The smell of water : grazer-induced colony formation in Scenedesmus

In aquatic systems, the phytoplankton - zooplankton relation is of major importance because it is the first step in the pelagic food chain. It is well known that zooplankton feed with a highly variable success on phytoplankton, primarily owing to algal characteristics such as size, shape, cell wall texture, nutritional quality and toxicity. Algae are present in a broad variety of shapes and may express an enormous variability in their morphology, physiology and behavior depending on environmental variables. Because algae depend on solar energy they have to remain in the upper water layers as long as possible. Moreover, they have to compete with other algae for dissolved nutrients. This means that in an aquatic environment selection pressure exists for small organisms since these have the most efficient uptake of nutrients and light and lowest sinking losses. By contrast, mortality through grazing by an entire assemblage of protozoan and metazoan grazers will exert a strong selection for traits that reduce this mortality through grazing. An effective way to resist grazing is by a dramatic increase in size. However, this confronts the algae with conflicting allometries of selection pressures.Since algae are small relative to their predatory enemies, they may not survive an encounter with a grazer. Therefore, it may be profitable to detect a grazer before they encounter each other in order to elicit a defensive strategy. In a predictable environment temperature and day length could be good predictor of danger. However, in aquatic systems, grazing fluctuates considerably on temporal and spatial scales and chemical cues may be used instead. All organisms exchange constantly chemicals with their environment and those chemicals that are essential in the biology of the grazer and are detectable by the algae may prove potential indicators of danger. They convey information and are referred to as infochemicals. This thesis focuses on the role of infochemicals in the interaction between algae and zooplankton, with emphasis on the <em>Scenedesmus</em> (algae) - <em>Daphnia</em> (waterflea) relation.In the presence of filtered medium from a <em>Daphnia</em> culture, the non-spiny <em>Scenedesmus acutus</em> formed numerous eight-celled colonies (coenobia) (Chapter 2). However, in control populations, i.e. in the absence of <em>Daphnia</em> -infochemical, <em>S. acutus</em> remained unicellular and formed only four-celled colonies when cultures reached stationary phase. The induced colony formation appeared reversible as eight-celled colonies gradually disappeared from the treated populations.A prerequisite for further exploration of the phenomenon of <em>Daphnia</em> -induced colony formation is the development of a reliable biotest (Chapter 3). Inoculum algal density, carbon availability and filter-type are some of the factors that affected the <em>Daphnia</em> -induced colony formation. Analysis of filter extractables revealed that at least two detergents might cause <em>S. acutus</em> to shift rapidly from a completely unicellular population to one dominated with colonies. The production of the <em>Daphnia</em> -infochemical is related to the amount of food processed by the animals. Starved animals or animals fed with ingestible but non-digestible beads were ineffective in inducing colonies. Neither algal homogenates nor auxins and several organic carbon sources have colony inducing activity. The infochemical does seem to originate from the <em>Daphnia</em> -food interaction, or better from the grazer-algal food interaction as several herbivorous zooplankters were able to induce colonies in <em>S. acutus</em> , whereas carnivorous zooplankton and fish were ineffective (Chapter 3). Simple excretion products, such as ammonia and urea alone or in combination with organic carbon sources were ineffective as colony inducing agents (Chapters 3 & 4).