Toxic Cyanobacteria and Plant Defenses Against a Generalist Herbivore by Michael F. Chislock A thesis submitted to the Graduate Faculty of Auburn University in partial fulfillment of the requirements for the Degree of Master of Science Auburn, Alabama August 6, 2011 Keywords: Eutrophication, harmful algal blooms, toxic cyanobacteria, chemical defenses, arms race, Daphnia Copyright 2011 by Michael F. Chislock Approved by Alan Wilson, Chair, Assistant Professor of Fisheries and Allied Aquacultures Dennis DeVries, Professor of Fisheries and Allied Aquacultures James Stoeckel, Assistant Professor of Fisheries and Allied Aquacultures ii Acknowledgments First and foremost, I would like to thank my family for their love, support, and encouragement. I don?t get to see you as often as I?d like, but I?d do anything in the world for y?all. I?d also like to thank all of my friends for keeping me sane ? life as a graduate student certainly wasn?t easy, and I owe y?all more than I?ll ever be able to give back. Thank you to Susan Smith, Karen Booker, Jeanie Harry, Carolyn Jones (?C.J.?), T. Speir, and the rest of the folks working in the Fisheries main office ? y?all are awesome!! Thank you so much for fixing all of my mistakes ? I appreciate your time and efforts, and I?ll try not to cause too many headaches as a doctoral student! I would also like to thank my research mentors and friends, Andy Turner and Alan Wilson. As a ?budding field ecologist,? I?m sure I helped contribute to all of Andy?s gray hair! When I interviewed to be in Alan?s lab at Auburn, I?m sure I was probably the shyest and quietest student that Alan had met. I?m still definitely not an extrovert but am working on it! I appreciate the opportunity to continue working in Alan?s lab as a doctoral student. I would also like to thank my graduate committee, Jim Stoeckel and Dennis DeVries, for their advice and guidance. Finally, I would like to thank Rusty Wright for insight on teaching, experiments, and pond management (I truly am sorry about the fish kill on S1 during my first year at Auburn?). I thank Gary Mittelbach for loaning his pump to fill enclosures, Eric Park, Kristin Adamson, Jonathan Buck, and Jessica Davenport for assistance with experimental set-up, sampling, and iii laboratory analyses, Cova Arias, Oscar Olivares-Fuster, RajReni Kaul, and Lauren Jernigan for help and support with genetic analyses, and the staff of the Auburn University E. W. Shell. Fisheries Research Center for logistical support. My thesis was supported by an Auburn University Graduate School Thesis and Dissertation Research Award, a Sigma Xi Grant-in-Aid of Research, and by NSF grants DEB-0841864 and DEB-0841944. Lastly, I would like to thank all of my lab mates for hanging out, going fishing, and for subway/pizza/ice cream runs. I would especially like to thank Steve Cason, Matt Ferrell, and Lauren Jernigan for their help and support. I am sure that I could never have gotten a fellowship allowing me to stay on as a doctoral student at Auburn without all of your help. iv Table of Contents Acknowledgments??????????????????????????? ??.......ii List of Tables????????????????????????????????.. .v List of Figures??? ?????????????????????????????.vi Chapters 1 Do High Levels of Cyanobacterial Toxins Prevent Daphnia Control of Phytoplankton Biomass?..............................................................................................................................1 Abstract??????????????????????????????... 1 Introduction?????????????????????????????. 2 Methods??????????????? ???????????????.. .4 Results??????????????????????????????... ..9 Discussion?????????????????????????????. 11 References?????????????????????????????. 15 2 Nitrogen-to-Phosphorus Ratio Mediates the Outcome of Intraspecific Competition in Daphnia???????????????????????????????.. 27 Abstract??????????????????????????????. 27 Introduction????????????????????????????... 28 Methods??????????????????????????????. 30 Results??????????????????????????????... 35 Discussion??............................................................................................................. 37 References?????????????????????????????. 42 v List of Tables Table 1.1. Source lakes and genetic characterizations of the three Daphnia pulicaria clones. ....21 Table 2.1. Source lakes and genetic characterizations of the six Daphnia pulicaria clones. ?...47 Table 2.2. Effect of N:P ratio on the relative abundance of each of the six Daphnia pulicaria genotypes at the conclusion of the experiment (10 December 2009). ??????.47 vi List of Figures Figure 1.1. Dynamics of (A) Daphnia pulicaria biomass, (B) chlorophyll a, and (C) microcystin concentration for Experiment 1. ?????????????.23 Figure 1.2. Relative biomass of dominant phytoplankton taxa on day 28 (30 July 2007) of Experiment 1. ??????????????????????????.24 Figure 1.3. Dynamics of (A) Daphnia pulicaria biomass and (B) chlorophyll a for Experiment 2. ?????????????????????????...25 Figure 1.4. Relative biomass of dominant phytoplankton taxa and microcystin concentration on day 28 (5 May 2008) of Experiment 2. ??????? ???..26 Figure 2.1. Mean juvenile somatic growth rates for the six Daphnia pulicaria genotypes on two diets: a nutritious green alga (100% Ankistrodesmus) and toxic cyanobacteria (100 % Microcystis aeruginosa). ??????????50 Figure 2.2. Dynamics of (A) chlorophyll a and (B) cyanobacterial dry biomass over the ten-week enclosure experiment. ????????????????...51 Figure 2.3. Concentration of the toxin microcystin for each nitrogen-to-phosphorus (N:P) treatment immediately prior to stocking of Daphnia pulicaria genotypes (16 October 2009). ?????????????????????52 Figure 2.4. Relative biomass of phytoplankton taxa at the beginning (16 October 2009), middle (3 November 2009), and end (10 December 2009) of the ten-week enclosure experiment. ??????????????????...53 Figure 2.5. Daphnia pulicaria (A) density and (B) biomass for each nitrogen-to- phosphorus (N:P) treatment at the conclusion of the experiment (10 December 2009). ??????????????????????????54 Figure 2.6. Relative abundance of (A) each Daphnia pulicaria genotype and (B) D. pulicaria genotypes from oligotrophic and eutrophic lakes at the conclusion of the experiment (10 December 2009). ???????????55 1 Chapter 1 Do High Levels of Cyanobacterial Toxins Prevent Daphnia Control of Phytoplankton biomass? ABSTRACT Toxin-producing cyanobacteria have frequently been hypothesized to limit the ability of herbivorous zooplankton (such as Daphnia) to control phytoplankton biomass by inhibiting Daphnia feeding, population growth, and in extreme cases, causing Daphnia mortality. Using two limnocorral experiments in eutrophic ponds, we tested the hypothesis that high levels of cyanobacterial toxins prevent Daphnia from strongly reducing phytoplankton abundance. At the start of the first experiment, phytoplankton communities were dominated by toxic Microcystis and Anabaena (~96% of total phytoplankton biomass), and concentrations of the toxin microcystin were ~3 ?g L-1. Daphnia pulicaria were collected from a eutrophic, fishless lake, cultured in the laboratory, and stocked at a low density (~0.3 L-1) into half (n = 4) of the enclosures. Over the first two weeks after adding Daphnia, toxin levels increased to ~6.5 ?g L-1, yet Daphnia populations increased. By the third week, Daphnia had suppressed phytoplankton biomass by ~74% relative to ?no Daphnia? controls, and reduced phytoplankton biomass persisted until the conclusion of the experiment (6 August 2007). In the second experiment, phytoplankton communities were dominated by Anabaena and Microcystis, and our four treatments included a no Daphnia control and each of three genetically-distinct D. pulicaria clones that varied in tolerance to toxic Microcystis in a previously published laboratory experiment. We found no differences among the three D. pulicaria genotypes in their population growth rates and carrying capacities, and all three D. pulicaria genotypes reduced phytoplankton biomass similarly relative to controls. By the conclusion of the experiment (19 May 2008), D. pulicaria had completely eliminated cyanobacteria, regardless of their genotype. The ability of 2 Daphnia to increase and suppress phytoplankton biomass, despite high levels of cyanobacteria and associated toxins, suggests that toxin-producing cyanobacteria do not prevent strong Daphnia control of phytoplankton biomass. INTRODUCTION Within freshwater ecosystems, cyanobacteria are the most important phytoplankton taxa associated with harmful algal blooms (HABs) (Paerl 1988). Toxigenic cyanobacteria, including Anabaena, Cylindrospermopsis, Microcystis, and Oscillatoria (Planktothrix), tend to dominate nutrient-rich freshwater systems due to their superior competitive abilities under high nutrients, low nitrogen-to-phosphorus ratios, low light levels, reduced mixing, and high temperatures (Reynolds et al. 1987; Downing et al. 2001; Paerl and Huisman 2009). Despite large-scale efforts to limit nutrient enrichment, such as the Clean Water Act (Carpenter et al. 1998), cultural eutrophication (and concomitant HABs) continues to be the leading cause of water pollution for most freshwater and coastal marine ecosystems in the world and is a rapidly growing problem in the developing world (Smith and Schindler 2009). Cyanobacteria pose one of the most serious threats to nutrient-rich freshwater ecosystems by producing toxic secondary metabolites that can poison fishes, pets, livestock, and humans (Paerl 1988; Carmichael 1992). In addition to posing significant public health risks, cyanobacteria have been shown to be poor quality food for most zooplankton grazers in laboratory studies (Wilson et al. 2006; Wilson and Hay 2007; Tillmanns et al. 2008). Numerous studies have demonstrated that cyanobacteria can inhibit zooplankton feeding, growth, and reproduction through several mechanisms including (1) colonial or filamentous morphologies, (2) nutritional deficiencies, and (3) production of toxins (Porter 1977; Lampert 1987; Wilson et al. 2006). Therefore, an important consequence of cyanobacterial dominance with nutrient enrichment is that 3 phytoplankton communities tend to be dominated by harmful species that are relatively resistant to zooplankton grazing (Sommer et al. 1986; Ghadouani et al. 2003). However, recent research has shown that local populations of Daphnia may evolve