James W. Moffett

Phosphonate utilization by the globally important marine diazotroph Trichodesmium

The factors that control the growth and nitrogen fixation rates of marine diazotrophs such as Trichodesmium have been intensively studied because of the role that these processes have in the global cycling of carbon and nitrogen, and in the sequestration of carbon to the deep sea. Because the phosphate concentrations of many ocean gyres are low, the bioavailability of the larger, chemically heterogeneous pool of dissolved organic phosphorus could markedly influence Trichodesmium physiology. Here we describe the induction, by phosphorus stress, of genes from the Trichodesmium erythraeum IMS101 genome that are predicted to encode proteins associated with the high-affinity transport and hydrolysis of phosphonate compounds by a carbon–phosphorus lyase pathway. We show the importance of these genes through expression analyses with T. erythraeum from the Sargasso Sea. Phosphonates are known to be present in oligotrophic marine systems, but have not previously been considered to be bioavailable to marine diazotrophs. The apparent absence of genes encoding a carbon–phosphorus lyase pathway in the other marine cyanobacterial genomes suggests that, relative to other phytoplankton, Trichodesmium is uniquely adapted for scavenging phosphorus from organic sources. This adaptation may help to explain the prevalence of Trichodesmium in low phosphate, oligotrophic systems.

Developing standards for dissolved iron in seawater

In nearly a dozen open- ocean fertilization experiments conducted by more than 100 researchers from nearly 20 countries, adding iron at the sea surface has led to distinct increases in photosynthesis rates and biomass. These experiments confirmed the hypothesis proposed by the late John Martin [Martin, 1990] that dissolved iron concentration is a key variable that controls phytoplankton processes in ocean surface waters. However, the measurement of dissolved iron concentration in seawater remains a difficult task [Bruland and Rue, 2001] with significant interlaboratory differences apparent at times. The availability of a seawater reference solution with well- known dissolved iron (Fe) concentrations similar to open- ocean values, which could be used for the calibration of equipment or other tasks, would greatly alleviate these problems [National Research Council (NRC), 2002]. The Sampling and Analysis of Fe (SAFe) cruise was staged from Honolulu, Hawaii, to San Diego, Calif., between 15 October and 8 November 2004 to collect data and samples that were later used to provide this reference material. Here we provide a brief report on the cruise results, which have produced a tenfold improvement in the variability of iron measurements, and announce the availability of the SAFe dissolved Fe in seawater standards.

Complexation of cobalt by natural organic ligands in the Sargasso Sea as determined by a new high-sensitivity electrochemical cobalt speciation method suitable for open ocean work

A high-sensitivity cobalt speciation method was developed and applied to a profile in the North Atlantic. Method development included examining the redox chemistry of the analytical system and calibrating the electroactive cobalt ligand dimethylglyoxime (DMG) using EDTA as a model ligand. The method was applied to a depth profile at the Bermuda Atlantic Time Series Station (BATS) during a September 1999 cruise. Total dissolved cobalt, measured using adsorptive cathodic stripping voltammetry (ACSV) on ultraviolet light irradiated samples, revealed a nutrient-like profile for cobalt. Co speciation, measured using CLE–ACSV (competitive ligand exchange), showed a cobalt binding ligand concentration that was similar to that of total cobalt throughout the profile. An excess of ligand was observed in the chlorophyll maximum where Prochlorococcus and Synechococcus numbers were highest. A conditional stability constant for CoHDMG2 was measured to be logKCoHDMG2cond=11.5±0.3 at pH 8.0. A pH dependence for KCoHDMG2cond was observed and is consistent with model calculations based on the protonation constants for H2DMG. The conditional stability constant for CoL was determined to be logKCoLcond=16.3±0.9 and total ligand concentrations varied from 9 to 83 pM as calculated by a one-ligand non-linear fit using the Levenberg–Marquardt algorithm. Alternate interpretations of the data are discussed, including the possibility for an underestimation of ligand concentrations and stability constants caused by the existence of Co(III) ligands, and kinetic and thermodynamic competition for natural ligands by Ni and Co(II).

Oxidation of cobalt and manganese in seawater via a common microbially catalyzed pathway

Abstract Cobalt and manganese uptake onto suspended particles was studied in waters collected from Waquoit Bay, Massachusetts and the upper water column of the Sargasso Sea using radiotracers, coupled with protocols used previously for Mn and Ce to distinguish biological and redox processes. Cobalt uptake onto suspended particles in Waquiot Bay was dominated by microbial oxidation. Moreover, there was a close relationship between Mn(II) and Co(II) oxidation, with Mn(II) specific rates approximately 7–10x faster. Oxidation of each element obeys Michaelis Menten kinetics, with identifical values of Km in a given sample and values of Vmax are 7× higher for Mn. Lineweaver-Burk plots, generated from saturation plots for Co and Mn oxidation at different Mn and Co concentrations, demonstrated competitive inhibition between Co and Mn. The results indicate that both elements are co-oxidized via the same microbial catalytic pathway, and that this is probably an important mechanism for the incorporation of Co into marine Mn oxides. In the Sargasso Sea, by contrast, Mn and Co uptake onto suspended particles were completely decoupled. Cobalt uptake was nonoxidative, biologically mediated, and enhanced by low to moderate levels of light. It is probably due primarily to uptake by phytoplankton. Manganese uptake was almost exclusively oxidative and was inhibited by light even at low intensities. The differences probably reflect a higher biological demand for Co in the Sargasso Sea (Co is a biologically essential element), where Co concentrations are low, and lower activity of Mn oxidizing bacteria. Results suggest that higher specific uptake rates of Co than Mn by phytoplankton in oceanic regimes could result in Co having a geochemistry intermediate between Mn and a more nutrient-type element, such as Zn. Nevertheless, Co and Mn cycling are expected to be closely coupled in regions of high microbial Mn oxidizing activity.

