides insights into the relationships of this neurotoxic cyanobacterium with microbial components of the biofilm community. Benthic biofilms were sampled every two to four days for 32 days from three sites in the Hutt River (Wellington) following a high flow event. A combination of microscopy and molecular techniques (bacterial ARISA and Illumina™ sequencing) were used to identify the micro-algal and bacterial components of the biofilm throughout its development. Variation in total anatoxin production was measured using LC-MS and changes in toxic P. autumnale cell numbers were quantified using QPCR. A suite of environmental variables (point velocity, depth, flow, conductivity, temperature and nutrients) were also monitored throughout the study period. Three distinct phases of microbial succession were identified (early, mid and late) using non-metric multidimensional cluster analyses. The micro-algal community composition (including P. autumnale) shifted from early to mid-phase ca. 16 days after the flushing flow and from mid to late phase at ca. day 30. The ARISA and Illumina™ sequencing showed the bacterial community shifts occurred ca. 4 and 9 days before the respective micro-algal community shifts. These analyses indicate a close coupling of the micro-algal and bacterial communities and may suggest bacterial driven succession. However, bacteria are likely to depend on micro-algal by-products for nutrition from the mid-phases onward and assessment of the metabolic processes occurring within the biofilms is needed to clarify this. Phormidium autumnale was dominant in the biofilm from an early stage in development and grew exponentially despite an influx of diatoms at day 20. None of the environmental parameters measured could explain the temporal variation in micro-algal and bacterial communities, which suggested that intrinsic rather than extrinsic factors were more important in regulating succession. This further supports the hypothesis that biofilm microbes may facilitate P. autumnale dominance. There was a significant variation in anatoxins per cell over time (p = 0.034). Production of anatoxins was greatest in the mid-phase of succession (208 fg cell⁻¹), coinciding with an increase in diatom biomass, which could implicate anatoxins as allelopathic chemicals that alleviate the effects of competition on P. autumnale. Changes in proportions of the different anatoxin variants produced over time also aligned with the three successional phases in both the micro-algal and bacterial communities, providing further evidence of a relationship between anatoxin production and microbial biofilm components. Bacterial taxa of the Alphaproteobacteria were dominant within the early bacterial community, but were surpassed by the Betaproteobacteria and Flavobacteria in mid and late phases. Bacterial genera involved in exopolysaccharide production, alkaline phosphatase activity and biopolymer degradation were identified. These attributes are important in the formation, maintenance and break-down of biofilms and therefore strengthen the likelihood of linkages between the micro-algal and bacterial community. Further investigations into functional roles of the biofilm components are needed to infer relationships between P. autumnale and the bacterial community. A clear pattern of microbial succession is described here and linkages between the micro-algal and bacterial communities are evident. Future work should focus on the functional attributes of microbes occurring at different stages of succession to further understand how P. autumnale dominates these benthic communities.