Experimental studies on the accumulation of polonium-210 by marine phytoplankton

Bioaccumulation of polonium-210 (t1/2 5 138 d) in marine phytoplankton can introduce this naturally occurring radioisotope into food chains, where it accounts for most of the radiation dose to marine organisms and to human consumers of seafood. Moreover, this isotope could be useful as a tracer of the flux of organic matter in ocean surface waters. We performed laboratory experiments with eight algal species representing six algal divisions to quantify the uptake, cellular partitioning, and retention of 210Po by algae. Biological uptake was unaffected by temperature or light, and volume concentration factors (VCFs) for these algal species ranged between 0.5 and 3.0 3 104. Interspecific differences in VCFs could be explained by considering the surface area to volume ratios of the cells and cellular protein content. Once associated with the cells, between 30 and 60% of the total cellular 210Po was in the cytoplasm of the different species. 210Po was not irreversibly bound to the cells but displayed a biological half-life of ;23 d. Because 210Po associates appreciably with organic matter inside cells, unlike other particlereactive nonessential metals, it could have promise as a tracer of sinking organic matter in the ocean. Polonium-210 is both a major source of natural radiation to marine organisms as well as a potential chemical tracer for organic matter sinking out of ocean surface waters. In order to better evaluate the usefulness of 210Po as an organic matter tracer and to comprehend its bioconcentration in marine food webs, the association of this element with marine phytoplankton needs to be better understood. There have been few controlled experiments in which the uptake of 210Po by plankton, in particular phytoplankton, has been examined (Fisher et al. 1983b). Studies on the partitioning of 210Po between particulate and dissolved phases have shown that this element, like Pb, is very particle-reactive, with reported Kd values for various types of particles in the range of 103– 105 (Carvalho 1997; Nozaki et al. 1997; Hong et al. 1999). Results of earlier studies suggest that 210Po also becomes incorporated into the cytoplasm of some species of phytoplankton (Fisher et al. 1983b) and bacteria (Cherrier et al. 1995; LaRock et al. 1996), where, unlike 210Pb, its partitioning is similar to that of protein and sulfur within the cell. Thus, there is evidence that potentially two pools of 210Po are associated with cells: a surface-bound pool and an internal pool. The polonium in the internal pool’s association with cytoplasm and protein could result in efficient assimilation of 210Po by grazing zooplankton (Reinfelder and Fisher 1991) that, in turn, could lead to its bioconcentration in marine food webs. The external pool, in contrast, could be expected to behave more like other surface-bound particle-reactive radionuclides such as 210Pb and 234Th, which are not assimilated in animal tissue and sink in fecal material (Fisher and Reinfelder 1995). The final alpha-emitting product of the radioactive decay of 238U, 210Po (t1/2 5 138 d), is produced by the beta decay 1 Corresponding author ([email protected]). Acknowledgments We thank R. Armstrong, S. Baines, K. Cochran, S. Hook, P. Masque, and two anonymous reviewers for helpful comments on the manuscript. This research was supported by NSF OPP-9986069 and by an NDSEG fellowship to G.M.S. This is MSRC contribution 1252. of 210Pb (t1/2 5 22 yr) via the short-lived intermediate 210Bi (t1/2 5 5 d). 210Pb is delivered to the ocean largely by atmospheric deposition (Shannon et al. 1970; Carvalho 1997); thus, its input into the ocean is not constant. Most 210Po in the ocean is produced by the in situ decay of 210Pb, although point sources near phosphate ore–processing operations and seamount volcanoes could contribute to higher polonium concentrations regionally (LaRock et al. 1996; Rubin 1997). Polonium is found in subtrace concentrations in surface seawater, on the order of 10220 mol L21 or 0.3–3.0 mBq L21 (Cochran et al. 1983; Hong et al. 1999). The vertical profile of dissolved polonium is like that of nutrient elements: the concentration is lowest at the surface, despite the atmospheric source from wet and dry deposition, and rises slightly with depth (Cochran et al. 1983). 210Po is quickly scavenged out of the surface layer, leading to estimated residence times in surface waters of approximately 0.6 yr (Bacon et al. 1976; Cochran et al. 1983), a value that is about half its residence time in deeper waters (Kadko et al. 1987). 