The influence of iron and siderophores on eukaryotic phytoplankton growth rates and community composition in the Ross Sea

Shipboard manipulation experiments were conducted during the 2011 austral summer to assess the responses of eukaryote phytoplankton assemblages in the Ross Sea to ferric chloride, Fe-siderophore complexes, and apo-siderophore addition. An emphasis was placed on the model catecholate siderophore enterobactin, including limited experiments to test hypothesized interactions between Cu availability and the capacity to access Fe from Fe-enterobactin. Stations included those on Pennell Bank, along the western flank of Pennell Bank and at the shelf break. Growth rate estimates based on chlorophyll a (chl a) accumulation and nitrate drawdown under constant illumination, as well as silicate:nitrate (Si:N) drawdown and variable fluorescence (Fv/Fm), were used to evaluate bulk phytoplankton responses to Fe and siderophores. Responses to ferric chloride, copper and Fe-enterobactin plus Cu manipulations revealed widespread Fe limitation but no evidence of Cu limitation of enterobactin availability. These results were consistent with ambient surface water dissolved Fe concentrations ranging from 110 to 320 pM, known to limit Southern Ocean diatoms, and with dissolved Cu profiles showing non-limiting surface water concentrations of ~ 1.8 nM. Fe-enterobactin additions (2 nM) stimulated growth of Ross Sea populations to the same extent as ferric chloride based on either chl a accumulation or nitrate drawdown, except on Pennell Bank (station 7) where the chl a-based growth rate was slightly lower. All Fe-enterobactin treatments resulted in decreases in Si:N drawdown and increases in Fv/Fm by the end of the incubations. Given the general availability of Fe-enterobactin among these treatments, we hypothesized that apo-enterobactin might be an effective shuttle from less available reservoirs of Fe, and therefore increase Fe availability. Apo-enterobactin led to a modest improvement in nitrate-based growth at one station. We amplified the V7-V9 hypervariable region of 18S ribosomal DNA with high phylogenetic resolution to investigate whether the source of Fe elicited a shift in community composition, which would suggest clade-specific capacities to access siderophore bound Fe. Despite differences in community structure at Pennell Bank (dominated by Chaetoceros clades) and other locations (dominated by Pseudonitzschia), there was no evidence for changes in community composition during incubations in response to ferric chloride or enterobactin treatments. Preliminary results from one incubation suggest that Fe limitation was relieved with Fe added as any one of four model siderophores (enterobactin, ferrichrome, ferrioxamine, and pyoverdin) of differing coordination, stability and dissociation rates, suggesting that extant species may be able to utilize Fe, either directly or indirectly, from a variety of strong ligands on these time scales.