Beschreibung: In this doctoral project, I investigated the effects of abiotic factors on the composition and cell density of epibacterial communities associated with the ecologically important brown macroalga Fucus vesiculosus in laboratory and outdoor experiments. During the experiments, specimens of the macroalgal host and their microbial biofilm were exposed to different levels of either temperature, irradiance, or salinity. Subsequently, bacteria were harvested from the macroalga, and bacterial community composition was analyzed by culture-independent analyses of 16S rRNA gene sequences. Cell densities were enumerated via epifluorescence microscopy directly on the algal surface. Furthermore, I studied the effects of natural microbial assemblages on the attachment rates of barnacle larvae. For this purpose, the microbial biofilm community was harvested from the macroalgal surface and exposed to barnacle larvae during their peak settlement season in the Kiel Fjord in a specially constructed bioassay apparatus. As revealed by redundancy analyses, epibacterial communities significantly clustered along the temperature (5, 10, 15, 20, 25°C) and salinity (5, 19, 25) gradients but not along the gradient of irradiance (100%, 44%, 23%, 12%, 5%, and 0% of sunlight reaching the algal surface). Temperature and salinity are important factors in structuring epibacterial communities since they each accounted for about 20% of the variation in the bacterial community composition. In general, 〈10% of bacterial OTUs (operational taxonomic units grouped at the 97% sequence similarity level) siginificantly correlated with the abiotic factors. The differences in community composition due to temperature were largely attributed to the Rhodobacteraceae, the bacterial family whose mean relative abundance was twice as high at the high temperatures than at the low temperatures. Members of the subphylum Betaproteobacteria made up about a third of the bacterial community at the low salinity level (5) on average and were nearly absent at the higher salinity levels. Members of the Gammaproteobacteria and of the Actinobacteria were on average at least twice as abundant at the higher salinities than at the low salinity. Furthermore, the type of substrate (stones were included in the salinity experiment as non-living reference substrate) had a significant effect on the composition of epibacteria. Epiphytic and epilithic bacterial communities differed significantly due to a higher mean relative percentage of members of the phylum Cyanobacteria. They made up about 8% on average in the epilithic community, while they were nearly absent on F. vesiculosus. Furthermore, members of the subphylum Gammaproteobacteria were at least twice as abundant in epiphytic than in epilithic bacterial communities. Epibacterial richness (i.e., the number of OTUs) significantly differed between treatment levels, although variability between algal individuals of one treatment level was generally high. OUT richness was highest at 15°C, lowest at a salinity of 5, and lowest at 44% irradiance reaching the algal surface. The most even distribution of epibacterial groups (evenness expressed with the Inverse Simpson Index) showed similar patterns as richness, although in the temperature experiment this was a non-significant trend. Epibacterial densities did not significantly differ between levels of temperature but decreased with shading. However, variability between individuals of one treatment level was often larger than between replicates of one treatment level. Whether direct or indirect effects via altered biotic interactions may have caused the observed shifts is discussed in the appropriate sections. In bioassays conducted in the Kiel Fjord, barnacle cyprids attached significantly less to filters treated with microbial assemblages harvested from F. vesiculosus than to non-treated seawater filters. This indicates a repellent effect of the microbes. The most pronounced repellent effect was found with microbial assemblages that had been kept at 15°C for 14 days (54% fewer attached cyprids than on filters without microbes). My study contributes to the basic understanding of the ecology of host-associated bacterial communities and the interactions between host, epibacteria, and macrofoulers: with the tool of high-throughput sequencing, high bacterial diversity for this macroalgal host was confirmed and a detailed description of community composition under different abiotic conditions was achieved. The impact that abiotic factors can have on this highly complex epibacterial community has been elucidated and may become more important in the future taking into consideration predicted climate change scenarios. The suspected potential of the epibacterial biofilm of playing a mutualistic role for the macroalgal host by assisting the host in its own defense against macrofouling was confirmed, adding to the notion that microbial communities should be considered when studying macroalgae and other marine hosts.

Beschreibung: Alga-herbivore interactions have a great impact on the structure of coastal benthic communities. Herbivores, of different sizes and belonging to different taxa, can exert severe grazing pressure, thus, influencing seaweed abundance and distribution. This top-down control is counteracted by algal defense mechanisms, many of which are chemical deterrents. Previous studies have unveiled the existence of defenses which are triggered by grazer attack. These so-called inducible defenses have been reported for all alga divisions and have been detected in many temperate and tropical coastal regions. However, little is known about inducible defenses in macroalgae from the Lower St. Lawrence Estuary, East Canada. In a series of laboratory experiments, I investigated the potential of four brown and two red macroalgae species from the Lower St. Lawrence Estuary to regulate defenses against direct grazing by two littorinid snail and one amphipod species. Two-choice feeding assays with live algal material were conducted to detect changes in algal palatability after exposure to grazers for a time span of 2 to 4 weeks. Algal material that was kept in the absence of grazers served as controls. Grazer preference for the control over the previously grazed algae tissue should indicate defense induction. Interestingly, none of the screened algae species reduced their palatability in response to herbivore attack. Furthermore, consumption rates of the gastropod L. littorea feeding on Fucus vesiculosus were lower than has been reported for other study systems, possibly due to a constitutive defense of the algae, among other factors. A chemical analysis, however, did not confirm such a defense. This study adds information to the concept of inducible defenses with regard to its reported variability by suggesting that the absence of inducible defenses at the Lower St. Lawrence Estuary may reflect a large scale intraspecific biogeographic variation in defense induction. However, other factors that may have lead to the absence of defense induction must be discussed.