Special Articles

Trichodesmium, a Globally Significant Marine Cyanobacterium

See allHide authors and affiliations

Science  23 May 1997:
Vol. 276, Issue 5316, pp. 1221-1229
DOI: 10.1126/science.276.5316.1221


Planktonic marine cyanobacteria of the genusTrichodesmium occur throughout the oligotrophic tropical and subtropical oceans. Their unusual adaptations, from the molecular to the macroscopic level, contribute to their ecological success and biogeochemical importance. Trichodesmium fixes nitrogen gas (N2) under fully aerobic conditions while photosynthetically evolving oxygen. Its temporal pattern of N2 fixation results from an endogenous daily cycle that confines N2 fixation to daylight hours.Trichodesmium colonies provide a unique pelagic habitat that supports a complex assemblage of consortial organisms. These colonies often represent a large fraction of the plant biomass in tropical, oligotrophic waters and contribute substantially to primary production. N2 fixation by Trichodesmium is likely a major input to the marine and global nitrogen cycle.

Trichodesmium, a colonial marine cyanobacterium (1) (Fig. 1), has intrigued naturalists, biologists, and mariners for well over a century (2). These cyanobacteria have been reported throughout the tropical and subtropical Atlantic, Pacific, and Indian oceans, as well as the Caribbean and South China seas (Fig. 2) (3,4). Modern interest in Trichodesmium dates to the early 1960s with the recognition that the biological productivity of large expanses of the ocean is often limited by the availability of nitrogen (5) and the observation that Trichodesmium is diazotrophic (that is, an N2 fixer). The current focus in assessing the global role of the upper ocean in assimilating atmospheric CO2 has elevated the importance of quantifying marine N2 fixation.

Figure 1

Examples of Trichodesmium colonies. (A) Fusiform or tuft of Trichodesmium culture IMS 101; (B) radial or puff colony of T. thiebautii. Colonies are typically ∼2 to 5 mm in length (fusiform) or diameter (radial) and are composed of tens to hundreds of aggregated filaments (trichomes). Each trichome consists of tens to hundreds of cells (typically ∼100); cells are generally 5 to 15 μm in diameter but can range up to 50 μm in length (15) (photos by H. Paerl).

Figure 2

Location of process-oriented studies and distribution of Trichodesmium in the world’s oceans based on maps compiled by Carpenter (3) with reference to surface nutrient and productivity distributions from Berger et al. (85) to exclude areas of coastal upwelling and equatorial divergence. Dashed line indicates approximate extent ofTrichodesmium penetration into subtropical waters. AlthoughTrichodesmium is found in waters colder than 20°C, growth and activity are usually restricted to waters above 20°C (3,4). See Table 1 for key to study locations.

Although major advances in understanding the biology ofTrichodesmium have recently occurred on diverse fronts, several important questions remain largely unresolved: (i) Where doesTrichodesmium fit in the broader scheme of cyanobacterial phylogeny? (ii) How does Trichodesmium sustain simultaneous photosynthetic O2 evolution with nitrogenase activity, and why does it fix N2 only during daylight periods? (iii) What physiological, morphological, and behavioral adaptations contribute toTrichodesmium’s ecological success in the oligotrophic marine environment? (iv) What environmental and ecological factors control production and N2 fixation inTrichodesmium in situ, and to what extent does it contribute to productivity, nutrient cycling, and trophodynamics in tropical and subtropical seas? (v) What is the overall importance ofTrichodesmium N2 fixation and primary production in the global marine N and C cycles?

Molecular and Physiological Features

Open-ocean marine N-cycle studies and basic N2fixation research converged in 1961 when Dugdale et al. first identified Trichodesmium as a putative light-dependent N2 fixer (6). However, as a nonheterocystous N2 fixer, Trichodesmium represented a clear exception to the prevailing dogma (7) and, in the absence of a pure culture or other definitive evidence, its ability to fix N2 was viewed with some caution (8). A comparison of the DNA sequence of the gene that codes for the Fe protein of nitrogenase (nifH) obtained from natural populations of Trichodesmium with that from aTrichodesmium isolate (9) provided the first direct evidence of the cyanobacterial origin of the nitrogenase activity associated with Trichodesmium. Later immunolocalization studies showed that the nitrogenase protein occurs within Trichodesmium cells (10-12).