tolerance to toxin-producing cyanobacteria in the diet under eutrophic conditions (Hairston et al. 1999; Gustafsson and Hansson 2004; Sarnelle and Wilson 2005). Genetically-based tolerance to toxic Microcystis has been observed within species of both European and North American Daphnia across habitats varying in nutrient enrichment (Sarnelle and Wilson 2005) and over time in a eutrophied lake (Hairston et al. 1999; Hairston et al. 2001). It is well known that the presence of the generalist herbivore Daphnia is critically important to the success of biomanipulation efforts to improve water quality in eutrophic lakes (Leibold 1989; Schindler 2006), and the evolution of tolerance to toxic cyanobacteria may have important consequences for the response of lakes to nutrient enrichment (Hairston et al. 2001; Sarnelle and Wilson 2005). To date, numerous studies demonstrate that Daphnia can greatly reduce the abundance of non-resistant phytoplankton in eutrophic lakes when freed from predation by planktivorous fish (Sarnelle 1992; Mazumder 1994). However, whether Daphnia can control grazing-resistant cyanobacteria continues to be a controversial question. There is considerable evidence that Daphnia may promote the dominance of grazing-resistant cyanobacteria (Carpenter et al. 1995; Brett and Goldman 1997; Ghadouani et al. 2003), but there are also several well-documented examples in which Daphnia have greatly reduced the abundance of cyanobacteria (Lynch and Shapiro 1981; Vanni et al. 1990; Sarnelle 2007). The presence (or absence) of cyanobacterial toxins may help explain the variable suppression of cyanobacterial abundance observed in previous studies. Toxin-producing cyanobacteria have frequently been hypothesized to limit the ability of Daphnia to control phytoplankton (and cyanobacterial) biomass (Porter 1977). 4 Furthermore, laboratory experiments have demonstrated that toxic cyanobacteria may inhibit Daphnia feeding, population growth, and in extreme cases, cause Daphnia mortality (Gliwicz and Lampert 1990; L?rling and van der Grinten 2003; reviewed by Wilson et al. 2006). Here we conduct two field experiments to test the hypothesis that high levels of toxic Microcystis (and the toxin microcystin) prevent Daphnia from suppressing cyanobacterial abundance. To our knowledge, no field experiments have examined the detrimental effects of cyanobacterial toxins on Daphnia, and previous field experiments were conducted prior to the advent of simple methods for measuring cyanobacterial toxins. METHODS Study sites The first experiment (Experiment 1) was conducted at Michigan State University, Michigan, in a small, eutrophic pond (MSU Lake 2). MSU Lake 2 was created as part of a project to treat secondary sewage effluent in 1973 (Spencer 1981) and is a shallow, polymictic pond with a surface area of approximately 3 ha, maximum depth of 2 m, and total phosphorus concentrations in the mixed layer averaging about 200 ?g L-1 (O. Sarnelle unpubl.). Cyanobacteria begin to dominate phytoplankton communities in MSU Pond 2 in late May, and Microcystis typically comprises greater than 90% of total phytoplankton biomass by August each year (O. Sarnelle unpubl.). The second experiment (Experiment 2) was conducted at the E.W. Shell Fisheries Research Station at Auburn University, Alabama, in a small, eutrophic reservoir pond (Pond S1). This pond is shallow and polymictic with a surface area of approximately 8 ha, maximum depth of 3.5 m, and total nitrogen (TN) and total phosphorus (TP) concentrations in the mixed layer averaging about 1900 ?g L-1 and 100 ?g L-1, respectively (Boyd and Shelton 1984; A. E. Wilson 5 and M. F. Chislock unpubl.). Cyanobacteria begin to dominate phytoplankton communities in the pond during late spring (April), and cyanobacterial blooms typically persist into early fall (September) (A. E. Wilson and M. F. Chislock unpubl.). Daphnia sources D. pulicaria for both experiments were collected from three small (<0.3 km2), eutrophic lakes in southern Michigan that have high cyanobacterial abundance during the summer months (Sarnelle and Wilson 2005; Table 1). D. pulicaria for Experiment 1 were collected from a nearby eutrophic pond lacking fish (MSU Lake 1) on 15 June 2007 and mass cultured in the laboratory in a 20-L glass carboy filled with autoclaved lake water, bubbled with an airstone, and fed a nutritious green algae (Ankistrodesmus) prior to the start of the experiment. The three D. pulicaria clones used in Experiment 2 were descendants of one female isolated in 2004 from each of three eutrophic lakes in southern Michigan and were shown to be tolerant to a diet of toxic Microcystis aeruginosa in a previous growth experiment (Sarnelle and Wilson 2005; Table 1). Prior to the field experiment, each D. pulicaria clone was first maintained in the laboratory (20-25oC, 12h light: 12h dark) in 20-L glass carboys filled with autoclaved lake water, bubbled with an airstone, and fed a nutritious green algae (Chlorella) grown in a nutrient-rich medium (modified BG-11 medium ? Vanderploeg et al. 2001). Each D. pulicaria clone was then transferred to separate outdoor 160-L tanks filled with lake water filtered through a 35-?m sieve and supplemented with Chlorella as a food source several weeks before the start of the field experiment. Therefore, all three D. pulicaria clones were grown under common-garden conditions to provide animals for stocking the limnocorrals. 6 Daphnia genetic analyses D. pulicaria genotypes for each of the three clones used in the second experiment were characterized using variation in two microsatellite loci (Dp464, Dp496 ? Colbourne et al. 2004) that have proven to be highly polymorphic for D. pulicaria collected from several of our study lakes in Michigan (Wilson and Hay 2007, A. E. Wilson unpubl.) (Table 1). We subsequently characterized D. pulicaria genotypes from each enclosure two weeks after initial stocking and at the conclusion of the experiment to confirm establishment of each D. pulicaria genotype treatment and that we had maintained single-genotype treatments of D. pulicaria in our enclosures. We genetically characterized 20-25 randomly selected D. pulicaria individuals from 95% ethanol-preserved macrozooplankton samples for all Daphnia enclosures for both dates (detailed methods in Wilson and Hay 2007). Enclosure experiments Experiment 1 was conducted in 4200-L, clear polyethylene enclosures that were sealed at the bottom, open to the atmosphere at the top, and suspended from a floating platform (EZ-Dock) anchored in MSU Pond 2. Eight enclosures were filled on 2 July 2007 by pumping pond water through a 75-?m mesh net to initially exclude resident Daphnia. We then randomly assigned half of the enclosures as no D. pulicaria controls, and half of the enclosures were stocked with D. pulicaria on 2 July 2007 to achieve initial densities of ~0.3 animals L-1. Two enclosures D. pulicaria enclosures were damaged allowing planktivorous fish (bluegill) to invade. We therefore only included the remaining 2 D. pulicaria enclosures in all statistical analyses. We sampled all enclosures weekly beginning on 2 July 2007 (immediately prior to Daphnia addition), and the experiment was concluded on 6 August 2007. 7 Experiment 2 was conducted in 2500-L, clear polyethylene enclosures that were sealed at the bottom, open to the atmosphere at the top, and suspended from a floating platform (EZ-Dock) anchored in the middle of Pond S1. Twelve enclosures were filled on 30 March 2008 by pumping pond water through a 75-?m mesh net to initially exclude resident Daphnia from treatments. Our four treatments included a no Daphnia control and each of three genetically- distinct D. pulicaria clones (Table 1), and we randomly assigned each enclosure to one of the four treatments. Enclosures for each D. pulicaria genotype treatment were stocked with Daphnia on 6 April 2008 to achieve initial densities of ~0.3 animals L-1. We sampled all enclosures weekly beginning on 5 April 2008, and the experiment was concluded on 19 May 2008. Sample collection and data analysis Depth-integrated water samples for chlorophyll a, phytoplankton biomass and species composition, microcystin, and macrozooplankton biomass and species composition were collected from the mixed layer of each enclosure with a tube sampler (inside diameter = 51 mm) for both experiments. Chlorophyll a concentrations were measured by extracting phytoplankton collected on Pall A/E filters in 90% ethanol for 24 h in the dark at 4oC followed by measurement with a fluorometer (Sartory and Grobbelaar 1984). Microcystin concentrations in particles were quantified using enzyme-linked immunosorbent assay (ELISA) (An and Carmichael 1994) after extraction from filters with 75% aqueous methanol. Phytoplankton species abundance and composition were determined for selected dates via the inverted microscope technique (Uterm?hl 1958) using water samples preserved in 1% Lugol?s solution. Biovolumes for each species were calculated using cell counts and estimates of cell volume based on measurements of cell dimensions. We then converted biovolumes (mm3 L-1) to dry biomass (?g L-1) assuming a specific gravity of 1 g cm-3 and a dry biomass: wet biomass ratio of 0.40 (Riemann et al. 1989; 8 Sarnelle et al. 2005; Knoll et al. 2008). Daphnia were measured and counted at 40X in a Sedgwick-Rafter cell, and total body lengths were converted to biomass using a length-weight regression for D. pulicaria (O. Sarnelle unpubl.). For Experiment 1, the effect of D. pulicaria on chlorophyll a and microcystin concentrations was tested using repeated measures analysis of variance (ANOVA, sampling date = repeated measure) across all sampling dates for the five-week experiment. Daphnia effects on phytoplankton community composition were examined for samples collected on day 28 (30 July 2007) of the experiment. Population growth rates for D. pulicaria were calculated as the slope of the linear regression of the natural logarithm of D. pulicaria density versus day of experiment during the period when populations were increasing (9 July to 23 July 2007). For Experiment 2, we used repeated measures ANOVA to analyze treatment effects on chlorophyll a across all sampling dates for the six-week experiment. ANOVA was used to analyze treatment effects on phytoplankton community composition and microcystin concentrations on day 28 (5 May 2008). Tukey?s post hoc test was then used to assess pairwise differences across the three D. pulicaria genotypes treatments. Population growth rates for each D. pulicaria genotype in each replicate enclosure were calculated as in Experiment 1, using D. pulicaria densities from 14 April to 28 April 2008. We then used ANOVA to compare mean population growth rates across the three D. pulicaria genotype treatments. To compare the Daphnia effects on phytoplankton abundance for Experiments 1 and 2, we calculated effect sizes by dividing the mean chlorophyll concentration in control enclosures by the mean chlorophyll concentrations in enclosures with D. pulicaria (Sarnelle 1992). All data were checked for normality and homogeneity and log-transformed when necessary. 