Relation Between Microcystis aeruginosa and Aphanizomenon flos-aquae

Unlike A. flos-aquae, M. aeruginosa does not fix nitrogen, which, at times, appears to be a growth-limiting nutrient for phytoplankton in Upper Klamath Lake, given the dominance of diazotrophic cyanobacteria, such as A. flos-aquae (Reuter and others, 1993). However, members of the Microcystis genus are highly adaptable and frequently become the dominant organism in eutrophic systems, including reservoirs of the Klamath River (Jacoby and Kann, 2007; Moisansder and others, 2009) and other Oregon lakes (Barbiero and Kann, 1994). Microcystis sp. have a high affinity to absorb dissolved inorganic nitrogen (Takamura and others, 1987), can store large amounts of intracellular phosphorus as polyphosphate (Jacobson and Halmann, 1982), and can take up phosphorus directly from attached bacteria (Jiang and others, 2007). Microcystis sp. cells also can tolerate strong irradiance (Paerl and others, 1985), overwinter as vegetative cells (without sporulating; Preston and others, 1980), and, like A. flos-aquae, they can adjust their buoyancy to occupy the best position for receiving optimum light intensity for photosynthesis (Ibelings and others, 1991). Such characteristics contribute to the success of Microcystis sp. in temperate and tropical climates, but the ability of this microorganism to coexist in direct competition with A. flos-aquae is unknown because there have been few, if any, direct comparisons between their critical physiological parameters (Yamamoto, 2009). Recent in situ nutrient enrichment experiments in the Copco and Iron Gate Reservoirs on the Klamath River, California, showed that, during the summer, nitrogen was frequently the primary nutrient limiting growth of M. aeruginosa and microcystin concentration, although changes in per-cell toxin content or the ratio in abundances of toxic versus nontoxic strains could have contributed to the observed trends (Kann and Corum, 2009; Bozarth and others, 2010). The effects of nitrogen addition were clearest when biomass and overall microcystin concentrations were lowest, and, on several occasions, secondary phosphorus limitation was observed (cell abundance increased when phosphorus was added in combination with nitrogen; Moisander and others, 2009). 


The ecological relation between M. aeruginosa and A. flos-aquae, whether it is competitive, facilitative, or neutral, has implications for the management of Upper Klamath Lake. If the relation is competitive, the occurrence of toxigenic M. aeruginosa may increase if nutrient management (reduction of phosphorus inputs, the focus of most proposed plans) successfully limits the A. flos-aquae bloom. However, if the relation is facilitative, an overall decrease or elimination of A. flos-aquae also may eliminate M. aeruginosa, given that this species already occurs in low abundance here. These alternatives can not be evaluated definitively with the data collected in the current study, and the relation between A. flos-aquae and M. aeruginosa may be more complex and variable throughout the season, but the results of this work do permit the creation of testable hypotheses, primarily the hypothesis that toxigenic M. aeruginosa and microcystin occurrence are associated with the second of two A. flos-aquae-dominated blooms observed in most years. The most notable difference in lake conditions between 2007 and 2009 was the timing of the dominant bloom cycle and severity of the first bloom decline. In years with the highest microcystin concentrations, the first bloom declined sharply in July (mid-July in 2007 and late July in 2009) and was followed by a second bloom about 2 weeks later. In contrast, only one bloom was observed in 2008, which declined later and did not culminate until about the third week in August (Kann, 2010). This resulted in the highest concentrations of DIP and DIN occurring relatively late that year and, together with the patterns in dissolved nutrient concentrations and ratios observed other years, leads to an additional hypothesis that toxigenic M. aeruginosa growth and (or) microcystin occurrence is stimulated directly by the release of DIN during the major A. flos-aquae-dominated bloom decline but is dependent, overall, on the presence of phosphorus to regulate growth and decline of A. flos-aquae.