The diversity among forms of Trichodesmium with respect to cellular and colonial morphology has prompted extended debate concerning its taxonomy and phylogenetic position within the cyanobacteria. With the relatively recent isolation and maintenance of cultures of the organism (13, 14) and the development of structural and molecular biological approaches to characterizing natural populations, the taxonomic status of Trichodesmiumpopulations has gained firmer footing. Sequence analysis ofnifH and 16S ribosomal DNA (rDNA) has enabled inference of the taxonomic identity of the Trichodesmium sp. isolates and has provided additional bases for distinguishingTrichodesmium species as well as the means to determine the relative phylogenetic relation of Trichodesmium to other bacteria and cyanobacteria (15, 16). The data are consistent with differences in trichome dimensions and colony morphology that had been used in previous taxonomic schemes (15). The nitrogenase DNA sequences of field samples of T. thiebautiiand both field and culture samples of T. erythraeum were very similar (98% over 325 nucleotides), in contrast to comparisons of nitrogenase nifH gene sequences between species of other cyanobacterial genera, which are as low as 75% similar.Trichodesmium nifH sequences from three species form a deeply branching cluster within the cyanobacteria (Fig.3), implying a very early radiation in cyanobacterial evolution. Although the nifH sequence indicates thatTrichodesmium sp. NIBB 1067 is distantly related to other cyanobacteria, it appears relatively closely related to an oscillatorian on the basis of its 16S rDNA sequence (17). The reason for the differences in phylogenetic association based on the two genes is not yet clear but could involve convergent evolution (in nifH) or lateral transfer (ofnifH or 16S ribosomal RNA genes). The rather distant relation of the Trichodesmium nifH gene to that of other cyanobacteria suggests that Trichodesmium nifH evolution may be constrained by the structural requirements for aerobic N2 fixation.

Figure 3

Evolutionary relation of the genusTrichodesmium to other diazotrophs on the basis ofnifH amino acid sequence data. The evolutionary distances between deduced amino acid sequences were used to create a phenogram using the neighbor-joining method of PHYLIP (96). The Fe protein gene (nifH) of Trichodesmium was amplified from cultures by means of the polymerase chain reaction (9). Trichodesmium nifH sequences cluster relatively distantly from other cyanobacterial genera, implying that the Trichodesmium genus diverged early in evolution. ThenifH DNA sequences obtained from field collections ofT. erythraeum and T. thiebautii are very similar (98%) and could be distinguished by only a few signature nucleotides (short distances among Trichodesmium speciesnifH sequences cannot be distinguished here).

One of the most intriguing aspects of Trichodesmium biology is the simultaneous occurrence of N2 fixation and photosynthesis. Several hypotheses explaining howTrichodesmium might fix N2 aerobically have been proposed. First, it is possible that the properties ofTrichodesmium nitrogenase are unique with respect to their resistance to O2 inactivation. Second, nitrogenase may be transiently modified to protect it from permanent O2deactivation, or it may be continually synthesized to replace protein being inactivated by O2. Third, there may be intracellular O2-consumptive processes that maintain O2 at concentrations compatible with N2 fixation. Finally, N2 fixation and photosynthesis may be spatially segregated in some manner—for instance, either within specific regions in a colony or by cell differentiation within a trichome—such that N2 fixation and photosynthesis are mutually exclusive within individual cells.

Mapping of the Trichodesmium nitrogenase operon and subsequent sequencing of nitrogenase structural genes demonstrated that the gene size and organization were similar to those of other heterocystous and nonheterocystous cyanobacteria (16). Modeling of the three-dimensional structure of the Fe protein from the deduced amino acid sequence did not reveal unique or unusual features of the Trichodesmium nitrogenase Fe protein, relative to those of other cyanobacteria, that might explain a higher O2 tolerance (18). Moreover, elevated O2 concentrations lead rapidly to inhibition of N2 fixation, whereas decreased partial pressure of O2 often stimulates activity (19); hence, nitrogenase activity in Trichodesmium is O2-sensitive.