210Po concentrates appreciably in marine animals (Heyraud and Cherry 1979; Bulman et al. 1995; Dahlgaard 1996; Durand et al. 1999), providing the largest radiation doses to aquatic organisms. The concentration of 210Po is $148 mBq g21 in phytoplankton (Cherry 1964), 3,145 mBq g21 in shrimp hepatopancreas (Cherry and Heyraud 1981), #700 and #1,026 mBq g21 in mussel soft tissues (Germain et al. 1995) and hepatopancreas (Stepnowski and Skwarzec 2000), respectively, and #262 mBq g21 in fish (pyloric caecum, Clulow et al. 1998). In contrast, humans have an average 210Po concentration of 0.148 mBq g21, yielding a dose that is up to two orders of magnitude lower than those received by marine animals (Cherry 1964). In fact, studies in the Irish Sea within the path of the discharges from the Sellafield nuclear fuel reprocessing plant have found higher doses to organisms from 210Po than from anthropogenic radionuclides (Pentreath and Allington 1988). Besides the exposure to the animals themselves, the highly enriched concentrations of 210Po in marine organisms can contribute significantly to human radioactivity exposure through seafood consumption (Bulman et al. 1995; Dahlgaard 1996; Carvalho 1997). Un1194 Stewart and Fisher derstanding the biological interactions of polonium is thus important in risk assessments that consider the health of ecosystems and public health. 210Po has been used to trace particle transfer (Radakovitch et al. 1999), ocean circulation (Moore and Smith 1986; Carvalho 1997), and vertical flux of particulate matter in the ocean (Nozaki et al. 1997, 1998; Friedrich and Rutgers van der Loeff 2002). Like uranium and thorium, radioactive lead and polonium should come to secular equilibrium, in this case on a time scale of roughly 2 yr (assuming steady state), but the two are often in disequilibrium in the surface layer of the ocean (Cochran et al. 1983; Nozaki et al. 1997, 1998). In fact in the upper ocean, the average dissolved 210Po/210Pb activity ratio is near 0.5 (Hong et al. 1999), implying that 210Po is preferentially removed from surface waters because of the different affinities of these elements for particulate matter. Nozaki et al. (1998) found a strong correlation between the dissolved 210Po removal rate constant and chlorophyll a concentration, but no similar correlation for 210Pb, suggesting a link between phytoplankton abundance and 210Po removal. Here, we present results of a study to assess more fully the uptake and distribution of 210Po in marine phytoplankton. We compared the concentration of 210Po by phytoplankton species that differed in size, surface characteristics, and protein content to investigate the relationship between these parameters and 210Po uptake. We also examined the reversibility of the 210Po-phytoplankton associations to further examine the consequences of the partitioning of polonium in the cells. Understanding the relationship between 210Po and various cellular components can provide a mechanistic basis for evaluating differences in the flux of this radionuclide, its possible concentration in higher trophic levels, and its potential use as a tracer of organic carbon flux in the sea. Materials and methods 210Po uptake by eight marine algal species was determined in a series of laboratory experiments. The species studied were the centric diatom Thalassiosira pseudonana (clone 3H), the chlorophytes Dunaliella tertiolecta (DUN) and Chlorella autotrophica (CCMP 243), the coccolithophore Emiliania huxleyi (CCMP 2112), the prymnesiophyte Isochrysis galbana (ISO), the cryptophyte Rhodomonas salina (CCMP 1319), the prasinophyte Tetraselmis levis (PLATY 1), and the dinoflagellate Heterocapsa triquetra (OB 21019305). All inocula came from axenic stocks; cultures were handled aseptically throughout the experiments. Cultures were maintained in sterile-filtered (0.2 mm Nuclepore polycarbonate membrane) surface seawater collected 8 km off Southampton, New York, and enriched with f/2 nutrients (Guillard and Ryther 1962). Experimental inocula were harvested either by resuspension off Nuclepore membranes or by centrifugation at 3,000 3 g (R. salina and I. galbana) from stock cultures in late log phase and grown in sterile 200-ml ground glass-stoppered erlenmeyer flasks, each containing 100 ml of filtered growth medium. Control treatments were identical to algal cultures, except the growth medium remained uninoculated. We also measured the binding of 210Po to sterilized acid-washed glass beads (5–15 mm, median diameter 10 mm) to assess the adsorption of 210Po onto inorganic mineral surfaces. Suspensions of the glass beads were treated identically to the algal cultures. All treatments were run in triplicate flasks and cell counts were made daily. Cultures were incubated at 158C under cool white fluorescent lamps producing about 100 mmol quanta m22 s21 at the culture surface (14 : 10 light : dark [LD]). The initial biomass (dry wt) in the cultures ranged from 100 to 500 mg C L21, and growth was monitored periodically throughout the experiments using a Coulter Counter (Multisizer II) and checked microscopically with a hemacytometer. Cell volumes, also measured using the Coulter Counter, ranged between 26 (C. autotrophica) and 2,693 mm3 (H. triquetra) (Table 1). Surface areas, calculated using microscopy and geometric formulas, ranged from 42 to 979 mm2. Surface area to volume ratios ranged between 0.75 and 2.0 mm21 (Table 1). Dimensions of the glass beads fell in the middle of all ranges and were very similar to the values for T. pseu-