APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Aug. 1994, p. 2677-2683 0099-2240/94/$04.00+0 Copyright ? 1994, American Society for Microbiology Vol. 60, No.8 Accumulation of Selenium in a Model Freshwater Microbial Food Web ROBERT W. SANDERS1* AND CYNTHIA C. GILMOUR2 Academy of Natural Sciences, Patrick Center for Environmental Research, Philadelphia, Pennsylvania 19103,1 and Academy of Natural Sciences, Benedict Estuarine Research Laboratory, Benedict, Maryland 206122 Received 3 March 1994/Accepted 15 May 1994 The transfer of selenium between bacteria and the ciliated protozoan, Paramecium putrinum, was examined in laboratory cultures. The population growth of the ciliate was not inhibited in the presence of the highest concentrations of dissolved selenite or selenate tested (1~ JJ.g liter-I). Experiments with radioactive 75selenite or 75selenate indicated that accumulation of selenium by ciliates through time was low when feeding and metabolism were reduced by incubating at O?C. However, selenium accumulated in ciliate biomass during incubation with dissolved 75Se and bacteria at 24?C and also when bacteria prelabeled with 75Se were otfered as food in the absence of dissolved selenium. When 75Se-labeled bacterial food was diluted by the addition of nonradioactive bacteria, the amount of selenite and selenate in ciliates decreased over time, indicating depuration by the ciliates. In longer-term (>S-day) fed-batch incubations with 75selenite-labeled bacteria, the selenium concentration in ciliates equilibrated at approximately 1.4 JJ.g of Se g (dry weight)-I. The selenium content of ciliates was similar to that of their bacterial food on a dry-weight basis. These data indicate that selenium uptake by this ciliate occurred primarily during feeding and that biomagnification of selenium did not occur in this simple food chain. Selenium is a nonmetallic element found in a variety of chemical forms in both marine and freshwaters. Selenate (Se VI) and selenite (Se IV), the common inorganic forms in natural waters, occur together with a poorly understood suite of Se-containing organic compounds. Selenium is required in trace amounts by organisms ranging from bacteria to phyto? plankton to mammals. However, as a consequence of natural, agricultural, or industrial processes, selenium can reach levels of parts per billion and lead to death or reproductive failure of fish and waterfowl (14, 20, 22). The deleterious effects of selenium are attributed to body burdens accumulated from food or via uptake of dissolved selenium (19, 26). The degree to which selenium is transferred through the food web and the mechanisms of transport are not well understood. However, it is likely that microorganisms contrib? ute substantially to these processes. Bacteria are important in the transformation between the organic and inorganic forms of selenium (7, 10,23, 24), and both bacteria and phytoplankton readily take up dissolved selenite, selenate, and organic sele? nium compounds via active and passive mechanisms (12, 30, 39, 42). Microorganisms incorporate selenium into amino acids and other macromolecules (9, 36), and reversible adsorption may also result in the accumulation of selenium (30). Regard? less of the mechanism of microbial uptake, once selenium is associated with bacteria or algae, it is potentially available to bacterivorous and algivorous predators. Heterotrophic protists (protozoa) are important predators of both bacteria and phytoplankton in aquatic ecosystems (29, 31, 35, 40) and in turn are ingested by common freshwater zooplankton (33). Consequently, as intermediaries in aquatic food webs, hetero? trophic protists may affect the transfer of selenium. The major objectives of this study were to determine if selenium was accumulated by a bacterivorous protist feeding * Corresponding author. Mailing address: Academy of Natural Sciences, 1900 Benjamin Franklin Parkway, Philadelphia, PA 19103? 