Continuous nitrogenase synthesis, as observed in other nonheterocystous cyanobacteria (7), could replace nitrogenase inactivated by O2 and thereby provide a mechanism for simultaneous N2 fixation and photosynthesis. However, nitrogenase activity in Trichodesmium is sustained for several hours in the presence of the protein synthesis inhibitor chloramphenicol (19, 20) and well after the disappearance ofnifH messenger RNA (21), which indicates a relatively low turnover rate of nitrogenase protein during the active period of N2 fixation. In some diazotrophs, nitrogenase may be protected from permanent inactivation by O2 through conformational changes or covalent modification (7). The Fe protein of Trichodesmium nitrogenase can be modified in response to O2 stress (19, 22), although the nature of the modification and whether this modification confers protection have yet to be determined.

Experimental evidence exists for the role of O2 removal processes in maintaining low intracellular O2concentrations and thereby facilitating contemporaneous N2fixation and photosynthesis. Respiration rates inTrichodesmium appear high relative to other cyanobacteria, resulting in high compensation points (typically 100 to 200 μmol m−2 quanta s−1) (23). Several other biochemical or physical means of O2consumption have been suggested (24).

Fogg (25) first offered the provocative hypothesis that a spatial segregation of O2 evolution and nitrogenase activity (analogous to that between vegetative cells and heterocysts) occurs within Trichodesmium colonies. He postulated that photosystem II (PS II)–associated O2 production took place in trichomes near the periphery of the colony, whereas nitrogenase activity was confined to the inner portions of the colony that lacked PS II. Experimental evidence supporting this idea was provided (26, 27) soon after Fogg’s original suggestion. However, other findings suggest that nitrogenase and oxygenic photosynthesis may co-occur in cells and that colony integrity is not an absolute prerequisite for activity (28). Early suggestions thatTrichodesmium might differentiate N2-fixing and photosynthesizing cells along individual trichomes have recently gained support (29). If cellular delegation of activity through differentiation is conclusively demonstrated as a general mechanism inTrichodesmium, it would provide an invaluable model for molecular-level studies of a simple differentiated system.

The proposed mechanisms summarized above are not mutually exclusive.Trichodesmium most likely uses several strategies to permit nitrogenase activity during photosynthesis. Other mechanisms that enable the co-occurrence of nitrogenase activity and oxygenic photosynthesis in this organism may yet be identified. Regardless of how Trichodesmium is able to fix N2 in the light, it is equally curious that in natural populations N2fixation occurs only during the day and is not performed during the night, as is characteristic for other nonheterocystous cyanobacteria (7). Saino and Hattori (30) first observed that nitrogenase was only active in samples of Trichodesmiumcollected during daylight hours: Samples collected at night were incapable of N2 fixation under artificial light. They suggested, 8 years before circadian rhythms were identified in any prokaryote (31), that the daily cycle of N2fixation in Trichodesmium might be attributable to an endogenous rhythm. Research into the dynamics of the nitrogenase pool and nifH transcription in natural populations revealed a dynamic daily cycle of synthesis, activity, and degradation, directly coupled to the light cycle (19-21), thus providing a mechanistic explanation for the original observations.

A primary criterion for establishing the endogenous nature of N2 fixation in Trichodesmium—its persistence over several cycles in constant light—has recently been provided inTrichodesmium sp. IMS 101 (32). The circadian clock, which appears to be set by illumination patterns, is likely to be of adaptive significance in ensuring that synthesis is induced in anticipation of the light period, thus optimizing the efficiency of light-driven N2 fixation in the relatively stable environment of tropical and subtropical seas. This is the first endogenous rhythm to be confirmed in a prokaryote other than a coccoid cyanobacterium (31).


Trichodesmium primarily inhabits surface waters of oligotrophic, tropical, and subtropical oceans, and is encountered in high abundance in western boundary currents (for example, the Gulf Stream, Kuroshio), in tropical portions of the central gyres, and in several ocean margin seas (3) (Fig. 2). The water column of these environments is generally very stable, with the upper mixed layer often around 100 m. This zone is characterized by low nutrient concentrations, very clear waters, and deep light penetration.