Iron stress in open-ocean cyanobacteria (Synechococcus, Trichodesmium, and Crocosphaera spp.): identification of the IdiA protein.

ABSTRACT Cyanobacteria are prominent constituents of the marine biosphere that account for a significant percentage of oceanic primary productivity. In an effort to resolve how open-ocean cyanobacteria persist in regions where the Fe concentration is thought to be limiting their productivity, we performed a number of Fe stress experiments on axenic cultures of marine Synechococcus spp., Crocosphaera sp., andTrichodesmium sp. Through this work, we determined that all of these marine cyanobacteria mount adaptive responses to Fe stress, which resulted in the induction and/or repression of several proteins. We have identified one of the Fe stress-induced proteins as an IdiA homologue. Genomic observations and laboratory data presented herein from open-ocean Synechococcus spp. are consistent with IdiA having a role in cellular Fe scavenging. Our data indicate that IdiA may make an excellent marker for Fe stress in open-ocean cyanobacterial field populations. By determining how these microorganisms respond to Fe stress, we will gain insight into how and when this important trace element can limit their growth in situ. This knowledge will greatly increase our understanding of how marine Fe cycling impacts oceanic processes, such as carbon and nitrogen fixation.

Temporal and spatial variability of copper complexation by strong chelators in the Sargasso Sea

Copper(II) complexation in the upper water column was studied at the Bermuda Atlantic Time Series Station (BATS) from January 1992 to March 1993, and in the southern Sargasso Sea in April 1992, using adsorptive cathodic stripping voltammetry (ACSV). Copper titration data, analysed using a one ligand model, indicated that speciation was dominated by a strong ligand or ligand class, with a conditional stability constant of 1013.2. Total concentrations of copper and ligand were very similar, ranging from 0.9 nM to 2 nM. Concentrations of free cupric ion (Cuf2+) varied widely in the upper water column depending on whether Cu exceeded the ligand concentration or vice versa. Temporal and spatial variability in these parameters showed trends with hydrographic and biological parameters consistent with biological production and near-surface photochemical decomposition of the strong ligand. Under well stratified oligotrophic conditions, such as those prevailing year round at the southern Sargasso station and in the summer and autumn at BATS, the ligand showed a subsurface maxima coinciding roughly with the chlorophyll maximum. Ligand concentrations decreased below total Cu concentrations in the mixed layer, leading to pronounced increases in Cuf2+ concentrations. However, samples collected at BATS during or following periods of intense vertical mixing and biological activity showed excess ligand concentration throughout the upper water column and extremely low cupric ion concentrations. The spatial and temporal variability of the strong ligand at BATS is similar to that observed by Coale and Bruland (1990) in the NE Pacific, suggesting that Cu speciation in both regions is controlled by common processes. However, in that study, copper concentrations were always below the ligand concentration, so cupric ion concentrations did not display the great variability observed in the Sargasso Sea in this study.

A radiotracer study of cerium and manganese uptake onto suspended particles in Chesapeake Bay

The oxidation kinetics of Ce (III) and Mn(II) were studied in Chesapeake Bay in March and July 1990 to establish the role of water column redox processes in contributing to Ce anomalies observed in this estuary (Sholkovitz and Elderfield, 1988; Sholkovitz et al., 1992). Oxidation was measured by adding Mn(II) and Ce (III) to freshly collected water samples as radiotracers and measuring their uptake onto the ambient suspended particle assemblage. Mn(II) oxidation was measured by following the uptake of 54Mn(II) onto suspended particles and utilizing protocols established by other workers to distinguish oxidation from Mn(II) adsorption. The same protocols were applicable to Ce (III), using 139Ce(III)and were supported by the use of 152Eu(III) as a nonredox reactive control. Specific rates of Ce (III) and Mn(II) oxidation measured at a station in the North Bay (depth = 4 m) in July were 2016% per day and 4032% per day, respectively. In March at the same station, the specific rate of Mn(II) oxidation was only 10% per day and Ce (III) oxidation was undetectable. Both Ce (III) and Mn(II) oxidation processes were inhibited by azide, indicating that they were microbially mediated. The seasonal differences probably reflect strong seasonal variation in the abundance of Mn oxidizing bacteria. No Ce (III) oxidation occurred in samples collected below the oxic/anoxic interface in July. The specific rates of oxidation for both elements were over 1000 times higher than those measured in the Sargasso Sea. However, the specific rates for Ce (III) and Mn(II) were very similar to each other. This fact, coupled with similar spatial and temporal trends for specific oxidation rates, suggests a common mechanism of oxidation of both elements which may be significant in a wide range of marine environments.