1195. Phone: (215) 299-1003. Fax: (215) 299-1028. on bacteria and, if accumulated, whether it was bioconcen? trated to levels exceeding those in the bacterial food and the water. We chose Paramecium putrinum as an experimental organism because ciliates in this genus are common in fresh? waters and have previously been used as models for a variety of ecological and physiological processes. Additionally, prelimi? nary tests indicated a tolerance of selenium by P. putrinum, their size allowed for efficient separation from bacterial prey, and their high feeding rates allowed for the accumulation of measurable amounts of radioactive selenium during short-term feeding experiments. MATERIALS AND METHODS Cultures and media. The ciliate P. putrinum (CCAP 1660/ 14) from the Culture Collection of Algae and Protozoa (Cum? bria, United Kingdom) was maintained in a 0.1 % Cerophyll ryegrass broth (media 802 [2]) with mixed bacteria as a food source. The ciliates were approximately 75 f.Lm in length and had an average calculated biovolume of 1.4 x 105 f.Lm-3. For experiments in which bacteria and protozoa were added sep? arately, unidentified bacteria from the Paramecium culture were grown in 0.25% Cerophyll broth or 0.2% yeast extract; cultures of the bacteria Pasteurella sp. (IHP#3 strain) (32) and Corynebacterium sp. were maintained on 0.2% yeast extract. All experiments were run in a synthetic growth medium (30) that otherwise simulated the major ion composition of a selenium-impacted lake (Hyco Lake, N.C.). The Hyco medium had the following composition (in milligrams per liter of distilled water): CaCI2 ? H20, 32.2; MgS04 ? 7H20, 36.9; NaHC03, 25.2; K2HP04 , 8.7; NaN03, 85; Na2Si03 ? 9H20, 28.4; KBr, 0.12; NaF, 0.4; FeEDTA, 4.3; CuS04 ? 5H20, 0.01; ZnS04 ? 7H20, 0.022; CoCl2 ? 6H20, 0.01; MnCl2 ? 4H20, 0.18; Na2Mo04 ? 2H20, 0.006; H3B03 , 0.62. HEPES (N-2-hydroxy? ethylpiperazine-N'-2-ethanesulfonic acid) buffer was used at 2.5 mM, and the pH was adjusted to 7.0 with NaOH. Selenium toxicity. Chronic toxicity of selenium to P. putri? num was examined by comparing population growth in con- 2677 2678 SANDERS AND GILMOUR APPL. ENVIRON. MICROBIOL. TABLE 1. Parameters of growth and preparation of 75selenium-Iabeled Pasteurella sp. for ciliate feeding experiments shown in Fig. 3 Final Selenium accumulation by bacteriaSelenium addition to bacterial Dissolved selenium carryover in Treatment bacterial cultures density ciliate cultures spiked with (fJ.g liter-I) (cells ng cell- I fJ.g g (dry wt)-I labeled bacteria (fJ.g liter-I) ml- I) Selenite 10 2.6 X 107 8.2 X 10-to 2.33 0.004 Selenate 50 2.6 X 107 9.0 X 10-to 2.58 0.011 troIs with growth in a series of dissolved selenite or selenate concentrations. Selenite and selenate (Sigma Chemical Co., St. Louis, Mo.) were stored dry at room temperature until just prior to the start of the experiments, when stock solutions of selenite and selenate were made up in sterile Hyco medium. Stock solutions were made to take into account the volume of medium added with the microorganism inocula to give final concentrations of 10, 100, and 1,000 IJ-g of selenite or selenate liter-I. Paramecium and bacterial cultures were centrifuged at 425 x g for 30 min and 8,600 x g for 15 min, respectively, to separate the microorganisms from their growth medium. The pelleted P. putrinum and bacteria were resuspended in sterile Hyco medium, enumerated, and added to the solutions of selenium. Three replicates were run for controls and each concentration of selenite and selenate. The controls were identical to the treatment flasks, except selenium was not added. Samples were taken from each replicate and preserved with 2% Lugol's iodine (ciliates) or 10% formalin (bacteria) at approximately 24-h intervals for a period of 120 h, at which time the ciliates had reached stationary growth phase in all treatments. Bacterial samples were enumerated by epifluores? cence microscopy (16). Ciliates from preserved samples were enumerated with phase-contrast microscopy at X 200 magni? fication. Specific growth rates were determined from the slopes of lines fit to In(Paramecium abundance) versus time. Bacterial uptake of 7sselenium. Uptake of selenium was determined for axenic cultures of Pasteurella sp. and Coryne? bacterium sp., using radioactive selenite and selenate (Na275Se03 and Na275Se04 ; Amersham Corp., Arlington Heights, Ill.) The specific activities of the selenite and selenate were 24 (3.29 IJ-Ci nmol- I) and 385 (205 nCi nmol- 1) ng IJ-Ci-1, respectively. Radioactive selenium was quantified with a Tennelec NaI(T1) gamma counting system equipped with