9 RESULTS Experiment 1 At the start of Experiment 1, phytoplankton communities in the enclosures were dominated almost entirely by toxic cyanobacteria, and Microcystis and Anabaena contributed over 96% of total phytoplankton biomass. Initial phytoplankton and cyanobacterial biomass were high across all enclosures (mean chlorophyll a = ~130 ?g L-1 and mean cyanobacterial biomass = ~19,000 ?g L-1), and concentrations of the toxin, microcystin, were above 3 ?g L-1 (Figure 1). Over the first two weeks of the experiment, toxin levels increased to approximately 6 ?g L-1 in enclosures with D. pulicaria (Figure 1C), yet D. pulicaria populations were increasing (r = 0.24 day-1, SE = 0.06 day-1) (Figure 1A). By the third week, D. pulicaria had suppressed phytoplankton biomass by ~74% relative to no D. pulicaria controls (Figure 1B), and this reduced phytoplankton biomass persisted until the conclusion of the experiment (6 August 2007) (F1,4 = 20.634, P = 0.010). D. pulicaria reduced chlorophyll a concentrations by a factor of 3.3, relative to no D. pulicaria controls, at the conclusion of the experiment. In general, D. pulicaria strongly suppressed the biomass of all species of cyanobacteria (Figure 2) relative to no D. pulicaria controls by day 28 (30 July 2007), reducing Microcystis, Oscillatoria, and Anabaena species by factors of ~4 (F1,4 = 14.539, P = 0.019), 56 (F1,4 = 15.746, P = 0.017), and 20 (F1,4 = 5.857, P = 0.073), respectively. Experiment 2 Initial phytoplankton biomass was high across all treatments for Experiment 2 (mean chlorophyll a = 75 ?g L-1) (Figure 3B), and phytoplankton communities were dominated by a mixture of the cyanobacteria Anabaena circinalis, Microcystis aeruginosa, and Chroococcus sp., which comprised approximately 85% of total phytoplankton biomass on 5 April 2008. 10 Characterization of D. pulicaria genotypes from each enclosure two weeks after initial stocking and at the conclusion of the experiment confirmed that the appropriate genotypes became established in each enclosure and that we were able to maintain single-genotype treatments of D. pulicaria throughout the experiment. 100% of the 20-25 randomly selected D. pulicaria individuals from each enclosure were the correct genotype two weeks after stocking and at the conclusion of the experiment. All three D. pulicaria genotypes rapidly increased in abundance and biomass during the first three weeks of the experiment (Figure 3A), and population growth rates did not differ across the three D. pulicaria genotypes (F2,6 = 0.235, P = 0.798). Initial microcystin concentrations were ~0.04 ?g L-1 for controls and enclosures for the three D. pulicaria genotypes. By day 28 (5 May 2008), microcystin concentrations had increased to greater than 1 ?g L-1 in control enclosures (Figure 4B), while concentrations of the toxin were less than 1 ng L-1 in enclosures for all D. pulicaria genotype treatments (F3,8 = 333.398, P < 0.0001). Over the course of the experiment, D. pulicaria abundance and biomass were similar for the three D. pulicaria genotype treatments (Abundance: F2,6 = 0.513, P = 0.623; Biomass: F2,6 = 1.434, P = 0.310), and by the final two weeks of the experiment, D. pulicaria populations had declined from their peak and leveled off (Figure 3A). In general, all three D. pulicaria genotypes significantly reduced chlorophyll a concentration over time (Figure 3B; F3,8 = 94.677, P < 0.0001). D. pulicaria from Baker Lake had the largest effect on chlorophyll a over time (Figure 3B; Tukey?s test: Baker vs. Wintergreen, P = 0.007; Baker vs. MSU Pond 1, P = 0.079). However, by the conclusion of the experiment (19 May 2008), all three D. pulicaria genotypes reduced chlorophyll a concentrations similarly relative to the control treatment (Tukey?s test for final date: P > 0.90), and D. pulicaria reduced chlorophyll 11 a concentrations by a factor of 26.1, relative to no D. pulicaria controls. Furthermore, all three D. pulicaria genotypes strongly suppressed the biomass of cyanobacteria (F3,8 = 82.669, P < 0.0001) and flagellates (F3,8 = 80.801, P < 0.0001) but did not affect the biomass of chlorophytes (F3,8 = 2.623, P = 0.122) and diatoms (F3,8 = 0.743, P = 0.556) (Figure 4A). DISCUSSION High levels of phytoplankton toxins are one explanation for the escape of phytoplankton from grazer control (Porter 1977). Within freshwater ecosystems, bloom-forming cyanobacteria (including Anabaena, Microcystis, and Oscillatoria) are the most important toxin-producing phytoplankton taxa (Carmichael 1992). Cyanobacterial toxins have frequently been shown to have negative effects on the fitness of zooplankton herbivores (including Daphnia) in laboratory studies (Gilbert 1990; DeMott et al. 1991, Lurling and van der Grinten 2003). Therefore, toxin production by cyanobacteria is hypothesized to drive declines of large daphnids with eutrophication and the concomitant escape of cyanobacteria from Daphnia control (Jeppesen et al. 2000; Porter 1977; Sommer et al. 1986). However, our results show that D. pulicaria can strongly suppress both cyanobacterial and phytoplankton biomass despite high levels of one of the most commonly measured and toxic of the cyanotoxins, microcystin. Recent laboratory experiments have revealed intraspecific variation within Daphnia species for traits conferring tolerance to toxic Microcystis (Hairston et al. 1999; Hairston et al. 2001; Sarnelle and Wilson 2005). Daphnia from environments with infrequent cyanobacterial blooms are more inhibited by Microcystis than Daphnia from environments with recurring cyanobacterial blooms (Sarnelle and Wilson 2005). Furthermore, phenotypic plasticity has been observed in Daphnia in response to exposure to toxic cyanobacteria in the diet (Gustafsson and Hansson 2004). Therefore, cyanotoxins are likely strong selective forces that potentially shape 12 Daphnia population genetics seasonally within lakes and across lakes of varying trophic status. For Experiments 1 and 2, D. pulicaria were collected from environments that have high cyanobacterial abundance during the summer months. Therefore, the ability of D. pulicaria to increase in abundance and suppress toxic cyanobacteria may have been driven by local adaptation to recurring cyanobacterial blooms. In Experiment 2, we compared the ability of three D. pulicaria clones, identified to be tolerant to a laboratory diet of toxic Microcystis in a previously published laboratory experiment (Sarnelle and Wilson 2005), to reproduce and control toxic cyanobacteria in the field. Numerous studies have demonstrated that the results of laboratory microcosm experiments are not always consistent with the results of manipulative field experiments and whole-lake studies (Carpenter 1996). Consequently, several recent studies have examined the relevance of generalizations based on microcosms to communities and ecosystems (Levin 1992; Sarnelle 1997; Spivak et al. 2011). Our results show that D. pulicaria genotypes identified to be tolerant to Microcystis in laboratory experiments can increase dramatically and eliminate cyanobacteria in the field. Therefore, the presumed role of cyanobacteria in driving seasonal declines of large daphnids may need to be re-evaluated in light of these results. The negative correlation between lake productivity and the abundance of large daphnids is more likely driven by the increasing abundance of planktivorous fish with nutrient enrichment (Jeppesen et al. 1997). As the Microcystis-tolerant D. pulicaria genotypes in our experiment were collected from three lakes spanning a large total phosphorus gradient (21 to 300 ?g L-1), local adaptation in response to increased cyanobacterial dominance may be a common occurrence in nature, particularly in lakes with a long history of eutrophication (Hairston et al. 1999). 13 D. pulicaria strongly suppressed cyanobacterial and phytoplankton abundance in both Experiments 1 and 2. However, D. pulicaria had a much larger effect on phytoplankton abundance in Experiment 2, reducing chlorophyll a concentrations by a factor of 26.1 (compared to 3.3 in Experiment 1). Furthermore, D. pulicaria completely eliminated all cyanobacterial species from enclosures by the conclusion of Experiment 2, while cyanobacteria comprised approximately 75% of the phytoplankton biomass on day 28 of Experiment 1 (data not shown). One potential explanation for D. pulicaria failing to eliminate all cyanobacteria in Experiment 1 is that the remaining cyanobacterial colonies were too large to be consumed by Daphnia. In Experiment 2, the majority of Microcystis colonies were edible (<10 cells per colony). Contrastingly, colonies with greater than ten cells (> 40?m) contributed over 80% of Microcystis biomass in Experiment 1. Therefore, larger colony size likely contributed to the greater overall grazing-resistance in the phytoplankton in Experiment 1. However, the equivalent diameter of Microcystis colonies in control and D. pulicaria enclosures did not differ on day 28 (F1,4 = 0.969, P = 0.381) for Experiment 1. In addition, the proportion of edible Microcystis (i.e., single cells and colonies of less than 10 cells) did not differ between control and D. pulicaria enclosures (F1,4 = 0.784, P = 0.426). Therefore, D. pulicaria presence did not favor larger Microcystis colonies. Concentrations and dynamics of the cyanotoxin microcystin also differed between Experiments 1 and 2. In Experiment 1, microcystin concentrations were greater than 3 ?g L-1 immediately prior to the addition of D. pulicaria. In contrast, initial microcystin concentrations for Experiment 2 were nearly two orders of magnitude lower (~0.04 ?g L-1) than in Experiment 1. Although microcystin concentrations increased