Vitamin B12 and cobalt cycling among diatoms and bacteria in Antarctic sea ice microbial communities

Within McMurdo Sound’s annual sea ice, assimilation and concentrations of vitamin B12 (cobalamin), microbial community productivity, and biomass were examined among three 100-m2 quadrats where light penetration was manipulated by snow cover during austral summer. From late October through December, B12 concentrations (6–32 pmol L21) and assimilation rates (17–780 pmol m22 d21) in congelation ice covaried with primary productivity (0.0001–250 mmol C m22 d21 ) and chlorophyll a (0.6–36 mg m22). Within ice core samples, incorporation of Co-B12 into the .1.0-mm size fraction (mostly diatoms) was almost 100 times faster than into 0.2–1.0-mm particles (mostly heterotrophic bacteria) and was dependent on light and active transport across cell membranes. Microalgal B12 : C cell quotas in field communities varied widely (2.6–150,000 nmol B12 (mol C)21; x̄ 5 500) and generally exceeded those of cultured temperate diatoms (0.4–55 nmol B12 (mol C)21; x̄ 5 4.1). Comparisons of dissolved B12 pools in the ice and their turnover (0.02–0.6 d21) with underlying seawater suggest that this vitamin is produced in situ rather than delivered from waters below. Production and uptake of B12 and uptake of cobalt, required for B12 synthesis, were then examined among bacteria isolated from these communities. Only 23% of 78 bacterial isolates were incapable of B12 uptake, but these clones assimilated dissolved cobalt. Intracellular B12 production was evident in 9 of the 11 isolates screened and their cell quotas varied widely, 0.6–6,800 nmol B12 (mol C)21. Mass balance analyses and published kinetics data independently suggest that microalgal growth in sea ice was not limited by vitamin B12 in most of our field observations and that in situ bacterial B12 production could potentially meet microalgal demands. Similar analyses, however, suggest that cobalt supply from underlying waters may have limited community growth and B12 production. Consistent with Liebig’s law of the minimum, numerous aquatic studies have demonstrated that the nutrient in shortest supply (N, P, Si, or Fe) relative to cellular demand at any particular time or site limits system productivity (Liebig 1855). Furthermore, relative stoichiometries and speciation of these elements can influence phytoplankton community structure as a consequence of cell nutrient quotas, uptake kinetics, and specificity of individual population’s uptake systems (Tilman 1982; Karl 2002). Selective growth responses within phytoplankton communities can have important implications for carbon cycling, from influencing trophic structure to controlling carbon sequestration in the ocean’s interior. For example, nutrient regimes promoting growth of large, biomineralizing (CaCO3, SiO2) taxa tend to select for larger herbivores and favor higher carbon export to the mesopelagic as sinking aggregates and fecal pellets (de Baar et al. 1995). Recent research on iron limitation in high-nutrient–lowproductivity regions, such as the Southern Ocean (Martin et al. 1990; de Baar et al. 1995), has spawned a growing awareness that other micronutrients may also limit phytoplankton populations, either selectively or broadly. Vitamin B12 may be one such keystone micronutrient whose availability determines which major taxa dominate a microalgal community. Decades ago, investigations revealed that many eukaryotic phytoplankton species are Bvitamin auxotrophs, i.e., unable to biosynthesize one or more B vitamins (e.g., Droop 1957; Provasoli 1963; Carlucci and Silbernagel 1969). Revisiting the cultivation literature, Croft et al. (2005, 2006) reported that a high proportion of diatom (59%) and dinoflagellate (89%) species in contemporary culture collections are demonstrably B12 auxotrophs and require extracellular vitamins to grow. These observations suggest that growth of many algal taxa can be limited by availability of extracellular B12. The opposing view that dissolved cobalamin seldom falls to growth-limiting concentrations in the ocean has also been expressed on the basis of early laboratory studies (see 1 Corresponding author ([email protected]).