In an environment where the densities of microorganisms are very low and the waters highly transparent, Trichodesmium is unusual in that it is visually prominent, especially during surface blooms. The observed abundance of Trichodesmium in nutrient-depleted waters prompts the question of how it is uniquely adapted to this environment. Characteristics of Trichodesmium that appear to contribute to its success in the oligotrophic open ocean include its capability to fix N2; its natural buoyancy, which positions it in the upper water column; a photosynthetic apparatus adapted to a high-light regime; and a relatively low growth rate, which, coupled with a lack of major grazers, allows it to maintain relatively high biomass.

Despite its ability to fix N2 and its habitation in the high-light environment of the near surface, Trichodesmiumhas growth rates that are low relative to those of many eukaryotic phytoplankton (4). Even in culture, the doubling times ofTrichodesmium are slow (∼3 to 5 days), which suggests that a relatively low growth rate may in fact be an adaptation for exploiting the high-energy but low-nutrient conditions of the oligotrophic oceans (33). Because of its diazotrophic capacity, Trichodesmium growth rates are presumably limited by the availability of non-N nutrients. Field and laboratory data suggest that Fe is a key factor limiting TrichodesmiumN2 fixation and growth rates (34-36).

A key characteristic of Trichodesmium is the presence of gas vesicles, which provide buoyancy (37) and help maintainTrichodesmium populations in the upper surface waters. As has been noted in other planktonic cyanobacteria, the buoyancy ofTrichodesmium is a dynamic property: A daily cycle of rising and sinking of colonies is often observed, and this may be a result of cell ballasting through the progressive increase of relatively dense carbohydrate and protein accumulating from photosynthesis through the day (38).

Wind stress at the surface affects the qualitative distribution ofTrichodesmium populations. Relatively high and steady winds of the Trade Wind belts mix plankton populations throughout the upper euphotic zone, but natural buoyancy counteracts mixing to the bottom of the mixed layer; population densities are generally greatest at relatively shallow depths (20 to 40 m) in the upper water column (3, 4) (Fig. 4).

Figure 4

Average depth distributions of trichomes of T. thiebautii and volumetric N2fixation (A), and chlorophyll a (Chl a) biomass and primary productivity relative to microplankton (B), for a series of stations taken in the southwestern tropical North Atlantic in May 1994 (Fig. 2, map code 9) (97). Bars indicate ±SE of the mean.

Localization of the population in the high-irradiance upper water column is an adaptation that may provide a solution to the constraint of high compensation points (23) (equivalent to compensation depths of ∼50 to 70 m in oligotrophic waters) and the added energetic demands of N2 fixation. Pigment composition and photosynthetic parameters indicate a photosystem adapted for both maximum efficiency and photoprotection at high light (4,39). Moreover, Fe enters the open ocean primarily through atmospheric deposition, and location of these organisms in the upper water column may be advantageous with respect to Fe acquisition (35).

When wind stress is low for an extended period, the intrinsic buoyancy of Trichodesmium can have striking consequences, leading to the development of extensive surface blooms or “red tides” (3, 39, 40). These phenomena, which range in actual color from yellow to brown, are easily observed by satellite (Fig.5) (41, 42). The accumulation of the population at the surface for an extended period may result in photoinhibition (43) and, possibly, photooxidative damage (3, 4,39); however, bloom organisms appear to be metabolically active, growing (40, 41), and relatively resistant to photoinduced damage (39).

Figure 5

(A) Image of chlorophyll and (B) relative Trichodesmium abundance derived from a coastal zone color scanner (CZCS) image of the northwestern coast of Australia in the vicinity of the Dampier Archipelago and confirmed by contemporaneous sea-truth data [adapted from (42)]. A protocol based on reflectivity and absorption at 550 nm was used. Chlorophyll is reported as detected by CZCS, with lowest to highest chlorophyll a concentrations ranging from purple (<0.05 mg m−3) to red (>3.0 mg m−3). ForTrichodesmium, dark colors indicate its absence; lighter colors (light blue through orange) indicate its presence. Differences in color represent varying responses to the protocol, not necessarily differences in Trichodesmium concentration.