to greater than 1 ?g L-1 in enclosures lacking D. pulicaria in Experiment 2, microcystin concentrations decreased to less than 1 ng L-1 in enclosures with D. pulicaria as Daphnia were able to suppress cyanobacterial biomass before 14 toxin levels could increase. High levels of cyanobacterial toxins (microcystin > 3 ?g L-1) certainly do not prevent D. pulicaria from suppressing phytoplankton and cyanobacterial abundance. However, higher initial concentrations of the toxin microcystin may have contributed to the greater overall grazing-resistance in the phytoplankton in Experiment 1. It is also possible that nutritional deficiencies and the toxicity of undescribed (and unmeasured) cyanobacterial compounds mediated differences in the grazing-resistance of phytoplankton in Experiments 1 and 2. One reason why the world is green (Hairston et al. 1960) may be that many plants are chemically defended against herbivores (Hay and Fenical 1988). In green lakes, toxin-producing cyanobacteria have frequently been hypothesized to limit the ability of herbivorous zooplankton (such as Daphnia) to suppress phytoplankton biomass (Porter 1977). However, our study demonstrates that high levels of the cyanotoxin microcystin do not prevent strong Daphnia control of phytoplankton biomass and adds to a growing body of literature demonstrating that Daphnia can have large negative effects on the abundance of cyanobacteria in eutrophic lakes, when freed from predation by planktivorous fish (Vanni et al. 1990; Sarnelle 1993; Sarnelle 2007). In both experiments, we used Daphnia collected from lakes with recurring cyanobacterial blooms, and Experiment 2 demonstrated that adaptations by D. pulicaria to increased prevalence of cyanobacteria in three different eutrophic lakes seem to have similar community- and ecosystem-level effects. As Daphnia from eutrophic lakes tend to be more tolerant to toxic cyanobacteria than Daphnia from oligotrophic lakes, local adaptation by Daphnia in response to toxic cyanobacteria may play an important role in the response of lake ecosystems to cultural eutrophication and food-web manipulations (Hairston et al. 1999). 15 REFERENCES An, J. S., and W. W. Carmichael. 1994. Use of a colorimetric protein phosphatase inhibition assay and enzyme-linked-immunosorbent-assay for the study of microcystins and nodularins. Toxicon 32: 1495-1507. Boyd, C. E., and J. L. Shelton. 1984. Observations on the hydrology and morphometry of ponds on the Auburn University Fisheries Research Unit. Alabama Agricultural Experiment Station Bulletin: 1-64. Brett, M. T., and C. R. Goldman. 1997. Consumer versus resource control in freshwater pelagic food webs. Science 275: 384-386. Carmichael, W. W. 1992. Cyanobacteria secondary metabolites - the cyanotoxins. Journal of Applied Bacteriology 72: 445-459. Carpenter, S. R. 1996. Microcosm experiments have limited relevance for community and ecosystem ecology. Ecology 77: 677-680. Carpenter, S. R., D. L. Christensen, J. J. Cole, K. L. Cottingham, X. He, J. R. Hodgson, J. F. Kitchell, S. E. Knight, and M. L. Pace. 1995. Biological control of eutrophication in lakes. Environmental Science & Technology 29: 784-786. Carpenter, S. R., N. F. Caraco, D. L. Correll, R. W. Howarth, A. N. Sharpley, and V. H. Smith. 1998. Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecological Applications 8: 559-568. Colbourne, J. K., B. Robison, K. Bogart, and M. Lynch. 2004. Five hundred and twenty-eight microsatellite markers for ecological genomic investigations using Daphnia. Molecular Ecology Notes 4: 485-490. DeMott, W. R., Q. X. Zhang, and W. W. Carmichael. 1991. Effects of toxic cyanobacteria and 16 purified toxins on the survival and feeding of a copepod and 3 species of Daphnia. Limnol. Oceanogr. 36: 1346-1357. Downing, J. A., S. B. Watson, and E. McCauley. 2001. Predicting cyanobacteria dominance in lakes. Canadian Journal of Fisheries and Aquatic Sciences 58: 1905-1908. Ghadouani, A., B. Pinel-Alloul, and E. E. Prepas. 2003. Effects of experimentally induced cyanobacterial blooms on crustacean zooplankton communities. Freshwater Biology 48: 363- 381. Gilbert, J. J. 1990. Differential effects of Anabaena affinis on cladocerans and rotifers: mechanisms and implications. Ecology 71: 1727-1740. Gliwicz, Z. M., and W. Lampert. 1990. Food thresholds in Daphnia species in the absence and presence of blue-green filaments. Ecology 71: 691-702. Gustafsson, S., and L.-A. Hansson. 2004. Development of tolerance against toxic cyanobacteria in Daphnia. Aquatic Ecology 38: 37-44. Hairston, G. G., F. E. Smith, and L. B. Slobodkin. 1960. Community structure, population control, and competition. The American Naturalist 94: 421-425. Hairston, N. G., W. Lampert, C. E. Caceres, C. L. Holtmeier, L. J. Weider, U. Gaedke, J. M. Fischer, J. A. Fox, and D. M. Post. 1999. Lake ecosystems - Rapid evolution revealed by dormant eggs. Nature 401: 446-446. Hairston, N. G., C. L. Holtmeier, W. Lampert, L. J. Weider, D. M. Post, J. M. Fischer, C. E. Caceres, J. A. Fox, and U. Gaedke. 2001. Natural selection for grazer resistance to toxic cyanobacteria: Evolution of phenotypic plasticity? Evolution 55: 2203-2214. Hay, M. E. and W. Fenical. 1988. Marine plant-herbivore interactions: The ecology of chemical defense. Annual Review of Ecology and Systematics 19: 111-145. 17 Knoll, L. B. S., O.; Hamilton, S. K.; Kissman, C. E. H., Wilson, A. E., Rose, J. B., and Morgan, M. R. 2008. Corrigendum: Invasive zebra mussels (Dreissena polymorpha) increase cyanobacterial toxin concentrations in low-nutrient lakes. Canadian Journal of Fisheries and Aquatic Sciences 65: 2552-2552. Lampert, W. 1987. Laboratory studies on zooplankton-cyanobacteria interactions. New Zealand Journal of Marine and Freshwater Research 21: 483-490. Leibold, M. A. 1989. Resource edibility and the effects of predators and productivity on the outcome of trophic interactions. American Naturalist 134: 922-949. Levin, S. A. 1992. The problem of pattern and scale in ecology. Ecology 73: 1943-1967. L?rling, M., and E. Van Der Grinten. 2003. Life-history characteristics of Daphnia exposed to dissolved microcystin-LR and to the cyanobacterium Microcystis aeruginosa with and without microcystins. Environmental Toxicology and Chemistry 22: 1281-1287. Lynch, M., and J. Shapiro. 1981. Predation, enrichment, and phytoplankton community structure. Limnol. Oceanogr. 26: 86-102. Mazumder, A. 1994. Patterns of algal biomass in dominant odd- vs. even-link lake ecosystems. Ecology 75: 1141-1149. Paerl, H. W. 1988. Nuisance phytoplankton blooms in coastal, estuarine, and inland waters. Limnol. Oceanogr. 33: 823-847. Paerl, H. W., and J. Huisman. 2009. Climate change: a catalyst for global expansion of harmful cyanobacterial blooms. Environmental Microbiology Reports 1: 27-37. Porter, K. G. 1977. The plant-animal interface in freshwater ecosystems. American Scientist 65: 159-170. Reynolds, C. S., R. L. Oliver, and A. E. Walsby. 1987. Cyanobacterial dominance - the role of 18 buoyancy regulation in dynamic lake environments. New Zealand Journal of Marine and Freshwater Research 21: 379-390. Riemann, B., P. Simonsen, and L. Stensgaard. 1989. The carbon and chlorophyll content of phytoplankton from various nutrient regimes. Journal of Plankton Research 11: 1037-1045. Sarnelle, O. 1992. Nutrient enrichment and grazer effects on phytoplankton in lakes. Ecology 73: 551-560. Sarnelle, O. 1993. Herbivore effects on phytoplankton succession in a eutrophic lake. Ecological Monographs 63: 129-149. Sarnelle, O. 1997. Daphnia effects on microzooplankton: comparisons of enclosure and whole- lake responses. Ecology 78: 913-928. Sarnelle, O. 2007. Initial conditions mediate the interaction between Daphnia and bloom- forming cyanobacteria. Limnol. Oceanogr. 52: 2120-2127. Sarnelle, O., and A. E. Wilson. 2005. Local adaptation of Daphnia pulicaria to toxic cyanobacteria. Limnol. Oceanogr. 50: 1565-1570. Sarnelle, O., A. E. Wilson, S. K. Hamilton, L. B. Knoll, and D. F. Raikow. 2005. Complex interactions between the zebra mussel, Dreissena polymorpha, and the harmful phytoplankton, Microcystis aeruginosa. Limnol. Oceanogr. 50: 896-904. Sartory, D. P., and J. U. Grobbelaar. 1984. Extraction of chlorophyll-a from fresh-water phytoplankton for spectrophotometric analysis. Hydrobiologia 114: 177-187. Schindler, D. W. 2006. Recent advances in the understanding and management of eutrophication. Limnol. Oceanogr. 51: 356-363. Smith, V. H., and D. W. Schindler. 2009. Eutrophication science: where do we go from here? Trends Ecol Evol 24: 201-207. 19 Sommer, U., Z. M. Gliwicz, W. Lampert, and A. Duncan. 1986. The PEG-model of seasonal succession of planktonic event in fresh waters. Arch. Hydrobiol. 106: 433-471. Spencer, C. N. 1981. Regulation of algae and macrophytes in eutrophic ponds by interacting biological, chemical, and physical factors. M.S. thesis. Michigan State University. Spivak, A. C., M. J. Vanni, and E. M. Mette. 2011. Moving on up: can results from simple aquatic mesocosm experiments be applied across broad spatial scales? Freshwater Biology 56: 279-291. Tillmanns, A. R., A. E. Wilson, F. R. Pick, and O. Sarnelle. 2008. Meta-analysis of cyanobacterial effects on zooplankton population growth rate: species-specific responses. Fundamental and Applied Limnology 171: 285-295. Uterm?hl, H. 1958. Zur Vervollkommnung der quantitativen Phytoplankton-Methodik. Mitt. int. Ver. theor. angew. Limnol. 9: 1-38. Vanderploeg, H. A., J. R. Liebig, W. W. Carmichael, M. A. Agy, T. H. Johengen, G. L. Fahnenstiel, and T. F. Nalepa. 2001. Zebra mussel (Dreissena polymorpha) selective filtration promoted toxic Microcystis blooms in Saginaw Bay (Lake Huron) and Lake Erie. Canadian Journal of Fisheries and Aquatic Sciences 58: 1208-1221. Vanni, M. J., C. Luecke, J. F. Kitchell, Y. Allen, J. Temte, and J. J. Magnuson. 1990. Effects on lower trophic levels of massive fish mortality. Nature 344: 333-335. Wilson, A. E., and M. E. Hay. 2007. A direct test of cyanobacterial chemical defense: Variable effects of microcystin-treated food on two Daphnia pulicaria clones. Limnol. Oceanogr. 52: 1467-1479. Wilson, A. E., O. Sarnelle, and A. R. Tillmanns. 2006. Effects of cyanobacterial toxicity and morphology on the population growth of freshwater zooplankton: Meta-analyses of laboratory 20 experiments. Limnol. Oceanogr. 51: 1915-1924. 