a 7.6-cm well detector and multichannel analyzer. Experimental samples were calibrated against 75Se stock isotope solutions. Bacterial isolates in logarithmic growth were transferred to 0.2% yeast extract in Hyco medium, and incubations were run in duplicate at 22?C in the gresence of 10 IJ-g of selenium liter- as Na275Se03 or Na2 Se04 ? Subsamples (5 ml) were removed at intervals, filtered onto 0.2-lJ-m-pore-size membrane filters (Millipore GS), and washed with fresh Hyco medium prior to gamma counting. Growth for each species was moni? tored by using optical density (A660) previously calibrated against cell abundance and dry weight measurements. Due to an interruption in availability of the isotope, the uptake of selenate by Corynebacterium sp. was not determined. Se accumulation and depuration by P. putrinum. Gentle reverse filtration through a 5-J.Lm-mesh sieve and copious rinsing with fresh medium were used to concentrate ciliates and reduce bacterial abundance for all accumulation experi? ments. This technique removed spent culture medium and most bacteria, while avoiding injury to ciliates from compres? sion against a filter. Uptake of selenium by P. putrinum was examined in two ways. In the first type of experiment, dissolved Na275Se03 or Na275Se04 was added directly to mixed bacte- rial-ciliate cultures in Cerophyll-enriched (0.1%) Hyco me? dium. Bacterial abundances had been reduced to =105 cells ml- 1 in these cultures. Final concentrations of selenium in the cultures were 10 IJ-g liter- 1 for selenite and 50 IJ-g liter- I for selenate. For each selenium ion, duplicate samples were incubated at 24?C to examine active uptake of selenium by ciliates. Another set of samples was incubated at O?C to reduce bacterivory and the metabolic rates of both ciliates and bacte? ria. Any uptake of 75Se in the O?C cultures was assumed to be due to adsorption rather than active uptake. Samples were taken at 0, 1, 2, 4, 8.5, 20, and 27 h for gamma counting and microscopic enumeration of ciliates and bacteria. In a second type of experiment, bacteria were prelabeled with 75selenium before addition to P. putrinum in unenriched Hyco medium. Pasteurella sp. was labeled by incubation with 10 J.Lg of selenite liter-lor 50 IJ-g of selenate liter- I for 24 h during exponential growth in 0.2% yeast extract (Table 1). Dissolved 75selenium was separated from the bacteria by centrifugation and washing, and each of the 75Se-Iabeled bacterial suspensions was added to duplicate cultures of bac? teria-free ciliates in unenriched Hyco medium. Incidental transfer of dissolved selenium into the ciliate cultures was <0.02 IJ-g liter- 1 (Table 1). Cultures were maintained at 25?C in the dark, and samples for gamma counting and microscopic enumeration were taken immediately upon addition of the labeled bacteria and at 15 and 22 h. Accumulation of selenium in the ciliates was determined from the radioactivity retained on 5-lJ-m-pore-size polycarbonate filters (Nuclepore Corp.). Preliminary tests with 75Se-Iabeled bacteria indicated minimal retention of bacteria on these filters (3 to 6% retention on a 0.4-lJ-m-pore-size filter). After sampling at 22 h, bacteria that had not been exposed to selenium were added to a final density of 5 x 107cells ml- 1 . This swamped the culture with unlabeled bacterial cells, and ciliate predation was then primarily on selenium-free bacteria. Selenium depuration rates by the cili? ates were estimated from the reduction of ciliate-associated selenium during the next 23 h with the assumption that there was no further uptake during this time. In a similar experiment, P. putrinum ciliates were maintained in fed-batch culture in Hyco medium for 6 days with 75selenite? labeled Corynebacterium sp. as food. The initial addition of bacteria to the ciliates was from a Corynebacterium culture that had been incubated with 75selenite for 100 h. The selenium content of bacterial cells was 1.9 IJ-g of selenium g (dry weight) -1 and the addition represented a spike of 5.7 ng of selenium liter- I in bacterial cells. Additional 75Se-Iabeled bacteria were added 21 and 44 h after the initial inoculation. Radioactivity in the ciliates was monitored over the time course to determine if selenite reached an equilibrium concen? tration