Temporal and spatial variability of cobalt in the Atlantic Ocean

Abstract The spatial and temporal variability of cobalt in the Atlantic Ocean was investigated by means of adsorptive cathodic stripping voltammetry. A vertical profile of total dissolved cobalt at the Bermuda Atlantic Time Series station ranged from 17 to 73 pM and displayed surface depletion indicative of biological utilization. This profile when compared with a cobalt profile from the northeast Pacific shows no increase in deep-water concentrations with thermohaline circulation through the deep ocean basins. Moreover, the middepth maximum observed in northeast Pacific profiles is not present in the Sargasso Sea, perhaps because of the lack of cobalt scavenging by particulate manganese oxides in surface waters and to the absence of a suboxic oxygen minimum zone, which, if present, could dissolve the manganese oxides. Total dissolved cobalt measurements were also made on a surface transect from the Sargasso Sea to coastal Massachusetts, USA, and on time-series samples from the Moored In Situ Trace Element Serial Sampler. Dissolved cobalt on this transect correlated strongly with salinity ( r 2 = 0.93) and ranged from 19 to 133 pM, indicating mixing of cobalt from shelf waters into the Sargasso Sea. Time-series samples near Bermuda did not show an obvious response to the summer maximum in aeolian dust deposition, with an annual average of 20 ± 10 pM at 40- to 47-m depths. By use of this annual value and particulate cobalt data from the literature, 100-m surface-water residence times were calculated to be as low as 0.32 yr for cobalt. Several sharp decreases in cobalt were observed in the time series that occurred simultaneously with a shallowing of the thermocline depth. These decreases could be caused by nutrient drawdown associated with higher productivity mesoscale eddy events. A west-east surface transect across the South Atlantic showed high cobalt concentrations at the boundaries of the transect and low concentrations in the center despite the high precipitation rates in the intertropical convergence zone. Phosphate measurements showed the similar trends as the total cobalt transect. A regression of cobalt vs. phosphate reveals a slope that is an order of magnitude higher than that of the northeast Pacific and that is similar to the slopes observed for zinc vs. phosphate in the Pacific.

Cobalt and nickel in the Peru upwelling region: A major flux of labile cobalt utilized as a micronutrient

behaved as a micronutrient with correlations with major nutrients (nitrate, phosphate; r 2 = 0.90, 0.96) until depleted to � 50 pM of strongly complexed cobalt. Co:P utilization ratios were an order of magnitude higher than in the North Pacific, comparable to utilization rates of zinc in other oceanic regions. Cobalt speciation measurements showed that available cobalt decreased over 4 orders of magnitude in this region, with shifts in phytoplankton assemblages occurring at transitions between labile and nonlabile cobalt. Only small changes in total dissolved nickel were observed, and nickel was present in a labile chemical form throughout the region. In the Peru upwelling region, cobalt uptake was highest at the surface and decreased with depth, suggesting phytoplankton uptake was a more important removal mechanism than co-oxidation with microbial manganese oxidation. These findings show the importance of cobalt as a micronutrient and that cobalt scarcity and speciation may be important in influencing phytoplankton species composition in this economically important environment. INDEX TERMS: 1030 Geochemistry: Geochemical cycles (0330); 1050 Geochemistry: Marine geochemistry (4835, 4850); 1065 Geochemistry: Trace elements (3670); 4279 Oceanography: General: Upwelling and convergences;

Polarographic determination of half-wave potentials for copper-organic complexes in seawater

Abstract A methodology has been developed to determine the half-wave potentials of organic Cu chelators in seawater and algal culture media from pseudopolarograms obtained by anodic stripping voltammetry. A linear relationship between half-wave potentials determined from the pseudopolarogram, and thermodynamic stability constants was observed for a collection of model ligands that spanned a wide range of stability constants (log K =12–26.5) and included many functional groups likely to be important in copper complexation in seawater. Results suggested that it is possible to determine thermodynamic stability constants from half-wave potentials, as reported previously for Zn and Fe. Pseudopolarograms from culture media of three Cu stressed phytoplankton species showed very different thermodynamic stability constants, indicating the presence of chemically distinct ligands. The cyanobacterium Synechococcus produced at least two strong Cu chelators. One had a thermodynamic stability constant of log K =36.7±3.0 (obtained by extrapolation of the scale) and complexed from 30%–50% of the total Cu. The second was electrochemically inert, even at −1.6 V. The eukaryotic species produced several chelators with stability constants ranging from log K =22.6–39.1. Seawater samples collected from coastal waters showed similar characteristics. These thermodynamic constants are very high relative to model ligands in the literature, which could reflect coordination by specialized biomolecules. Alternatively, the large negative half-wave potentials could be due to Cu(I) complexes (with log K =10–20) or reflect reductive ligand decomposition of kinetically inert Cu(II) complexes.