21 TABLES Table 1.1. Source lakes and genetic characterizations of the three D. pulicaria clones. Each D. pulicaria clone was characterized using two microsatellite markers (Dp464 and Dp496), and microsatellite nucleotide lengths were determined for both loci. Location Total Dp464 Dp496 Lake Lat. N, Long. W) Phosphorus (?g l-1) length (bp) length (bp) Baker 42o26?27?, 85o21?03? 21-40 147, 147 201, 216 MSU Pond 1 42o40?53?, 84o28?57? 170-300 148, 148 198, 202 Wintergreen 42o23?50?, 85o23?07? 50-70 149, 149 196, 202 22 FIGURE LEGENDS Figure 1. Dynamics of (A) Daphnia pulicaria biomass, (B) chlorophyll a, and (C) microcystin concentration for Experiment 1. Data represent means ? one standard error for each treatment over the five-week enclosure experiment. Figure 2. Relative biomass of dominant phytoplankton taxa on day 28 (30 July 2007) of Experiment 1. Data represent means for phytoplankton taxa for each treatment. The ?other? phytoplankton taxon included a mixture of flagellates and chlorophytes. Figure 3. Dynamics of (A) Daphnia pulicaria biomass and (B) chlorophyll a for Experiment 2. Data represent means ? one standard error for each treatment over the six-week experiment. Figure 4. Relative biomass of dominant phytoplankton taxa and microcystin concentration on day 28 (5 May 2008) of Experiment 2. Data represent means for phytoplankton taxa and means ? one standard error for microcystin concentration for each treatment. 23 24 25 26 27 Chapter 2 Nitrogen-to-Phosphorus Ratio Mediates the Outcome of Intraspecific Competition in Daphnia ABSTRACT Despite the traditional emphasis on plant defenses in plant-animal arms races, herbivore offenses may have important consequences for community structure and ecosystem function. Cyanobacteria produce toxins that may reduce the fitness of herbivorous zooplankton (such as Daphnia). However, it has recently been discovered that Daphnia populations in eutrophic lakes can adapt to tolerate toxic cyanobacteria. We use lab-measured variation in tolerance to toxic cyanobacteria and juvenile growth rates on high quality food for D. pulicaria populations from eutrophic versus oligotrophic lakes to examine intraspecific competition among Daphnia and potential trade-offs associated with these adaptations in a ten-week limnocorral experiment. Enclosures were fertilized at high or low nitrogen-to-phosphorus (N:P) ratios to manipulate phytoplankton and cyanobacterial abundance (and concentrations of the toxin microcystin). We then stocked a mixture of six genetically-distinct D. pulicaria clones from oligotrophic and eutrophic lakes into half of the enclosures. High N:P resulted in over four-fold higher chlorophyll a concentrations and eight-fold higher cyanobacterial biomass relative to low N:P enclosures at the start of the experiment. By the conclusion of the experiment, relative abundance of Daphnia genotypes from eutrophic lakes was nearly 100% for high N:P enclosures, while all six Daphnia genotypes were equally represented at low N:P. However, these results were driven by the dominance of a single Daphnia genotype at high N:P. Thus, our results indicate that N:P ratio can mediate the outcome of intraspecific competition in Daphnia by affecting phytoplankton abundance and species composition. 28 INTRODUCTION A central goal in community ecology is to understand how consumer-resource interactions influence the dynamics of populations, communities, and ecosystems (Tilman 1982). Despite the traditional assumption of phenotypic homogeneity within a species (Tilman 1982; Leibold et al. 1997), a growing body of studies suggests that intraspecific trait variation may have important ecological consequences (Thompson et al. 2001; Whitham et al. 2006; Post et al. 2008). We now know that genetically-based phenotypic variation, phenotypic plasticity, and inducible defenses can all be important in mediating species interactions (Turner and Mittelbach 1990; Tollrian and Harvell 1999; Werner and Peacor 2003). While numerous studies have documented trade-offs in the competitive abilities of different consumer species for resources (Desmarais and Tessier 1999; Tessier et al. 2000; Tessier and Woodruff 2002), the consequences of similar trade-offs within species are less studied. Recent research has revealed that there is a large amount of inter- and intraspecific variation in the ability of herbivores to tolerate plant defenses (Karban and Agrawal 2002). Theory and evidence in support of plant defenses have accumulated over the past several decades, and plant chemical defenses against herbivory have been documented for terrestrial and freshwater vascular plants (Howe and Westley 1988), freshwater and marine macroalgae (Hay and Fenical 1988), and some phytoplankton taxa (Carmichael 1992; Kirk and Gilbert 1992). In contrast, the study of herbivore offenses, the adaptations that increase herbivore fitness by countering plant defenses has received less emphasis (Karban and Agrawal 2002). Here we use intraspecific variation within a generalist herbivore in freshwater lakes (i.e., Daphnia) to examine the relative competitive ability of Daphnia that vary in tolerance to cyanobacteria and their associated toxins (Sarnelle and Wilson 2005). 29 Cyanobacteria pose one of the most serious threats to nutrient-rich freshwater ecosystems by producing toxic secondary metabolites that can poison fishes, pets, livestock, and humans (Paerl 1988; Carmichael 1992). In addition to posing significant public health risks, cyanobacteria have been shown to be poor quality food for herbivorous zooplankton (such as Daphnia) in laboratory studies (Wilson et al. 2006; Wilson and Hay 2007; Tillmanns et al. 2008). However, recent research has shown that local populations of Daphnia may evolve to tolerate toxic cyanobacteria following prolonged exposure to cyanobacterial blooms (Hairston et al. 1999; Hairston et al. 2001; Sarnelle and Wilson 2005). Genetically-based tolerance to toxic cyanobacteria has now been documented within Daphnia species using common-garden laboratory experiments comparing populations of European D. galeata before and after a major eutrophication event (Hairston et al. 1999; Hairston et al. 2001) and populations of North American D. pulicaria across lakes of varying nutrient enrichment (Sarnelle and Wilson 2005). It is well known that the presence of the generalist herbivore Daphnia is critically important to the success of biomanipulation efforts to improve water quality in eutrophic lakes (Leibold 1989; Schindler 2006), and the evolution of tolerance to toxic cyanobacteria may have important consequences for the response of lakes to nutrient enrichment (Hairston et al. 2001; Sarnelle and Wilson 2005). Nutrient enrichment of freshwater ecosystems leads to a simultaneous increase in total phytoplankton biomass and the relative abundance of cyanobacteria (Smith 1983; Watson et al. 1997; Downing et al. 2001). It is well known that many herbivores (including Daphnia) often trade-off the ability to exploit high resource levels (i.e., high food quality and quantity) with the ability to depress resources to low levels (Tessier et al. 2000; Tessier and Woodruff 2002). However, the costs and competitive consequences of tolerance to toxic cyanobacteria for Daphnia (and for herbivore adaptations to cope with plant chemical defenses, in general) are 30 unexplored (Karban and Agrawal 2002). Daphnia from eutrophic lakes are more tolerant of toxic cyanobacteria than Daphnia from oligotrophic lakes (Sarnelle and Wilson 2005). Adaptation to toxic cyanobacteria is likely beneficial when cyanobacterial biomass is high. However, tolerance to toxic cyanobacteria may be costly in environments where cyanobacteria are rare. Here we use variation in tolerance to toxic cyanobacteria for D. pulicaria populations from eutrophic versus oligotrophic lakes to examine intraspecific competition among Daphnia and potential trade-offs associated with these adaptations. We manipulated nitrogen-to-phosphorus (N:P) ratios in a limnocorral experiment to favor dominance by toxic cyanobacteria in half of the enclosures. We then evaluated the effect of N:P ratio on intraspecific competition among six genetically-distinct D. pulicaria clones that were collected from oligotrophic and eutrophic lakes and varied in tolerance to toxic cyanobacteria in a previously published laboratory experiment (Sarnelle and Wilson 2005). METHODS Daphnia clones The six D. pulicaria clones used in this experiment were descendants of one female isolated in 2004 from each of six small glacial lakes (<0.3 km2) in southern Michigan and were shown to vary in tolerance to a laboratory diet of toxic Microcystis aeruginosa and in juvenile growth rates on high quality food (the green alga Ankistrodesmus) in a previous growth experiment (Figure 1; Sarnelle and Wilson 2005). Three of the D. pulicaria source lakes (Lawrence, Sixteen, and Warner) are oligotrophic and cyanobacteria are rare, while the other three lakes are eutrophic and have high cyanobacterial abundance during the summer months (Table 1; Sarnelle and Wilson 2005). Prior to the field experiment, each D. pulicaria clone was first maintained in the laboratory (25oC, 12h light: 12h dark) in 1-L glass beakers filled with autoclaved lake water and 31 fed a nutritious green algae (Chlorella) grown in a nutrient-rich medium (modified BG-11 medium ? Vanderploeg et al. 2001). Each D. pulicaria clone was then transferred to separate outdoor 160-L tanks filled with lake water filtered through a 35-?m sieve and supplemented with Chlorella as a food source several weeks before the start of the experiment. Therefore, all six D. pulicaria clones were grown under common-garden conditions to provide animals for stocking the enclosures. Daphnia genetic analyses D. pulicaria genotypes for each of the six clones used in the experiment were genetically discriminated using variation in two microsatellite loci (Dp3, Dp339 ? Colbourne et al. 2004) that have proven to be highly polymorphic for D. pulicaria collected from several of our study lakes in Michigan (Table 1; A. E. Wilson unpubl.). D. pulicaria from ethanol-preserved samples (see ?Sample collection and data analysis? section) were rinsed thoroughly with distilled water to remove attached bacteria and phytoplankton. Genomic deoxyribonucleic acid (DNA) was extracted by heating individual D. pulicaria to 95oC in 10 ?L Lyse-N-Go PCR reagent (Pierce Chemical Co., Rockford, IL). Forward primers were modified