in P. putrinum. Paramecium feeding and egestion. Ingestion rates for ciliates feeding on bacteria were estimated by the disappearance of bacteria from cultures in unenriched Hyco medium over time (15), assuming that the lack of added organic substrate pre? cluded bacterial growth. In a separate experiment, Parame- VOL. 60, 1994 SELENIUM UPTAKE BY PROTOZOA 2679 a Growth rates in treatments were not significantly different from those in controls with no added selenium (P > 0.05; analysis of covariance). TABLE 2. Specific growth rates of P. putrinum during incubation with dissolved selenite and selenatea Specific daily growth rate (?SE) Time (h) 2412 186 1.0 ~------------""T8.0 0.9 A 7.0 >. 0.8 _ ~ 0.7 6.0 3= 55 0.6 5.0 ~ o 0.5 4.0 _Q)O> ~ 0.4 3.0 en 8- 00 .. 32 2.0 ~Selenate 0.1 1.0 0.0 +-- -+--+--_--+-__--+-__--+....L 0.0 o Selenate 0.48 ? 0.05 0.43 ? 0.10 0.44 ? 0.08 0.49 :?: 0.09 Selenite 0.32 ? 0.06 0.39 ? 0.02 0.39 ? 0.05 0.38 ? 0.05 o 10 100 1,000 Concn of added selenium (f.Lg liter- 1) cium egestion rates were determined by a modification of the method of Sherr et al. (34). Briefly, tracer amounts of nondi? gestible 0.9-lJ-m Fluoresbrite carboxylated microspheres (Poly? sciences, Warrington, Pa.) were added to cultures of P. putri? num and bacteria. Aliquots were fixed over a time course, and microspheres within individual ciliates were counted by epi? fluorescence microscopy. Microsphere accumulation in food vacuoles was monitored until the number of microspheres per cell plateaued. The microsphere concentration was then re? duced by dilution with a large volume of Hyco medium containing bacteria, so that microspheres were rarely encoun? tered and ingested. The time required (after dilution) for the average number of microspheres per ciliate to reach a constant background level was considered the retention time for any undigestible portion of the bacterial biomass. RESULTS 1.0 8.0 0.9 B 7.0 0.8 ~ 6.0 i ?00 0.7 c: 0.6 . 5.0 ~Q) "C0 0.5 4.0. ~co 0.4 Q)(.) 3.0 enli 0.3 0>0 2.0 =l 0.2 0.1 1.0 0.0 0.0 0 6 76 Time (h) FIG. 1. Weight-specific selenium contents and optical densities of bacterial cultures incubated with dissolved 75selenite or 75selenate (10 J.Lg liter-I). (A) Pasteurella sp. Optical density was plotted only for the selenate incubation, but optical density in the selenite incubation was nearly identical at every sampling. (B) Corynebacterium sp. 75Selenate was not available at the time of the experiment. FIG. 2. Selenium contents of P. putrinum (means ? 1 standard error) during incubation at 0 and 24?C in the presence of dissolved 75selenite (10 J.Lg liter-I) or 75selenate (50 J.Lg liter-I). Live bacteria were present in all incubations. 246 12 1 8 Time (hours) ~ ---tr- 1j 5 ~ -0- ~ 4 "E 3Q) "E8 2 E .~ c:: "* 0 ~3E~E=:::;:::;~~=::;==::;===;=3..........--~ en 0 -"i 6 -r-----S-e-le-n-ite-O-oC---r-------r-------, Selenite 24?C Selenate O?C Selenate 24?C ciliates equilibrated after 2 to 4 h at approximately 0.1 and 0.2 IJ-g of Se g (dry weight) -1 for selenite and selenate, respectively (Fig. 2). In the incubations at 24?C, Paramecium abundances remained constant (approximately 200 ciliates ml- 1), but bacteria increased from an initial density of 1.3 x 105 to 1.5 X 107 cells ml- 1 after 20 h. Uptake of selenium by ciliates at 24?C was initially similar to uptake at O?C (Fig. 2). However, after 8.5 h, concentrations of selenium in ciliates from 24?C incuba? tions exceeded those from O?C incubations. At 20 h, concen? trations of selenite and selenate in ciliates from the 24?C Toxicity. Neither acute nor chronic toxicity was observed for P. putrinum exposed to dissolved selenite and selenate at concentrations of up to 1,000 IJ-g of Se liter-I. Microscopically observed swimming behavior was normal in all selenium treatments, and ciliate population growth was at least as high in selenium-exposed cultures as in unexposed controls (Table 2). Differences in ciliate growth between treatments and the controls were not statistically significant (Table 2) (P > 0.05; analysis of covariance). The growth of Pasteurella sp. and Corynebacterium sp. in the presence of 10 IJ-g of Se liter- 1 (Fig. 1) was similar to growth in the absence of added selenium (data not shown). Bacterial uptake of See Accumulation of selenium by