bioRxiv preprint doi: https://doi.org/10.1101/111559; this version posted February 24, 2017. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. Eelgrass leaf surface microbiomes are locally variable and highly correlated with epibiotic eukaryotes Mia M. Bengtsson1*, Anton Bühler2, Anne Brauer1, Sven Dahlke3, Hendrik Schubert2 and Irmgard Blindow3 1 Institute of Microbiology, University of Greifswald, Greifswald, Germany 2 Institut für Biowissenschaften, University of Rostock, Rostock, Germany 3 Biological Station of Hiddensee, University of Greifswald, Kloster, Germany *correspondence: [email protected] ABSTRACT Eelgrass (Zostera marina) is a marine foundation species essential for coastal ecosystem services around the northern hemisphere. Like all macroscopic organisms, it possesses a microbiome which may play critical roles in modulating the interaction of eelgrass with its environment. For example, its leaf surface microbiome could inhibit or attract eukaryotic epibionts which may overgrow the eelgrass leading to reduced primary productivity and subsequent eelgrass meadow decline. We used amplicon sequencing of the 16S and 18S rRNA genes of prokaryotes and eukaryotes to assess the leaf surface microbiome (prokaryotes) as well as eukaryotic epibionts in- and outside lagoons on the German Baltic Sea coast. Bacterial microbiomes varied substantially both between sites inside lagoons and between open coastal and lagoon sites. Water depth, leaf area and biofilm chlorophyll a concentration explained a large amount of variation in both bacterial and eukaryotic community composition. Communities of bacterial and eukaryotic epibionts were highly correlated, and network analysis revealed disproportionate co-occurrence between a limited number of eukaryotic taxa and several bacterial taxa. This suggests that eelgrass leaf surface biofilms are a mosaic of the microbiomes of several eukaryotes, in addition to that of the eelgrass itself, and underlines that eukaryotic microbial diversity should be taken into account in order to explain microbiome assembly and dynamics in aquatic environments. INTRODUCTION invertebrates, algae, fish and microorganisms. Due to its ability to tolerate Seagrasses are aquatic flowering plants that low and fluctuating salinity levels, eelgrass form underwater meadows critically is common in estuaries as well as in the important for coastal ecosystems around the biggest brackish water habitat in the world, world. Seagrass meadows are nursing the Baltic Sea. However, significant grounds for juvenile fish which can hide and declines in the depth limit and areal cover of forage between the seagrass leaves, eelgrass meadows in the Baltic Sea have sediments are stabilized by the seagrass been observed since several decades roots and the biomass of seagrass and (Boström et al., 2014). This echoes the associated organisms sequester carbon with distressing situation for seagrass ecosystems implications for climate change mitigation around the world, which experience threats (Fourqurean et al., 2012). The seagrass by human activities and global change (Orth eelgrass (Zostera marina) is an important et al., 2006). foundation species along soft-bottom coasts The mechanisms that are responsible in the northern hemisphere. Eelgrass for eelgrass meadow decline appear to meadows are biodiversity hotspots, depend on many different factors. providing a home to a myriad of Historically, outbreaks of the pathogenic 1 bioRxiv preprint doi: https://doi.org/10.1101/111559; this version posted February 24, 2017. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. protist Labyrinthula zosterae (“wasting has determined that bacterial abundance and disease”) has caused catastrophic die-off of productivity on eelgrass leaves vary during eelgrass (Muehlstein et al., 1991), although the year, with a peak in early autumn its importance as a pathogen under current (Tornblom and Sondergaard, 1999), and that conditions in Europe seems limited (Brakel eelgrass leaves have an associated bacterial et al., 2014). Instead, factors such as water community (the “microbiome”), with only clarity, eutrophication, grazing pressure, and some compositional overlap with other interspecific competition have been aquatic macrophytes (Crump and Koch, identified as culprits for eelgrass growth 2008). (Baden et al., 2010; Duffy et al., 2015). There is currently little known about Especially important is the competition what factors influence eelgrass between eelgrass