with (-29)/IRDye labeled 19-mer M13 primer sequence in order to visualize polymerase chain reaction (PCR) products on a Li- Cor 4300 DNA Analyzer (Li-Cor Biosciences, Lincoln, NE). Amplification of microsatellite alleles was performed using polymerase chain reaction (PCR) in 12.5 ?L volumes (~40 ng of DNA, 1X buffer [Promega Go Green Colorless Buffer], 1.5 mM MgCl2, 0.2 mM dNTP, 1 pmol M13 labeled forward microsatellite primer, 2 pmol reverse microsatellite primer, 0.5 pmol IRDye labeled M13 primer, and 0.5 units of Taq DNA polymerase). Each locus was separately analyzed for each D. pulicaria individual. PCR used a touchdown protocol under the following conditions: 95oC for 3 min followed by 10 cycles of 94oC for 35 s, 65oC (-1oC/cycle) for 35 s, 32 72oC for 45 s, followed by an additional 30 cycles with a constant annealing temperature of 55oC with a final extension at 72oC for 10 min. Reactions were stopped with 6 ?L stop buffer, diluted as necessary with deionized water, denatured for 3 minutes at 90oC, and snapped cold before loading on a 6.5% polyacrylamide gel. Study site The experiment was conducted at the E.W. Shell Fisheries Research Station at Auburn University, Alabama, in a small, eutrophic reservoir pond (Pond S1). Pond S1 is a shallow, polymictic pond with a surface area of approximately 8 ha, maximum depth of 3.5 m, and total nitrogen (TN) and total phosphorus (TP) concentrations in the mixed layer averaging about 1000 ?g L-1 and 150 ?g L-1, respectively in the fall (Boyd and Shelton 1984; A. E. Wilson and M. F. Chislock unpubl.). Cyanobacteria begin to dominate phytoplankton communities in Pond S1 during late spring (April), and cyanobacterial blooms typically persist into early fall (September (A. E. Wilson and M. F. Chislock unpubl.). Immediately prior to the initiation of the experiment (5 October 2009), chlorophyll a in Pond S1 was approximately 30 ?g L-1, and cyanobacteria comprised less than 10% of total phytoplankton biomass. Enclosure experiment We manipulated nitrogen-to-phosphorus (N:P) ratio in 2500-L, clear polyethylene enclosures that were sealed at the bottom, open to the atmosphere at the top, and suspended from a floating platform (EZ-Dock) anchored in the middle of the pond. Twenty-four enclosures were filled on 5 October 2009 by pumping pond water through a 75-?m mesh net to initially exclude resident Daphnia. All enclosures were enriched at a rate of 400 ?g L-1 phosphorus added as K2HPO4. Low N:P enclosures received no addition of nitrogen, while high N:P enclosures were also enriched with 7000 ?g L-1 nitrogen added as NH4Cl (n=12 enclosures per N:P ratio). Therefore, 33 the ratio of total nitrogen to total phosphorus (TN:TP) in low N:P enclosures was ~2:1, by mass, and high N:P enclosures had a TN:TP ratio of ~14:1, by mass. We subsequently stocked approximately equal densities of six genetically-distinct D. pulicaria clones from oligotrophic and eutrophic lakes (n=3 clones per lake type) into half (n=12) of the enclosures at a total density of 0.1 L-1 on 17 October 2009, and the remaining twelve enclosures served as no D. pulicaria controls. Thus, there were six replicate enclosures for each N:P ratio/D. pulicaria treatment. During D. pulicaria addition, we preserved two subsamples of the Daphnia inoculum in 95% ethanol for genetic analysis to confirm that we stocked approximately equal densities of each D. pulicaria genotype. We sampled all enclosures biweekly-to-monthly beginning on 5 October 2009, and the experiment was concluded on 10 December 2009. Sample collection and data analysis Depth-integrated water samples for chlorophyll a, phytoplankton biomass and species composition, and microcystin were collected from each enclosure with a tube sampler (inside diameter = 51 mm) on each sampling date. On the final sampling date (10 December 2009), we also collected depth-integrated samples for D. pulicaria density, biomass, and genotypic composition. Chlorophyll a concentrations were measured by extracting phytoplankton collected on Pall A/E filters in 90% ethanol for 24 h in the dark at 4oC followed by measurement with a fluorometer (Sartory and Grobbelaar 1984). Microcystin concentrations in particles were quantified using enzyme-linked immunosorbent assay (ELISA) (An and Carmichael 1994) after extraction from 75% aqueous methanol. Phytoplankton species abundance and composition were determined via the inverted microscope technique (Uterm?hl 1958) using water samples preserved in 1% Lugol?s solution. Biovolumes for each species were calculated using cell counts and estimates of cell volume based on measurements of cell dimensions. We then converted 34 biovolumes (mm3 L-1) to dry biomass (?g L-1) assuming a specific gravity of 1 g cm-3 and a dry biomass: wet biomass ratio of 0.40 (Riemann et al. 1989; Sarnelle and Wilson 2005; Knoll et al. 2008). We characterized D. pulicaria genotypes for the inoculum used to stock the enclosures to confirm that we added approximately equal densities of each clone and for all D. pulicaria enclosures at the conclusion of the experiment to determine the outcome of intraspecific competition among the clones. We genetically characterized 20-25 randomly selected D. pulicaria individuals from ethanol-preserved macrozooplankton samples for each inoculum subsample and for all Daphnia enclosures at the conclusion of the experiment. D. pulicaria were measured and counted at 40X in a Sedgwick-Rafter cell, and D. pulicaria lengths were converted to biomass using a length-weight regression (O. Sarnelle unpubl.). We used analysis of variance (ANOVA) to compare chlorophyll a, cyanobacterial dry biomass, relative abundance of dominant phytoplankton taxa, and microcystin concentrations for low and high N:P ratios immediately prior to the addition of D. pulicaria (16 October 2009). The effects of N:P ratio and D. pulicaria presence on chlorophyll a and cyanobacterial dry biomass over time were tested using repeated measures ANOVA (sampling date = repeated measures), and pairwise differences among treatments were assessed with Tukey?s test. We used ANOVA to analyze the effects of N:P ratio and D. pulicaria presence on the relative abundance of dominant phytoplankton on 3 November and 10 December 2009. The effect of N:P ratio on D. pulicaria density and biomass at the conclusion of the experiment was tested using ANOVA. We then compared the relative abundance of each D. pulicaria clone initially stocked into the enclosures to its relative abundance at the conclusion of the experiment for low and high N:P ratios using ANOVA and Tukey?s test for pairwise comparisons. ANOVA and Tukey?s test were also used to compare the relative abundance of D. pulicaria clones from oligotrophic and 35 eutrophic lakes stocked into the enclosures versus their relative abundance at the conclusion of the experiment for low and high N:P ratios. Relative abundance data for dominant phytoplankton taxa and D. pulicaria genotypes were arcsine-square-root-transformed prior to all statistical analyses. All other data were checked for normality and homogeneity and log-transformed when necessary. RESULTS N:P ratio had a large and significant positive effect on chlorophyll a, cyanobacterial biomass, and concentrations of the toxin microcystin over the first eleven days of the experiment (Figures 2 and 3). Mean chlorophyll a and cyanobacterial biomass for high N:P enclosures were approximately four- and eight-times higher, respectively, than for low N:P enclosures (chlorophyll a: F1,22 = 19201.816, P < 0.0001; cyanobacterial biomass: F1,22 = 62.285, P < 0.0001), immediately prior to stocking of D. pulicaria genotypes (16 October 2009). Furthermore, the toxic cyanobacterium Microcystis aeruginosa contributed over 20% of total phytoplankton biomass (Figure 4) and mean concentrations of microcystin were over three-times higher at high N:P than low N:P on 16 October 2009 (F1,22 = 8.002, P = 0.010) (Figure 3). In contrast, M. aeruginosa comprised approximately 2% of total phytoplankton biomass and the filamentous cyanobacterium Anabaena circinalis contributed less than 10% of total phytoplankton biomass at low N:P on 16 October 2009 (Figure 4). In addition, the relative abundance of dinoflagellates was significantly higher at low N:P on 16 October 2009 (F1,22 = 22.769, P < 0.0001). By the fourth week of the experiment (~two weeks after D. pulicaria addition - 3 November 2009), D. pulicaria had suppressed cyanobacterial biomass by ~96% relative to no D. pulicaria controls at high N:P (Tukey?s test: P = 0.009). The relative abundance of M. aeruginosa increased to approximately 36% of total phytoplankton biomass in the no D. 36 pulicaria control while D. pulicaria reduced the relative abundance of M. aeruginosa to 2% of total phytoplankton biomass at high N:P (Tukey?s test: P < 0.0001). In contrast, the relative abundance of chlorophytes was increased in D. pulicaria enclosures at high N:P, relative to no D. pulicaria controls at high N:P (Tukey?s test P = 0.010). D. pulicaria had no effect on cyanobacterial biomass at low N:P by 3 November 2009 (Figure 4 - Tukey?s test: P = 1.000). Furthermore, the relative abundance of A. circinalis and other cyanobacteria (Cylindrospermopsis and Chroococcus species) did not differ with and without D. pulicaria at low N:P on 3 November 2009 (Tukey?s test: Anabaena ? P = 0.997; other cyanobacteria ? P = 0.866). On 10 December 2009, the relative abundance of Anabaena and Microcystis was significantly higher for the D. pulicaria/low N:P treatment than for each of the other three treatments (Tukey?s test: P < 0.05 for each pairwise comparison). In general, high N:P enclosures were largely dominated by chlorophytes at the conclusion of the experiment (~88% of total phytoplankton biomass) and M. aeruginosa contributed less than 5% of total phytoplankton biomass regardless of D. pulicaria presence. Repeated-measures ANOVA over the 10-week experiment revealed significant effects of N:P ratio (F1,20 = 31.837, P < 0.0001), Daphnia presence (F1,20 = 8.635, P = 0.008), and the interaction of N:P ratio and Daphnia presence (F1,20 = 8.241, P = 0.0009) on cyanobacterial biomass. Furthermore, N:P ratio (F1,20 = 519.876, P < 0.0001), D. pulicaria presence (F1,20 = 9.931, P = 0.005), and the interaction of N:P ratio and D. pulicaria presence (F1,20 = 6.221, P = 0.022) all had significant effects on chlorophyll a over time. At the conclusion of the experiment, D. pulicaria had reduced chlorophyll a by approximately 46%, relative to the no D. pulicaria control at high N:P (Tukey?s test: P = 0.005). In contrast, D. pulicaria had no effect on chlorophyll a at low N:P (Tukey?s test: P = 0.963). 