Pas? teurella sp. occurred in two phases. There was an initial rapid and comparable uptake in the presence of both selenite and selenate during the first hour, after which accumulation con? tinued, but at lower rates (Fig. 1A). Selenate uptake was not as rapid as cell growth, so that the selenate concentration per unit biomass decreased after the first hour of incubation (Fig. 1A). The relatively constant amount of selenate (=0.5 IJ-g of Se g [dry weight] -1) associated with Pasteurella sp. during the final portion of the growth phase indicated that uptake matched growth. Accumulation of selenite by Pasteurella sp. was more rapid than cell growth and about 10 times more rapid than selenate uptake. Corynebacterium sp. accumulated selenite throughout the growth phase. After 8 h, selenite reached a concentration (7.6 IJ-g of Se g [dry weight]-I) that was more than double that observed for Pasteurella sp. after 8 h of incubation (2.9 IJ-g of Se g [dry weight]-I) (Fig. 1). Accumulation of dissolved selenium by P. putrinum. Incuba? tion at O?C was used as an indicator of nonactive uptake (Le., by adsorption and diffusion) due to low metabolisms of ciliates and bacteria at this temperature. Ciliate and bacterial abun? dances remained constant at this temperature, and sorption by 2680 SANDERS AND GILMOUR APPL. ENVIRON. MICROBIOL. 144 1000 900 800 Q)(.) c_ 700 cu- u E 600 c CD::J .J:l a.500 ? 400 Q) ci -z 300 Jg- (3 200 100 144 120 Bacteria 96724824 A 1.5------------------., Toxicity of selenium to microorganisms. Selenium toxicity has been determined for only a few heterotrophic protozoa. A concentration of 10,000 f.1g of selenite liter-1 (the lowest concentration tested) inhibited the growth of the heterotrophic dinoflagellate, Crypthecodinium cohnii (28). Growth of another heterotrophic flagellate, Entosiphon sulcatum, was affected in the presence of only 3 f.1g of selenite liter-1 (4). The addition of 20 to 160 f.1g of selenite liter-1 to freshwater laboratory microcosms had no effect on protozoan biomass accumulation but did result in reduced protozoan diversity after 21 days of exposure (27). In outdoor experimental streams, protozoan species richness was unaffected relative to controls after 10 longer-term experiment in which ciliates were fed 75selenite? labeled Corynebacterium sp. daily for 3 days, P. putrinum accumulated selenium from ingested bacteria for a period of 6 days (Fig. 4A). The selenium content of the Corynebacterium sp. in this experiment was 1.86 f.1g of Se g (dry weight)-1. Selenium content of the ciliates reached 0.81 f.1g of Se g (dry weight) -1 within 24 h and equilibrated at approximately 1.35 f.1g of Se g (dry weight) -1 after 4 days. Egestion and digestion rates of P. putrinum. The loss of selenium from ciliates ingesting Se-Iabeled bacteria may have been due to excretion of dissolved selenium or from egestion of selenium-laced bacterial cell debris from the ciliates' diges? tive vacuoles. Fluorescent microspheres were used as nondi? gestible visual tracers of egestion by the ciliates to quantify the time scale of potential losses due to egestion of incompletely digested bacteria. Disappearance of microspheres from ciliates indicated that, at 25?C, any undigested material in vacuoles would be egested within 1 h of ingestion. Thus, losses of selenium with bacterial cell debris would have occurred soon after the addition of unlabeled bacteria (arrows in Fig. 3). DISCUSSION 24 48 72 96 120 Time (hours) FIG. 4. (A) Selenite content of P. putrinum versus time in a fed-batch culture of 75selenite-Iabeled bacteria. (B) Abundances of P. putrinum and bacteria in the fed-batch cultures. 50 1400 1200 E 1000 Ci5 a. 800 0 600 ~en 0> 400 ~ 200 C3 0 50 1400 1200 E 1000 Ci5 a. 800 0 ~ 600 en Q) 400 ~ 200 C3 30 40 -0- Bacteria ~ Paramecium 30 40 ___ Bacteria -.- Paramecium 2010 60-r-----------r-----., ~ A -0- Selenate ? ~ 50---+- Selenite ?c C3 ~.c 40 C/)~ co ~ 30 .~ "C 0> 0> ~ ~ 20 m g '#- ~ 10 ? 55 B Selenite 0> -g E 45 as "- "co> C a.