In Upper Klamath Lake, nitrogen fixation by A. flos-aquae early in the season, when concentrations of A. flos-aquae typically are highest, may facilitate the growth of toxigenic M. aeruginosa after the first major bloom decline by supplying new nitrogen to the system as A. flos-aquae cells lyse and decompose; this appears to be a stronger relation in years, such as 2009, with a well-defined, lakewide bloom cycle. However, because the A. flos-aquae­-dominated bloom cycle appears to be regulated more by changes in phosphorus availability (given that A. flos-aquae is able to fix N₂ when DIN is not available), support of M. aeruginosa growth by an increase in new nitrogen favors phosphorus availability indirectly as a more important factor overall for regulation of microcystin concentrations. This also helps to explain the observed seasonal patterns in microcystin and chlorophyll a concentrations and why correlations were significant between microcystin and total or dissolved phosphorus concentrations in 2009 and not significant between microcystin and nitrogen (TN or DIN) concentrations. The results of this study show that high concentrations of DIN and DIP followed the major lakewide bloom decline in 2009 and may have promoted growth of toxigenic M. aeruginosa (as indicated by the occurrence of microcystins) in concert with the development of a second large A. flos-aquae bloom. These patterns were also observed in previous analyses of cyanobacterial blooms in Clear Lake, California, where summer release of ammonia from the decomposition of the spring A. flos-aquae bloom and the presence of naturally abundant phosphorus stimulated Microcystis growth and recovery of the dominant bloom former, A. flos-aquae (Horne, 1975; Horne and Goldman, 1994). Similar to results of the Clear Lake study, the microcystin data presented here do not indicate growth of toxigenic cells during the first A. flos-aquae bloom. Results of monthly sampling collected in 2007 do not provide the same temporal resolution as 2009 data, but the highest microcystin concentrations measured that year occurred on August 1, about 2 weeks after the bloom decline and 2 weeks earlier than the highest concentrations measured in 2009. Results of alternate-week sampling in 2008 showed uniformly low concentrations relative to the other 2 years, which probably is related to the absence of a large, early A. flos-aquae-dominated bloom and the late (and less severe) bloom decline observed that year. 


In 2009, M. aeruginosa began to increase with the major decline in the A. flos-aquae bloom, was at very low concentrations prior to that decline, and continued to increase rapidly during the second A. flos-aquae bloom that followed the decline. Therefore, growth of toxigenic M. aeruginosa may be favored by the decrease in A. flos-aquae during the bloom decline (if these species compete directly) and by the increased availability of nutrients (both DIN and DIP) during this time, but the occurrence of toxigenic M. aeruginosa does not seem to be adversely affected by the return of the A. flos-aquae bloom later in the season. M. aeruginosa appears to continue growing and (or) producing microcystins while co-existing with A. flos-aquae during the second bloom. As such, the relation between these species appears to have shifted in the latter half of the season from being based on competition to being more neutral, although other factors, independent of resource competition with A. flos-aquae, may have kept microcystin concentrations low earlier in the season. However, the difference in nitrogen to phosphorus ratios (TN:TP and TPN:TPP) measured between the first and second A. flos-aquae blooms should be noted, in that they indicate changes in nitrogen-to-phosphorus stoichiometry in phytoplankton cells under different environmental conditions. The determinants of optimal cellular nitrogen:phosphorus stoichiometry under different ecological scenarios have been previously derived and modeled (Klausmeier and others, 2004) and, in Upper Klamath Lake, indicate that the higher nitrogen to phosphorus ratios during the first A. flos-aquae bloom may have resulted from the allocation of more nutrients to phosphorus-poor cellular resource-acquisition machinery (for photosynthesis and nitrogen fixation) under competitive equilibrium. Likewise, the lower nitrogen to phosphorus ratios characteristic of the second A. flos-aquae bloom may have resulted from cells using more nutrients for phosphorus-rich protein assembly machinery (ribosomes) during exponential growth under nutrient replete conditions. If so, it is plausible that A. flos-aquae may have lower physiological requirements for nitrogen than toxigenic M. aeruginosa and, therefore, may “share” more nitrogen during the second bloom. A facilitative relation between diazotrophic and non-diazotrophic groups, in which the presence of the diazotrophs supported a larger population of non-diazotrophs than possible in the absence of the diazotroph, was recently observed in a marine environment (Agawin and others, 2007). The degree to which the relation between M. aeruginosa and A. flos-aquae is facilitative or neutral may be an important factor for determining the effects of using nutrient reduction to diminish or eliminate the A. flos-aquae bloom on the presence of M. aeruginosa and microcystins in Upper Klamath Lake.


First posted May 30, 2012