Although physical processes most often dominate oceanic plankton distributions, large surface accumulations of Trichodesmiummay affect the bulk physical and chemical properties of surface waters. Mesoscale sea-surface features of the extent and intensity characteristic of some Trichodesmium blooms likely modify light penetration and the quality of the in situ light field (43) as well as heat and gas transfer across the ocean-atmosphere interface (44). Organic and inorganic nutrients accumulate during blooms and can affect subsequent phytoplankton succession and productivity (40).

Ecologically, Trichodesmium affects the structure and function of the oligotrophic ocean by contributing to its productivity and trophodynamics, as well as by providing a unique pelagic habitat. Upon close inspection, many colonies reveal a microcosm including bacteria, other cyanobacteria, protozoa, fungi, hydrozoans, and copepods (45). Nitrogen and carbon fixed by Trichodesmiumenters the food web, but not necessarily through increased C and N flux into classical food chains. Some species of Trichodesmiumproduce a toxin that deters grazing by calanoid and cyclopoid copepods, considered to be the major grazers in these systems (46). A specialized group of harpacticoid copepods appear adapted to the toxin and are capable of directly grazing Trichodesmium(47), although their quantitative role in the consumption ofTrichodesmium is unknown. Although there are anecdotal reports of the presence of Trichodesmium in the guts of gelatinous zooplankton and fish (4), this does not appear to be a quantitatively important fate for Trichodesmiumbiomass, and much of the C and N fixed by Trichodesmiumlikely enters upper trophic levels by other pathways. In natural populations of Trichodesmium, a large proportion of recently fixed N is released as organic N (48). Besides providing a means of exchange between N2-fixing and non–N2-fixing cells in the colonies, this exudation may be an important source of C and N for microbes inhabiting the colonies (49) or free-living in the water column, and ultimately may be a source of inorganic N for phytoplankton.


With its cosmopolitan distribution throughout much of the oligotrophic tropical and subtropical oceans (3) and its capacity to form extensive surface blooms (40-42),Trichodesmium may be one of the most globally important cyanobacteria and phytoplankton. However, there is insufficient quantitative information to substantiate this inference, possibly because of the unique problems associated with assessing the biomass and productivity of Trichodesmium by traditional methods as well as the inherent difficulty in sampling ephemeral blooms (50). As a result, the contribution ofTrichodesmium to marine C and N input has generally been considered relatively small (51). Recent data, however, suggest otherwise.

Accurately quantifying N2 fixation in the seas has direct bearing on our understanding of C flux in the tropical oceans. According to the “new production” concept (52), in steady-state marine systems the amount of organic N removed from the system (that is, the euphotic zone) should equal external (“new”) N inputs entering that system. Although N2 fixation is a potential source of new N, NO3 from depth is generally considered the quantitatively dominant source of new N in most open-ocean systems (52). Areal rates of total production in tropical oligotrophic regions are typically very low (53), and the relative amount of production dependent on new N (that is, “new” production) is thought to be small; nonetheless, the export of organic N and C from the upper water column of oligotrophic regions is important in the marine C and N budgets because the total area involved is vast (53).

The increase in atmospheric CO2 has provided added impetus to the quest to quantify oceanic primary production and its rate of removal to depth (54). N2 fixation and vertical NO3 flux from depth have different potentials for supporting primary production and effecting net removal of CO2 from the atmosphere. Vertical NO3 flux occurs with a concurrent upward flux of CO2 and PO4 3– from depth, often close to the stoichiometric requirement of phytoplankton. Thus, relative to N2 fixation, NO3 derived from depth has limited capacity for effecting net removal of atmospheric CO2. N2 fixation represents a source of new N entering the system that can account for a net sequestering of atmospheric CO2 into export production (55).

Recently, several independent lines of evidence have prompted speculation that marine N2 fixation has been severely underestimated and may play a larger role in global C and N fluxes. Large imbalances in the estimated N budgets for the North Atlantic, central North Pacific, Indian, and global oceans have prompted speculation as to unknown or poorly quantified N inputs, and N2 fixation may provide the missing source of N (56).