~- g E 45 as "- "C 0> C a. ~60%; Fig. 3) and the transfer of bacterium-associated selenium into Paramecium biomass (:535%) suggest that much of the sele? nium ingested by the ciliates was not assimilated. In the absence of dissolved selenium, selenium per unit weight actu? ally decreased in P. putrinum relative to its content in the bacterial food sources. The maximum concentration of sele? nium in P. putrinum during the 2-day experiment (Fig. 3) was <0.9 f..Lg of Se g (dry weight)-1 compared with approximately 2.6 f..Lg of Se g (dry weight)-1 in the bacteria (Table 1). Incorporation efficiency of selenium during the first 15 h of this experiment ranged from 28 to 34% as calculated from the total amount ingested with bacterial food and the selenium content of ciliates at the end of that time. In the experiment in which 75selenite-Iabeled bacteria were replenished at regular inter? vals over 3 days, selenite continued to increase in P. putrinum until concentration per unit weight equilibrated 2 days after the final addition of Se-Iabeled bacteria (Fig. 4). The concen? tration of selenite in this bacterial food (Corynebacterium sp.) was 1.9 f..Lg of Se g (dry weight)-t, and selenite in P. putrinum reached equilibrium at about 1.4 f..Lg of Se g (dry weight)-I. It appeared that selenium was not biomagnified in ciliates feed? ing on bacteria in these experiments. Rather, the equilibrium concentration of selenium in the ciliates was lower than, but generally reflected, the concentration in their food. 2682 SANDERS AND GILMOUR APPL. ENVIRON. MICROBIOL. The loss of selenium ingested by ciliates was calculated during two phases of the pulse-chase experiment (Fig. 3). The total amounts of selenite and selenate ingested per ciliate during the first 15 h of this experiment (calculated from the selenium content of bacteria and their disappearance over time) were 44.1 and 56.9 fg, respectively. Subtracting the selenium content of ciliates at 15 h from the total amount ingested, the maximum loss rates during that time were 1.9 (0.102 J.Lg of Se g [dry weight]-1 h-1) and 2.7 (0.144 J.Lg of Se g [dry weight] -1 h-1) fg of selenate ciliate-1 h-1. These rates were considerably greater than the depuration (0.022 J.Lg of Se g [dry weight] -] h-1) determined later in the same experiment from the decrease in 75Se associated with ciliate biomass after ingestion of 75Se-Iabeled bacteria was minimized. A similar pattern of selenite loss from P. putrinum (calculated by sub? tracting accumulated selenite from ingested selenite) was observed in the 6-day experiment with labeled Corynebacterium sp. (Fig. 4). A loss of 0.243 J.Lg of Se g (dry weight)-l h- 1 was determined from samples taken early in the experiment when large numbers of bacteria were ingested. A much lower depuration rate (0.050 J.Lg of Se g [dry weight]-l h- 1) was calculated for the period between days 2 and 4 when labeled bacteria had been reduced by grazing and ciliate ingestion rates were lower. These results could reflect a direct corre? spondence between the amount of selenium ingested and the amount excreted. However, it is also probable that a propor? tion of the selenium loss was from defecation of partially digested bacteria, i.e., that all of the ingested selenium was not assimilated. Egestion of any selenium-laced bacterial cell de? bris from the digestive vacuoles would be much less as labeled bacteria were depleted in the later stages of these experiments. Consequently, the lower rates of selenium loss reported here are probably more representative of the depuration of assim? ilated material. General considerations of selenium in the microbial food web. Selenium is incorporated into amino acids, and its trans? fer and depuration in the microbial food web may be analo? gous to those of nitrogen in some ways. Food web transfer and remineralization of nitrogen and other nutrients by protozoa can be predicted from the carbon/nutrient ratios of the pred? ator and prey if the respiration and gross growth efficiency are known (6). However, further study is needed before this model can be adapted to selenium. The degree to which selenite and selenate were biotransformed during accumulation is unknown for any of the experiments of the present study. Additionally, several factors may differentiate incorporation of selenium from that of nitrogen. These include the high proportion of selenium that binds abiotically to bacteria and phytoplankton food and the possibility that some seleno-amino acids synthe? sized by algae cannot be incorporated into proteins (11). On the bases of this study, our related unpublished field work, and our colleagues' work with phytoplankton (30), our working hypotheses are that most selenium accumulation occurs at the base of the food web and selenium concentration in predators generally reflects the concentration in their prey. Because bacteria are not grazed efficiently by most metazoans, passage of selenium in the aquatic food web may be primarily via phytoplankton and protozoa because they are ingested at high rates by zooplankton and benthic microfauna (11, 33) and also occur as epibionts. Zooplankton and microfauna, in turn, are irigested by fish and waterfowl. Consequently, factors that determine the rate and amount of selenium accumulation in microorganisms may in large part determine the eventual accumulation of selenium in predators such as fish, whose reproductive cycles may be strongly affected at selenium levels that are not toxic to many microorganisms. ACKNOWLEDGMENTS This work was supported by the Electric Power Research Institute under project RP2020-11 and in part by the Environmental Associates of the Academy of Natural Sciences. We thank J. G. Sanders, G. F. Riedel, and several anonymous reviewers for helpful comments on the manuscript, G. F. Riedel for assisting with gamma counting, and R. Aoki, G. S. Riedel, K. Gloersen, and L. Cole for technical assistance. REFERENCES 1. Allen, R. D. 1984. Paramecium phagosome membrane: from oral region to cytoproct and back again. J. Protozool. 31:1-6. 2. American Type Culture Collection. 1991. Catalogue of protists, 17th ed. American Type Culture Collection, Rockville, Md. 3. Besser, J. M., J. N. Huckins, E. E. Little, and T. W. La Point. 1989. Distribution and bioaccumulation of selenium in aquatic micro? cosms. Environ. Pollut. 62:1-12. 4. Bringmann, G., and R. Kuhn. 1980. Comparison of the toxicity thresholds of water pollutants to bacteria, algae, and protozoa in the cell multiplication inhibition test. Water Res. 14:231-241. 5. Burton, G. A., T. H. Giddings, P. DeBrine, and R. Fall. 1987. High incidence of selenite-resistant bacteria. from a site polluted with selenium. Appl. Environ. Microbiol. 53:185-188. 6. Caron, D. A., and J. C. Goldman. 1988. Dynamics of protistan carbon and nutrient cycling. J. Protozool. 35:247-249. 7. Cooke, T. D., and K. W. Bruland. 1987. Aquatic chemistry of selenium: evidence of biomethylation. Environ. Sci. Technol. 21:1214-1219. 8. Cutter, G. A. 19890; Freshwater systems, p. 243-262. In M. Ihnat (ed.), Occurrence and distribution of selenium. CRC Press, Boca Raton, Fla. 9. Cutter, G. A., and K. W. Bruland. 1984. The marine biogeochem? istry of selenium: are-evaluation. Limnol. Oceanogr. 29:1179? 1192. 10. Doran, J. W., and M. Alexander. 1977. Microbial transformations of selenium. Appl. Environ. Microbiol. 33:31-37. 11. Fisher, N. S., and J. R. Reinfelder. 1991. Assimilation of selenium in the marine copepod Acartia tonsa studied with a radiotracer method. Mar. Ecol. Prog. Sere 70:157-164. 12. Foda, A., J. H. Vandermuelen, and J. J. Wrench. 1983. Uptake and conversion of selenium by a marine bacterium. Can. J. Fish. Aquat. Sci. 4O(Suppl. 2):215-220. 13. Foe, C., and A. W. Knight. 1986. Selenium bioaccumulation, regulation, and toxicity in the green alga, Selenastrum capricomu? tum, and dietary toxicity of contaminated alga to Daphnia magna, p. 77-88. In J. Slocum (ed.), Selenium in the environment. California State University, Fresno. 14. Gillespie, R. B., and P. C. Baumann. 1986. Effects of high tissue concentration of selenium on reproduction by bluegills. Trans. Am. Fish. Soc. 115:208-213. 14a.Gilmour, C. C. Unpublished data. 15. Heinbokel, J. F. 1978. Studies of the functional role of tintinnids in the Southern California Bight. I. Grazing and growth rates in laboratory cultures. Mar. BioI. 47:177-189. 16. Hobbie, J. E., R. J. Daley, and S. Jaspar. 1977. Use of Nuclepore filters for counting bacteria by fluorescence microscopy. Appl. Environ. Microbiol. 33:1225-1228. 17. Hudman, J. F., and A. R. Glenn. 1984. Selenite uptake and incorporation by Selenomonas ruminantium. Arch. Mikrobiol. 140:252-256. 18. Karlson, U., and W. T. Frankenberger, Jr. 1988. Effects of carbon and trace element addition on alkylselenide production by soil. Soil Sci. Soc. Am. J. 52:1640-1644. 