The cyanobacterial nitrogen fixation paradox in natural waters

Nitrogen fixation, the biochemical conversion of “inert” atmospheric N (N₂) to biologically available ammonia (NH₃), is a microbially mediated process of global significance because it provides “new” N to aquatic ecosystems in which biological production is often controlled by N availability^1,2. N₂ fixation is an anaerobic process carried out by specific prokaryotes, including heterotrophic and chemolithotrophic bacteria and some cyanobacteria (blue-green algae)³. The process likely evolved during the oxygen (O₂)-devoid Precambrian period some 2+ billion years ago^4,5. Of the N₂-fixing microbial taxa, the cyanobacteria are of particular biogeochemical and ecological interest because they were also the first O₂-evolving photosynthetic organisms on Earth⁶; their proliferation during this period is thought to be an evolutionary “milestone” because it led to the generation of an O₂-rich atmosphere, a prerequisite for the evolution of O₂-requiring fungi, bacteria, animals, and higher plant species on our planet⁶.

Ironically, the development of an O₂-rich atmosphere, hydrosphere, and pedosphere constituted a formidable biochemical challenge for the cyanobacteria because, while they were capable of fixing N₂, the process had to be confined to an O₂-free micro-environment⁷. This requirement posed a serious dilemma, especially for aquatic cyanobacteria, because they require illuminated conditions in surface waters, but the high ambient O₂ levels produced by photosynthesis in these waters also represents an environmental barrier to O₂-sensitive N₂ fixation. Over their long evolutionary history, cyanobacteria have developed biochemical and structural adaptations as well as biotic associations in order to optimize N₂ fixation while relying on oxygenic photosynthesis to provide energy and organic carbon (C) compounds to support metabolism and growth. The adaptions include (1) confining N₂ fixation to night-time when photosynthesis is “turned off”, (2) forming colonies and aggregates to reduce illumination and form low-O₂ “microzones”, (3) participating as endosymbionts in biological associations, and (4), forming heterocysts (non-photosynthetic, O₂-free cells) in some filamentous taxa, which allows N₂ fixation to proceed while receiving photo-reductant and organic C through photosynthesis from adjacent cells⁸.

These are all remarkably clever adaptations to a modern-day oxic biosphere, which help circumvent the “O₂ problem”⁶. From an ecosystem perspective, they have allowed N₂-fixing species to provide biologically available N from the vast reservoir of atmospheric N₂. However, on the ecosystem scale, recent N budget analyses indicate that N₂ fixation inputs fall far short of meeting ecosystem requirements when biologically available N inputs (from terrestrial and atmospheric sources) and losses (via denitrification, sedimentation and burial, and advection) are considered^9–11. As a result, freshwater, estuarine, and marine systems are often chronically N deficient^11–17. Pervasive N limitation has many implications for ecosystem function, especially when excessive external nutrient inputs lead to accelerating primary production (eutrophication), harmful algal blooms, and excessive O₂ consumption (hypoxia). If chronic N-limited conditions prevail in water bodies and N₂ fixation cannot meet ecosystem N requirements, then external N inputs often supply N to support eutrophication and its unwanted symptoms. From a management perspective, this means that the growing global glut of N inputs from agricultural, urban, and industrial sources^14,18–20 needs to be controlled, in addition to the broadly accepted phosphorus (P) input constraints, in order to protect our waterways and water supplies.

Why does N₂ fixation fall short of meeting ecosystem demands? Apparently, this process does not operate at sufficient rates in a modern-day, oxic world to compensate for losses via burial, export, and denitrification, even though it is protected and optimized by the various biological adaptations mentioned above. It is counteracted at larger scales by biogeochemical processes, such as denitrification, that run in the opposite direction (NO₃ → N₂). The N₂-fixing process is an energy-demanding one, requiring 16 ATP molecules to fix one molecule of N₂³. In cyanobacteria, this energy demand has to be met by photosynthesis, while in non-photosynthetic bacteria, organic matter and redox reactions serve as energy sources³. In highly productive (eutrophic), turbid waters where cyanobacteria and bacteria thrive, the availability of photosynthetically active radiation (PAR: 400–700 nm) is often restricted, causing a radiant energy deficit and suboptimal N₂ fixation rates. Secondly, cyanobacteria taxa that dominate in eutrophic waters often accumulate as thick surface “blooms”, in part to circumvent light limitation in subsurface waters¹¹. High rates of photosynthesis in such blooms lead to O₂ supersaturation, often in excess of 200% saturation²¹. These ambient O₂ levels inhibit N₂ fixation in situ, even in heterocystous taxa^22,23. Thirdly, N₂ fixation requires high levels of P (to support the energetics, e.g. ATP formation and nucleic acid production) and metals, most prominently iron (Fe), which is a co-factor in the enzyme complex nitrogenase³. In highly oxygenated surface waters, Fe occurs as the insoluble and biologically unavailable Fe³⁺ ion that may lead to Fe-limited conditions²⁴. Lastly, wind-induced turbulence and vertical mixing can reduce N₂ fixation potential by disrupting colonies and aggregates and enhancing inward diffusion of O₂ (Figure 1)²⁵ and deepening the mixed layer, reducing light availability.

Figure 1. The nitrogen fixing process, as mediated by cyanobacteria (utilizing oxygenic photosynthesis as an energy and carbon source) as well as heterotrophic and chemolithotrophic microorganisms, in eutrophic surface waters.

Potential environmental controls, including phosphorus (P) and iron (Fe) availability, energy sources, and dissolved oxygen inhibition, are shown in red. The background photo is of an O₂-supersaturated (during daytime) cyanobacterial surface bloom in Lake Taihu, China. Photograph by H. Paerl.

Thus, while N₂ fixation converts inert N₂ into biologically available NH₃ to support aquatic fertility in a remarkable fashion, it faces multiple constraints and limitations in aquatic environments, especially in surface waters, which are often N limited. Geochemists, some limnologists, and a few oceanographers have assumed that as long as P and Fe are readily available, N₂ fixation should make up for an N deficit, given the unlimited supply of N₂ available^26,27. However, this assumed linear stoichiometric relationship is not straightforward. Major environmental factors constrain this process, preventing it from functioning at optimal rates and supplying complete ecosystem N requirements^8,11. As a result, much of the world’s marine and freshwater environments remain chronically N deficient. In practical (management) terms, this limitation means that external inputs of N play a key role in providing adequate and excessive fertility (eutrophication) of many freshwater and most marine ecosystems^11,15,16. Tremendous increases in anthropogenically generated bioavailable N in the form of synthetic (Haber process) fertilizers, agricultural, industrial, and urban wastes, and N₂ emissions (as both oxides and reduced forms of N) far overshadow biological fixation of N₂ in providing available N to receiving waters. Effective future management and protection of our fresh and marine waters will depend on the control of external inputs of both N and P^11,27 instead of depending on the more traditional approach of controlling P inputs without N restrictions²⁸.