37 By the conclusion of the experiment, N:P ratio had large effects on the relative abundance of all six D. pulicaria genotypes (Figure 6A; Table 2). Tukey?s test revealed that the relative abundance of Lawrence Lake (P = 0.188), Lake Sixteen (P = 0.832), Warner Lake (P = 0.990), Baker Lake (P = 0.517), MSU Pond 1 (P = 0.166), and Wintergreen Lake (P = 0.723) D. pulicaria genotypes at low N:P did not differ from their relative abundance during stocking. In contrast, the relative abundance of Lawrence Lake, Lake Sixteen, Warner Lake, and Wintergreen Lake D. pulicaria genotypes was significantly lower at high N:P when compared to their relative abundance during stocking (Tukey?s test - Lawrence: P = 0.0187; Sixteen: P = 0.0250; Warner: P < 0.0001). The relative abundance of the Baker Lake D. pulicaria genotype was significantly higher at low N:P than at high N:P (Tukey?s test: P = 0.024); however, the relative abundance of the Baker Lake genotype at high N:P did not differ from its relative abundance during stocking (Tukey?s test: P = 0.541). In contrast, the relative abundance of the MSU Pond 1 D. pulicaria genotype was significantly higher at high N:P than during stocking, and the MSU Pond 1 genotype contributed over 95% of D. pulicaria in high N:P enclosures at the conclusion of the experiment. In general, the relative abundance of Daphnia genotypes from eutrophic lakes (Baker, MSU Pond 1, and Wintergreen) was nearly 100% for high N:P enclosures, while all six D. pulicaria genotypes were equally represented at low N:P (Figure 6B). DISCUSSION It is well known that species frequently trade-off high maximum growth rates for low minimum resource requirements (Tessier et al. 2000). Daphnia species from environments with low mortality risk and low food conditions (e.g., deep lakes) tend to have low minimum resource requirements (i.e., R*) but do poorly in resource-rich environments. In contrast, Daphnia species from environments with high mortality risk and abundant food (e.g., shallow lakes) typically 38 have high per capita reproductive rates but also high minimum resource requirements (Tessier and Woodruff 2002). In addition to variation in food quantity observed across the pond-size and permanence gradient, food quantity and quality also differ across lakes of contrasting trophic status. It is well known that nutrient enrichment of freshwater ecosystems leads to an increase in total phytoplankton biomass (i.e., food quantity) (Dillon and Rigler 1975). Furthermore, food quality may also decline with nutrient enrichment, particularly during the summer months, as phytoplankton communities are increasingly dominated by toxic cyanobacteria with eutrophication (Trimbee and Prepas 1987; Watson et al. 1997; Downing et al. 2001). The interplay between increasing phytoplankton biomass and cyanobacterial dominance with nutrient enrichment creates an interesting framework to study the consequences of the trade-off between ability to exploit resource-rich (with potentially toxic food) versus resource-poor environments. By manipulating N:P ratio, we created two contrasting environments ? one characterized by increased food quantity but also higher levels of cyanobacteria and associated toxins at high N:P (i.e., lower food quality). In contrast, food quantity, cyanobacterial biomass, and toxins were significantly lower at low N:P. The trade-off in resource exploitation across different species of daphnids has been shown to have important consequences for phytoplankton communities in previous studies (Tessier et al. 2000). Our results build on these previous studies by examining a similar trade-off within a single species of Daphnia. We demonstrate that Daphnia may trade-off ability to exploit high resource environments and tolerate toxic cyanobacteria for low minimum resource requirements. Furthermore, the outcome of intraspecific competition among Daphnia genotypes also had important consequences for phytoplankton communities. Daphnia genotypes from eutrophic lakes became dominant at high N:P and had large effects on chlorophyll a and cyanobacterial 39 biomass. In contrast, Daphnia genotypes from oligotrophic and eutrophic lakes were equally represented at low N:P and had no effect on chlorophyll a and cyanobacterial biomass. While Daphnia genotypes from eutrophic lakes became dominant at high N:P, this was driven by the relative abundance of a single genotype (MSU Pond 1) that comprised greater than 90% of all Daphnia genotypes at the conclusion of the experiment. Immediately prior to stocking D. pulicaria genotypes, the toxic cyanobacterium Microcystis aeruginosa contributed over 25% of total phytoplankton biomass. However, the phytoplankton community at high N:P was dominated by relatively nutritious phytoplankton (i.e., small, edible chlorophytes). Therefore, selection for Daphnia genotypes may have been driven by high maximum growth rates on high quantities of good quality food versus tolerance to toxic cyanobacteria. The Baker Lake genotype was most tolerant of toxic cyanobacteria in a previously published laboratory experiment. However, the relative abundance of the Baker Lake genotype was less than 1% of all Daphnia genotypes at the conclusion of the experiment. As we only extensively sampled Daphnia populations at the conclusion of the experiment to minimize contamination of Daphnia genotypes, the dynamics of relative abundances for Daphnia genotypes over time are unknown. Experiments examining how population growth rates and minimum resource requirements of Daphnia are affected by food quantity and quality (or toxicity) would be interesting and could help to identify potential mechanisms driving the results observed here. While Daphnia from environments with low food quantity (e.g., oligotrophic lakes) tend to have low minimum resource requirements, it would be interesting to see if cyanobacteria in the diet reverse the outcome of resource competition in Daphnia. Cyanobacteria have repeatedly been shown to be poor quality food for most zooplankton grazers (including Daphnia) in laboratory studies (Wilson et al. 2006; Wilson and Hay 2007; Tillmanns et al. 2008). Numerous studies have 40 demonstrated that cyanobacteria can inhibit zooplankton feeding, growth, and reproduction through several mechanisms including (1) colonial or filamentous morphologies, (2) nutritional deficiencies, and (3) production of toxins (Porter 1977; Lampert 1987; Wilson et al. 2006). While experimental examinations of resource competition typically manipulate food quantity (Tilman 1982), it is possible that the presence of poor quality food (e.g., cyanobacteria) may also affect the outcome of resource competition. Toxin-producing cyanobacteria have frequently been hypothesized to limit the ability of herbivorous zooplankton (such as Daphnia) to control phytoplankton biomass by inhibiting Daphnia feeding, population growth, and in extreme cases, causing Daphnia mortality (Gilbert 1990; DeMott et al. 1991, Lurling and van der Grinten 2003). Understanding the effects of cyanobacteria that are poor quality and potentially dangerous food for Daphnia consumers on minimum resource requirements may help explain the evolution of Daphnia tolerance to toxic cyanobacteria. While evolutionary biologists have been studying intraspecific trait variation since Darwin, we still have a poor understanding of what maintains this variability within a population. In addition to the intraspecific variation in tolerance to cyanobacteria between Daphnia populations in eutrophic versus oligotrophic environments, there is also considerable variation for traits conferring tolerance to cyanobacteria within eutrophic lakes (Sarnelle and Wilson 2005). Trade- offs associated with the presence (or absence) of cyanobacteria in the environment may help to explain the maintenance variability within Daphnia populations. Assuming tolerance to cyanobacteria is costly in the absence of cyanobacteria, it would be interesting to determine whether there are seasonal changes in Daphnia tolerance to cyanobacteria as a result of phytoplankton succession. 41 Ecological genetics is a rapidly growing field which merges aspects of population genetics and community ecology to understand the consequences of intraspecific variation for communities and ecosystems (Whitham et al. 2006; Wade 2007). In particular, intraspecific variation within keystone and foundation species is expected to have large effects on ecological interactions (Whitham et al. 2006; Post et al. 2008). In freshwater ecosystems, rapid adaptive evolution by major consumers may play an important role in the response of lake ecosystems to cultural eutrophication and food web manipulations (Hairston et al. 2001; Sarnelle and Wilson 2005). Hairston et al. (1999) made the seminal observation that a generalist herbivore in freshwater lakes (Daphnia) can evolve to tolerate toxic cyanobacteria in the diet in response to nutrient enrichment. Daphnia tolerance to toxic cyanobacteria has now been observed for a D. galeata population following eutrophication of Lake Constance in Europe (Hairston et al. 2001) and for D. pulicaria populations in North America (Sarnelle and Wilson 2005). Understanding the causes and consequences of intraspecific variation within Daphnia may aid in the future management of eutrophic lakes, and we can now accurately characterize intraspecific genetic variation within and across populations of Daphnia (Colbourne et al. 2004). An emphasis on intraspecific trait variation provides an interesting conceptual framework for linking diversity to ecosystem function, and previous studies suggest that this approach may be profitable, particularly for plankton food webs (Tessier et al. 2000; Post et al. 2008; Duffy et al. 2010). 42 REFERENCES An, J. S., and W. W. Carmichael. 1994. Use of a colorimetric protein phosphatase inhibition assay and enzyme-linked-immunosorbent-assay for the study of microcystins and nodularins. Toxicon 32: 1495-1507. Boyd, C. E., and J. L. Shelton. 