and algae that grow within microbiomes, or what functions the bacteria meadows, often as epiphytes on the eelgrass have on eelgrass leaves. Bacteria are likely itself. In some cases, eelgrass becomes to play a fundamental role in the competition extensively covered with a mixture of algae, of eelgrass and epibionts such as algae. For bacteria and sessile animals, which shade it example, certain bacteria may inhibit the and inhibit transport of solutes (Brodersen et attachment of algal spores while others may al., 2015) and may over time cause eelgrass promote further colonization (Celdrán et al., meadows to degrade. Recent research 2012; Mieszkin et al., 2013). Conversely, suggests that relative success of eelgrass and epibiotic eukaryotes including algae may in associated algae is determined by complex turn shape the bacterial communities on interactions between biotic processes such eelgrass leaves by release of bacterial as grazing on both algae and eelgrass by attractants or deterrents (Steinberg and de invertebrates, and environmental factors Nys, 2002) or through selective grazing such as nutrient concentrations and (Huws et al., 2005), for example. Biotic and temperature (Alsterberg et al., 2013; Eklöf abiotic environmental factors as well as et al., 2012). However, these mechanisms host-related factors such as eelgrass are not well understood and their complexity productivity and genotype are also likely to necessitates a holistic ecosystem approach shape the eelgrass microbiome, and thereby taking into account several organism groups modulate its function. that inhabit seagrass meadows and their In this study, we aimed to obtain a interactions to be resolved (Boström et al., first view onto eelgrass leaf microbiomes by 2014; Maxwell et al., 2016). investigating the community composition of An overlooked group of organisms prokaryotic and eukaryotic epibiotic in eelgrass meadows is the bacteria, which organisms in relation to abiotic and biotic cover eelgrass leaves and roots forming environmental variables. We sampled biofilms, are responsible for degradation of eelgrass in semi-sheltered lagoons and along eelgrass detritus and are also associated to exposed open shorelines around the Island all other organisms in eelgrass beds of Hiddensee on the eastern German Baltic including algae and other epiphytes. Sea coast and used high-throughput Illumina Bacteria are the first colonizers on new amplicon sequencing of the 16S and 18S seagrass leaves and thus initiate a rRNA genes. We hypothesized that (1) both successional process that may end with bacterial and eukaryotic communities would severe epiphytic overgrowth. Bacteria are vary substantially between lagoon and open also an important food source for microbial coast sites due to different abiotic conditions eukaryotic and invertebrate grazers and may and that (2) bacterial microbiome therefore support a substantial part of the composition would be influenced by the foodweb found within eelgrass meadows. composition of epibiotic eukaryotes. Research on eelgrass leaf bacterial communities is very limited, but early work 2 bioRxiv preprint doi: https://doi.org/10.1101/111559; this version posted February 24, 2017. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. Denmark Sweden lower average salinity (8.8 ± 1.0 PSU, range Open coast N º Libben sites 6.5 – 13.4 PSU) compared to the open coast 54.6 (Libben bay, 9.4 ±1.6 PSU, range 6.8 – 15.9 PSU). Nutrient levels are elevated in Germany Vitter Bodden Lagoon lagoons (total N: 38.2 ±12.3, total P: 1.2 Hiddensee sites ±0.62 µmol l-1) compared to open coast Rügen waters (total N 19.9 ±4.3, total P: 0.91 ±0.4 N º -1 Schaproder Bodden VBL1 µmol l ) mainly due to agricultural runoff 54.5 VBL2 Baltic Sea 0 2 4 km SBL1 and other human activities (Schiewer, SBL2 OC1 2008). Salinity and nutrient values are WGS84 32N 1: 340 000 OC2 Mapsource: OpenStreetMap OC3 yearly averages from monitoring data 2005 13.0ºE 13.2ºE – 2014 (Landesamt für Umwelt, Figure 1: Location of sampling sites around the Naturschutz und Geologie Mecklenburg- island of Hiddensee on the German Baltic Sea coast. The inset shows the geographic placement of the Vorpommern, unpublished data). area. Sampling: Zostera marina shoots were sampled via scuba diving at 7 sites around the island of Hiddensee (Fig. 1). At every site, 5 replicate shoots (above-ground parts) MATERIALS & METHODS were collected within an area of Study area: We chose to perform our approximately 1 m2. Shoots