19. Lemly, A. D., and G. J. Smith. 1987. Aquatic cycling of selenium? implications for fish and wildlife. Leaflet no. 12. U.S. Fish and Wildlife Service, Washington, D.C. 20. Marshall, E. 1985. Selenium poisons refuge, California politics. Science 229:144-146. 21. Nilsson, J. R. 1979. Phagotrophy in Tetrahymena, p. 339-379. In M. Levandowsky and S. H. Hutner (ed.), Biochemistry and physiology of protozoa. Academic Press, Inc., New York. 22. Ohlendorf, H. M.1989. Bioaccumulation and effects of selenium in wildlife, p. 133-177. In L. W. Jacobs (ed.), Selenium in agriculture VOL. 60, 1994 SELENIUM UPTAKE BY PROTOZOA 2683 and the environment. Soil Science Society of America, Madison, Wis. 23. Oremland, R. S., J. T. Hollibaugh, A. S. Maest, T. S. Presser, L. G. Miller, and C. W. Culbertson. 1989. Selenate reduction to elemen? tal selenium by anaerobic bacteria in sediments and culture: biogeochemical significance of a novel sulfate-independent respi? ration. Appl. Environ. Microbiol. 55:2333-2343. 24. Oremland, R. S., and J. P. Zehr. 1986. Formation of methane and carbon dioxide from dimethylselenide in anoxic sediments and by a methanogenic bacterium. Appl. Environ. Microbiol. 52:1031? 1036. 25. Patrick, R. P. 1978. Effects of trace metals in the aquatic ecosys? tem. Am. Sci. 66:185-191. 26. Porcella, D. B., G. L. Bowie, J. G. Sanders, and G. A. Cutter. 1991. Assessing Se cycling and toxicity in aquatic ecosystems. Water Air Soil Pollut. 57/58:3-11. 27. Pratt, J. R., and N. J. Bowers. 1990. Effect of selenium on microbial communities in laboratory microcosms and outdoor streams. Toxicity Assessment Int. J. 5:293-307. 28. Prevot, P., and M.-O. Soyer-Gobillard. 1986. Combined action of cadmium and selenium on two marine dinoflagellates in culture, Prorocentrum micans Ehrbg. and Crypthecodinium cohnii Biec? heler. J. Protozool. 33:42--47. 29. Rassoulzadegan, F., and R. W. Sheldon. 1986. Predator-prey interactions of nanozooplankton and bacteria in an oligotrophic marine environment. Limnol. Oceanogr. 31:1010-1021. 30. Riedel, G. F., D. P. Ferrier, and J. G. Sanders. 1991. Uptake of selenium by freshwater phytoplankton. Water Air Soil Pollut. 57/58:23-30. 31. Sanders, R. W., K. G. Porter, S. J. Bennett, and A. E. DeBiase. 1989. Seasonal patterns of bacterivory by flagellates, ciliates, rotifers, and cladocerans in a freshwater planktonic community. Limnol. Oceanogr. 34:673-687. 32. Sanders, R. W., K. G. Porter, and D. A. Caron. 1990. Relationship between phototrophy and phagotrophy in the mixotrophic chryso? phyte Poterioochromonas malhamensis. Microb. Ecol. 19:97-109. 33. Sanders, R. W., and S. A. Wickham. 1993. Planktonic protozoa and metazoa: predation, food quality and population control. Mar. Microb. Food Webs 7:197-223. 34. Sherr, B. F., E. B. Sherr, and F. Rassoulzadegan. 1988. Rates of digestion of bacteria by marine phagotrophic protozoa: tempera? ture dependence. Appl. Environ. Microbiol. 54:1091-1095. 35. Smetacek, V. 1981. The annual cycle of protozooplankton in the Kiel Bight. Mar. BioI. 63: 1-11. 36. Stadtman, T. C. 1974. Selenium biochemistry. Science 183:915? 921. 37. Steinberg, N. A., and R. S. Oremland. 1990. Dissimilatory selenate reduction potentials in a diversity of sediment types. Appl. Envi? ron. Microbiol. 56:3550-3557. 38. Thompson-Eagle, E. T., W. T. Frankenberger, Jr., and U. Karlson. 1989. Volatilization of selenium by Altemaria altemata. AppI. Environ. Microbiol. 55: 1406-1413. 39. Vandermuelen, J. H., and A. Foda. 1988. Cycling of selenite and selenate in marine phytoplankton. Mar. BioI. 49:231-236. 40. Weisse, T., H. Muller, R. M. Pinto-Coelho, A. Schweizer, D. Springmann, and G. Baldringer. 1990. Response of the microbial loop to the phytoplankton spring bloom in a large prealpine lake. Limnol. Oceanogr. 35:781-794. 41. Wheeler, A. E., R. A. Zingaro, and K. Irgolic. 1982. The effect of selenate, selenite and sulfate on six species of unicellular algae. J. Exp. Mar. BioI. EcoI. 57:181-194. 42. Wrench, J. J. 1978. Selenium metabolism in the marine phyto? plankters Tetraselmis tetrahele and Dunaliella minuta. Mar. BioI. 49:231-236. 43. Wrench, J. J., and C. I. Measures. 1982. Temporal variation in dissolved selenium in a coastal ecosystem. Nature (London) 299: 431--433.