1984. Observations on the hydrology and morphometry of ponds on the Auburn University Fisheries Research Unit. Alabama Agricultural Experiment Station Bulletin: 1-64. Carmichael, W. W. 1992. Cyanobacteria secondary metabolites - the cyanotoxins. Journal of Applied Bacteriology 72: 445-459. Colbourne, J. K., B. Robison, K. Bogart, and M. Lynch. 2004. Five hundred and twenty-eight microsatellite markers for ecological genomic investigations using Daphnia. Molecular Ecology Notes 4: 485-490. DeMott, W. R., Q. X. Zhang, and W. W. Carmichael. 1991. Effects of toxic cyanobacteria and purified toxins on the survival and feeding of a copepod and 3 species of Daphnia. Limnol. Oceanogr. 36: 1346-1357. Desmarais, K. H., and A. J. Tessier. 1999. Performance trade-off across a natural resource gradient. Oecologia 120: 137-146. Dillon, P. J. and F. H. Rigler. 1974. The phosphorus-chlorophyll relationship in lakes. Limnol. Oceanogr. 19: 767-773. Downing, J. A., S. B. Watson, and E. McCauley. 2001. Predicting cyanobacteria dominance in lakes. Canadian Journal of Fisheries and Aquatic Sciences 58: 1905-1908. Duffy, M. A. 2010. Ecological consequences of intraspecific variation in lake Daphnia. Freshwater Biology 55: 995-1004. 43 Gilbert, J. J. 1990. Differential effects of Anabaena affinis on cladocerans and rotifers: mechanisms and implications. Ecology 71: 1727-1740. Hairston, N. G., W. Lampert, C. E. Caceres, C. L. Holtmeier, L. J. Weider, U. Gaedke, J. M. Fischer, J. A. Fox, and D. M. Post. 1999. Lake ecosystems - Rapid evolution revealed by dormant eggs. Nature 401: 446-446. Hairston, N. G., C. L. Holtmeier, W. Lampert, L. J. Weider, D. M. Post, J. M. Fischer, C. E. Caceres, J. A. Fox, and U. Gaedke. 2001. Natural selection for grazer resistance to toxic cyanobacteria: Evolution of phenotypic plasticity? Evolution 55: 2203-2214. Hay, M. E., and W. Fenical. 1988. Marine plant-herbivore interactions: the ecology of chemical defense. Annual Review of Ecology and Systematics 19: 111-145. Howe, H. F. and L. C. Westley. 1988. Ecological relationships of plants and animals. Oxford Univ. Press. Karban, R., and A. A. Agrawal. 2002. Herbivore offense. Annual Review of Ecology and Systematics 33: 641-664. Kirk, K. L., and J. J. Gilbert. 1992. Variation in herbivore response to chemical defenses: zooplankton foraging on toxic cyanobacteria. Ecology 73: 2208-2217. Knoll, L. B. S., O.; Hamilton, S. K.; Kissman, C. E. H., Wilson, A. E., Rose, J. B., and Morgan, M. R. 2008. Corrigendum: Invasive zebra mussels (Dreissena polymorpha) increase cyanobacterial toxin concentrations in low-nutrient lakes. Canadian Journal of Fisheries and Aquatic Sciences 65: 2552-2552. Lampert, W. 1987. Laboratory studies on zooplankton-cyanobacteria interactions. New Zealand Journal of Marine and Freshwater Research 21: 483-490. Leibold, M. A. 1989. Resource edibility and the effects of predators and productivity on the 44 outcome of trophic interactions. American Naturalist 134: 922-949. Leibold, M. A., J. M. Chase, J. B. Shurin, and A. L. Downing. 1997. Species turnover and the regulation of trophic structure. Annual Review of Ecology and Systematics 28: 467-494. L?rling, M., and E. Van Der Grinten. 2003. Life-history characteristics of Daphnia exposed to dissolved microcystin-LR and to the cyanobacterium Microcystis aeruginosa with and without microcystins. Environmental Toxicology and Chemistry 22: 1281-1287. Paerl, H. W. 1988. Nuisance phytoplankton blooms in coastal, estuarine, and inland waters. Limnol. Oceanogr. 33: 823-847. Porter, K. G. 1977. The plant-animal interface in freshwater ecosystems. American Scientist 65: 159-170. Post, D. M., E. P. Palkovacs, E. G. Schielke, and S. I. Dodson. 2008. Intraspecific variation in a predator affects community structure and cascading trophic interactions. Ecology 89: 2019- 2032. Riemann, B., P. Simonsen, and L. Stensgaard. 1989. The carbon and chlorophyll content of phytoplankton from various nutrient regimes. Journal of Plankton Research 11: 1037-1045. Sarnelle, O., and A. E. Wilson. 2005. Local adaptation of Daphnia pulicaria to toxic cyanobacteria. Limnol. Oceanogr. 50: 1565-1570. Sartory, D. P., and J. U. Grobbelaar. 1984. Extraction of chlorophyll-a from fresh-water phytoplankton for spectrophotometric analysis. Hydrobiologia 114: 177-187. Schindler, D. W. 2006. Recent advances in the understanding and management of eutrophication. Limnol. Oceanogr. 51: 356-363. Smith, V. H. 1983. Low nitrogen to phosphorus ratios favor dominance by blue-green-algae in lake phytoplankton. Science 221: 669-671. 45 Tessier, A. J., and P. Woodruff. 2002. Trading off the ability to exploit rich versus poor food quality. Ecology Letters 5: 685-692. Tessier, A. J., M. A. Leibold, and J. Tsao. 2000. A fundamental trade-off in resource exploitation by Daphnia and consequences to plankton communities Ecology 81: 826-841. Thompson, J. N., O. J. Reichman, P. J. Morin, G. A. Polis, M. E. Power, R. W. Sterner, C. A. Couch, L. Gough, R. Holt, D. U. Hooper, F. Keesing, C. R. Lovell, B. T. Milne, M. C. Molles, D. W. Roberts, and S. Y. Strauss. 2001. Frontiers of ecology. BioScience 51: 15-24. Tilman, D. 1982. Resource competition and community structure. Princeton Univ. Press. Tillmanns, A. R., A. E. Wilson, F. R. Pick, and O. Sarnelle. 2008. Meta-analysis of cyanobacterial effects on zooplankton population growth rate: species-specific responses. Fundamental and Applied Limnology 171: 285-295. Tollrian, R., and C. D. Harvell. 1999. The ecology and evolution of inducible defenses. Princeton Univ. Press. Trimbee, A. M., and E. E. Prepas. 1987. Evaluation of total phosphorus as a predictor of the relative biomass of blue-green algae with emphasis on Alberta lakes. Canadian Journal of Fisheries and Aquatic Sciences 44: 1337-1342. Turner, A. M., and G. G. Mittelbach. 1990. Predator avoidance and community structure: interactions among piscivores, planktivores, and plankton. Ecology 71: 2241-2254. Uterm?hl, H. 1958. Zur Vervollkommnung der quantitativen Phytoplankton-Methodik. Mitt. int. Ver. theor. angew. Limnol. 9: 1-38. Vanderploeg, H. A., J. R. Liebig, W. W. Carmichael, M. A. Agy, T. H. Johengen, G. L. Fahnenstiel, and T. F. Nalepa. 2001. Zebra mussel (Dreissena polymorpha) selective filtration promoted toxic Microcystis blooms in Saginaw Bay (Lake Huron) and Lake Erie. Canadian 46 Journal of Fisheries and Aquatic Sciences 58: 1208-1221. Wade, M. J. 2007. The co-evolutionary genetics of ecological communities. Nat. Rev. Genet. 8: 185-195. Watson, S. B., E. Mccauley, and J. A. Downing. 1997. Patterns in phytoplankton taxonomic composition across temperate lakes of differing nutrient status. Limnol. Oceanogr. 42: 487- 495. Werner, E. E., and S. D. Peacor. 2003. A review of trait-mediated indirect interactions in ecological communities. Ecology 84: 1083-1100. Whitham, T. G., J. K. Bailey, J. A. Schweitzer, S. M. Shuster, R. K. Bangert, C. J. Leroy, E. V. Lonsdorf, G. J. Allan, S. P. DiFazio, B. M. Potts, D. G. Fischer, C. A. Gehring, R. L. Lindroth, J. C. Marks, S. C. Hart, G. M. Wimp, and S. C. Wooley. 2006. A framework for community and ecosystem genetics: from genes to ecosystems. Nat. Rev. Genet. 7: 510-523. Wilson, A. E., and M. E. Hay. 2007. A direct test of cyanobacterial chemical defense: Variable effects of microcystin-treated food on two Daphnia pulicaria clones. Limnol. Oceanogr. 52: 1467-1479. Wilson, A. E., O. Sarnelle, and A. R. Tillmanns. 2006. Effects of cyanobacterial toxicity and morphology on the population growth of freshwater zooplankton: Meta-analyses of laboratory experiments. Limnol. Oceanogr. 51: 1915-1924. 47 TABLES Table 2.1. Source lakes and genetic characterizations of the six D. pulicaria clones. Three D. pulicaria clones were collected from oligotrophic lakes where cyanobacteria were absent (Lawrence, Sixteen, and Warner), and three clones were collected from eutrophic lakes with abundant cyanobacteria (Baker, MSU Pond 1, and Wintergreen). Each D. pulicaria clone was characterized using two microsatellite markers (Dp3 and Dp339), and microsatellite nucleotide lengths were determined for both loci. Location Total Dp3 Dp339 Lake Lat. N, Long. W Phosphorus (?g l-1) length (bp) length (bp) Lawrence 42o26?27?, 85o21?03? 8-10 282, 291 175, 180 Sixteen 42o33?90?, 85o36?80? 9-12 282, 286 175, 180 Warner 42o28?16?, 85o31?30? 12-14 282, 282 175, 180 Baker 42o26?27?, 85o21?03? 21-40 291, 291 175, 180 Wintergreen 42o23?50?, 85o23?07? 50-70 282, 282 175, 175 MSU Pond 1 42o40?53?, 84o28?57? 170-300 282, 286 175, 190 Table 2.2. Effect of N:P ratio on the relative abundance of each of the six Daphnia pulicaria genotypes at the conclusion of the experiment (10 December 2009). D. pulicaria genotype Lake type DF F P Lawrence Lake Oligotrophic 2 23.757 <0.0001 Lake Sixteen Oligotrophic 2 28.210 <0.0001 Warner Lake Oligotrophic 2 13.807 0.001 Baker Lake Eutrophic 2 4.912 0.030 MSU Pond 1 Eutrophic 2 330.668 <0.0001 Wintergreen Lake Eutrophic 2 20.553 <0.0001 Error 11 48 FIGURE LEGENDS Figure 1. Mean juvenile somatic growth rates for the six Daphnia pulicaria genotypes on two diets: a nutritious green alga (100 % Ankistrodesmus) and toxic cyanobacteria (100% Microcystis aeruginosa). Error bars represent one standard error. Data are from Sarnelle and Wilson 2005. Figure 2. Dynamics of (A) chlorophyll a and (B) cyanobacterial dry biomass over the ten-week enclosure experiment. Data represent means ? one standard error for each treatment. Figure 3. Concentration of the toxin microcystin for each nitrogen-to-phosphorus (N:P) treatment immediately prior to stocking of Daphnia pulicaria genotypes (16 October 2009). Data represent means ? one standard error. Figure 4. Relative biomass of phytoplankton taxa at the beginning (16 October 2009), middle (3 November 2009), and end (10 December 2009) of the ten-week enclosure experiment. Data represent means for phytoplankton taxa. Since Daphnia pulicaria were added on 17 October 2009, relative biomass of phytoplankton taxa for ?No Daphnia? and ?Daphnia? enclosures were pooled for each N:P ratio on 16 October 2009. Figure 5. Daphnia pulicaria (A) density and (B) biomass for each nitrogen-to-phosphorus (N:P) treatment at the conclusion of the experiment (10 December 2009). Data represent means ? one standard error for each N:P treatment. 49 Figure 6. Relative abundance of (A) each Daphnia pulicaria genotype and (B) D. pulicaria genotypes from oligotrophic and eutrophic lakes at the conclusion of the experiment (10 December 2009). Data represent means ? one standard error for each N:P treatment. 50 51 52 53 54 55