For nonupwelling regions of the tropical and subtropical North Atlantic, the vertical flux of NO3 , assumed to represent the bulk of the influx of new N into the euphotic zone (52), is estimated to range from 30 to 150 μmol N m−2 day−1 (57). However, such estimates of new N inputs are often insufficient to satisfy the calculated demand for new N and organic export from the euphotic zone (57), which suggests the existence of sources of combined N in addition to vertical NO3 flux. Other biogeochemical approaches also reveal N demand in, or losses from, the upper water column exceeding current estimates of N inputs, further indicating unquantified inputs (58).

Independent evidence that N2 fixation may be of greater relevance than currently believed derives from studies of the natural isotopic abundance of 15N in surface particulate organic nitrogen (PON). Trichodesmium has a low δ15N, typical of N2-fixing organisms (59). Relative to the isotopic composition of PON in more nutrient-rich areas, the δ15N of suspended particles and zooplankton in surface waters of the western and central tropical Pacific and in the Caribbean is often low and inversely related to the abundance ofTrichodesmium. This suggests that the isotopically light PON in these waters is a result of N2 fixation byTrichodesmium (59).

Relative to conventional assessments of planktonic standing crop, primary productivity, and inorganic N uptake, there are few direct estimates of Trichodesmium biomass and its rate of N2 fixation (50) (Table 1) and even fewer estimates of its contribution to primary production. The most comprehensive studies to date have been made in the subtropical Sargasso Sea and in the tropical Caribbean and far western Pacific (Kuroshio, South and East China Sea) (Fig. 2), ocean margin regions that may not be representative of the tropical oceans. More limited data are available for the tropical North Atlantic and the vicinity of Hawaii. Virtually no information exists for large expanses of the tropical oceans, particularly in the Southern Hemisphere. Nonetheless, several of the studies conducted over the past two decades have provided direct evidence of the importance of Trichodesmiumrelative to other phytoplankton (60). With respect to subtropical waters such as the Sargasso Sea, previous studies have detected appreciable populations of Trichodesmium only for limited periods during the summer, and the calculated contributions ofTrichodesmium to C and N input were relatively small (Table1) (61).

Table 1

Summary of direct areal estimates of N2fixation, as originally reported or as derived. Studies based on C2H2 reduction determinations of N2fixation used a 3:1 conversion from C2H2 reduced to N2 fixed, unless otherwise noted. N, number of discrete observations.

View this table:

In tropical studies, despite low growth rates, the relatively high biomass of Trichodesmium in these regions implies that it may account for a quantitatively important input of organic C, and data from several studies have provided support for this idea (60) (Fig. 4). Not surprisingly, the most data exist for estimates of N2fixation (Table 1). The average rate of N2 fixation reported for a suite of stations throughout the Caribbean Sea was 77 μmol N m−2 day−1. One station in the tropical North Pacific yielded an N2 fixation rate of 134 μmol N m−2 day−1 (62-65). We calculated 126 μmol N m−2 day−1 on the basis of data (63) for the China Sea and Kuroshio current near the southern tip of Japan (Table 1). Our recent results from the open tropical Atlantic Ocean (Fig. 4) are generally consistent with earlier data from this and other tropical regions (Table 1). Scaling the average (nonbloom) tropical rate of 106 μmol N m−2day−1 to tropical oligotrophic waters that cover about 150 × 106 km2 (53) yields an annual input of ∼80 Tg N, considerably greater than previous estimates of pelagic N2 fixation input (3, 40).

These estimates of N input through N2 fixation in tropical areas are roughly equivalent to the estimates of vertical NO3 flux given above and suggest that these two processes are of comparable importance in new N input. Moreover, recent evidence indicates that actual eddy diffusivities are at the lower range of those values currently assumed (66). If this is correct, it would lower the estimate of N input by vertical flux of NO3 and correspondingly increase the relative importance of upper water column N2 fixation to overall N input. The extreme depth of the euphotic zone in the oligotrophic open ocean (>100 m), and the absence of measurable NO3 throughout much of the upper water column of the highly oligotrophic tropical areas, imply an uncoupling between the reservoirs of NO3 at depth and productivity in the near surface. Any NO3 that penetrates up from depth is likely to be assimilated at the chlorophyll maximum, typically located near the 1% light level near the nitracline.