<em>Scenedesmus</em> plasticity has, however, not only been shaped by the activity of grazers, but also by other selective forces. Several factors are known that may influence the growth and morphological development in <em>Scenedesmus</em> and among them nutrient availability (Chapter 4) and temperature (Chapter 5) are important ones. In culture, with relatively high algal densities carbon limitation may occur. The availability of inorganic carbon appeared ineffective in inducing colonies, but had a clear effect on cell size. Neither N- nor P-limitation resulted in the formation of numerous, eight-celled coenobia. In general, under nutrient limitation cultures were dominated by unicells. However, despite the limitation, by adding <em>Daphnia</em> water, colonies still could be induced. One of the criticisms on use of artificial growth media is the excessive amount of nutrients in most of them. However, using media of various strengths showed no differences in morphological appearance of <em>S. acutus</em> , both in the absence and presence of <em>Daphnia</em> water.It appears that as long as cell division is not hampered grazer-induced colony formation may occur.Temperature not only affected growth, but also the morphological development in <em>S. acutus</em> (Chapter 5). At low temperatures growth was reduced, but cell- and colony size increased. Under a broad range of temperatures from 9° to 29°C, the addition of <em>Daphnia</em> water significantly increased the proportion of eight-celled coenobia. The smaller size at higher temperature supports the hypothesis of a trade-off between sinking and size.An analysis of potential costs associated with grazer-induced colony formation was initially directed on metabolic costs (Chapter 6). However, no reductions in growth and photosystem II efficiency were detected in induced colonies. Higher sinking losses of induced colonial <em>Scenedesmus</em> populations were measured. Hence, costs may be assigned to enhanced sinking out of the euphotic zone into darker and colder water layers, thereby significantly reducing growth rates. The strategy may, however, not be completely lethal, as <em>Scenedesmus</em> is known to be capable of surviving for prolonged periods on the sediments.One of the prerequisites for interpreting the grazer-induced colony formation as an induced defense is that the response has to be effective in reducing mortality through grazing. In Chapter 7, the grazing success of several zooplankton species, such as the rotifers <em>Keratella</em> and <em>Brachionus</em> and the cladocerans <em>Bosmina</em> , <em>Ceriodaphnia</em> and <em>Daphnia</em> , was analyzed. Food intake was reduced in all smaller grazers, but not in the largest <em>Daphnia</em> species. Moreover, growth of the small <em>Daphnia cucullata</em> was reduced when offered colonial <em>S. acutus</em> . These reductions appeared not the result of an altered biochemical composition of induced colonies.So far, the effect of grazing-associated infochemicals had only been examined for the non-spiny <em>S. acutus</em> . In Chapter 8, 23 different <em>Scenedesmus</em> strains, 9 different other chlorophytes, 2 diatom species and 5 strains of cyanobacteria were investigated. In 35% of the <em>Scenedesmus</em> a positive response to the addition of <em>Daphnia</em> water was observed. Most responding appeared the non-spiny strains, i.e. 64% in contrast to the 4% for spined <em>Scenedesmus</em> . Not only is the trait colony formation only one of the potential defensive traits, it also appeared that the biotest was only suited for examining non-spiny <em>Scenedesmus</em> .The grazer-induced colony formation appeared not to be restricted to the genus <em>Scenedesmus,</em> since two <em>Coelastrum</em> strains were responsive too. Also in the diatom <em>Synedra</em> and the cyanobacterium <em>Microcystis</em> cell size was increased in the presence of <em>Daphnia</em> water. Moreover, the latter showed a tendency to higher toxin levels when cultured in the presence of medium from a <em>Daphnia</em> culture (Chapter 8).Another phenomenon often observed in the presence of live <em>Daphnia</em> , is the aggregation of live cells onto fecal pellets (Chapter 8). These large aggregates will undoubtedly be inedible to grazers and may be an additional process affecting the energy flow from algae to their consumers.In Chapter 9, experiments were performed to evaluate the ability of <em>Daphnia</em> to locate algae by means of chemical cues. No evidence for such a mechanism was detected. However, the animals did seem to avoid water with odors from congeners. Moreover, water from crowded <em>Daphnia</em> cultures had clear effects on growth and reproduction in two <em>Daphnia</em> species and may have an effect on the phytoplankton-grazer interaction.Summarizing the various experiments described in this thesis, the phenomenon of grazer-induced colony formation in <em>Scenedesmus</em> can be interpreted as an inducible defense at the expense of higher sinking losses. The phenomenon is not restricted to <em>Scenedesmus</em> and because of the enormous plasticity in phytoplankton, numerous species may eventually turn out not only to respond to abiotic but to biotic agents as well.