Although numerous blooms have been documented and their spatial extent and biomass density have been estimated (40-42), the effect of a given bloom on C and N input has been determined only on a few occasions. For three studies that directly measured N2fixation in surface blooms, the input of N in the bloom was about three times that occurring throughout the rest of the water column (67).

Taken together, these observations strongly indicate that N2 fixation is an important component of the marine N budget that needs to be considered in basin-scale studies of N cycling and in calculations of global N budgets. In addition to N2fixation, other potentially significant “new” N inputs, such as atmospheric deposition of dissolved inorganic and organic N, have been poorly quantified or ignored in the past (68). More accurate assessment of their contributions to the oceanic N cycle will further help to rectify current discrepancies in basin-scale and oceanic N budgets.

Future Research and Prospects

Research efforts from a molecular to a global perspective provide a new basis for understanding the biology and ecology ofTrichodesmium and inferring its role in global biogeochemical cycles. Physiological, genetic, and immunological evidence have confirmed Trichodesmium, rather than associated microorganisms, as the main N2 fixer in its colonial aggregates. Trichodesmium fixes N2aerobically, possibly in the same cell and at the same time as it evolves O2 through photosynthesis, making it a unique and valuable model organism for the study of N2 fixation in photosynthetic organisms. As one of the few prokaryotic systems identified as having an endogenous rhythm, Trichodesmium has a relative biochemical simplicity that makes it an attractive system for identifying the genetic and physiological factors regulating components of its biological clock.

Multiple characteristics have been identified that contribute to the ability of Trichodesmium to fix N2 aerobically, but a single critical component that allows simultaneous photosynthesis and nitrogenase activity of nitrogenase has yet to be found, and seemingly conflicting results need to be reconciled. The dichotomy in growth rate estimates based on C or N in natural populations needs also to be resolved. The availability of robust cultures will greatly improve our knowledge of the biochemical functioning of these unusual diazotrophs.

A variety of recognized morphological and functional features demonstrate Trichodesmium to be well adapted to the N-poor oligotrophic ocean environment. Of particular ecological relevance may be its cyclic patterns of vertical migration, N2 fixation, and photoprotective processes. However, we have yet to develop an integrative physiological model directly linking these ecological behaviors in cultures or in natural populations. More rigorous definition of the interplay of physical and chemical limiting factors will provide important constraints for modeling and predicting the in situ dynamics of these populations.

The accumulating evidence strongly indicates a much more important role of Trichodesmium in oceanic biogeochemistry than it is currently afforded. Synoptic estimates of the areal and temporal extent of Trichodesmium populations and, particularly, blooms by new and planned satellite sensors [for example, the ocean color and temperature scanner (OCTS) and the sea-viewing wide field-of-view sensor (Sea- WiFS)], coupled with the development of in situ methods of enumerating these populations (for example, fluorescence-based optical plankton counters and laser-induced fluorescence imagers), will further refine and advance our knowledge of occurrence in tropical and subtropical seas. This information will be useful in determining the potential role of blooms in the modification of system-scale features such as heat, material flux, and albedo. The more comprehensive database onTrichodesmium population biomass and distribution derived from satellite studies, along with expanded direct information on its in situ contributions to C and N cycling, will also allow for more precise extrapolation of its oceanic contributions to new N inputs as well as the inclusion of Trichodesmium as an explicit component of large-scale modeling of oceanic productivity.

Variations in oceanic productivity over glacial-interglacial time scales have been directly related to subtle variations in the extent of denitrification and cyanobacterial (that is, Trichodesmium) N2 fixation in the sea (69). The possibility exists for use of nucleotide probes of nifH to examine paleoecological trends of Trichodesmium populations in oceanic sediment cores (70). Developing a firm understanding of the dynamics and controls of Trichodesmium populations in the contemporaneous ocean, along with information on its population trends over geological time scales, will lead to important new insights about controls of oceanic productivity.


View Abstract

Stay Connected to Science

Navigate This Article