Interaction of the Marine Bacterium Marinobacter Adhaerens HP15 with the Diatom Thalassiosira Weissflogii

Home , Alteromonadaceae, Marinobacter, Planktotalea

Interaction of the Marine Bacterium Marinobacter adhaerens HP15

with the Diatom Thalassiosira weissflogii Analyzed by

Proteomics Approaches

Antje Stahl

A thesis submitted in partial fulfillment of the requirements for the degree of

Doctor of Philosophy in Marine Microbiology

Dissertation Committee

Prof. Dr. Matthias Ullrich Jacobs University Bremen

Prof. Dr. Frank Oliver Glöckner Jacobs University Bremen

Prof. Dr. Jens Harder Max-Planck-Institute for Marine Microbiology

Date of Defense: December 8, 2015

Contents

Acknowledgements ...... III Summary ...... V List of Abbreviations ...... VII 1. Introduction ...... 1 1.1. Oceanic carbon cycling ...... 1 1.1.1. Carbon fluxes and the biological carbon pump ...... 4 1.1.2. Features of marine aggregates and marine snow ...... 4 1.1.3. Transparent exopolymer particles trigger aggregate formation ...... 6 1.2. Bacteria-diatom interactions ...... 8 1.2.1. Diatoms are colonized by bacteria...... 8 1.2.2. Bacteria-diatom interactions and the global carbon cycling ...... 9 1.2.3. Bacterial chemotaxis promotes the encountering of nutritional hot spots ...... 9 1.2.4. Bacteria-diatom interaction on a molecular level ...... 11 1.2.5. Marinobacter adhaerens HP15 – Thalassiosira weissflogii : A model system ...... to understand bacteria-diatom interactions ...... 14 1.3. Heavy metal resistance in bacteria ...... 17 1.3.1. Bacterial strategies to cope with environmental heavy metal stress ...... 17 1.3.2. Marinobacter – a genus inhabiting environments, enriched in heavy metals? ...... 19 2. Aims of the Present Study ...... 21 3. Results ...... 23 3.1. Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii ...... 25 3.2. Stereo-tracking of chemosensing-deficient and motility impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach ...... 51 3.3. Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification ...... 69 4. Discussion and Future Scope ...... 93 4.1. Understanding Marinobacter adhaerens HP15 – Thalassiosira weissflogii interaction and suggestions on methodical improvements ...... 93 4.2. Chemotaxis and the colonization of marine particles ...... 99 4.3. Heavy metals resistance in Marinobacter adhaerens HP15 and further genus members ... 101 5. References ...... 103 6. Supplementary – Tables and Figures ...... 115 Statutory Declaration ...... 127

II Acknowledgements

Acknowledgements

I would like to thank my main supervisor Prof. D. Matthias Ullrich as well as my thesis committee members Dr. Katja Metfies, Prof. Dr. Frank Oliver Glöckner, and Prof. Dr. Jens Harder for continuous guidance, recommendations, and reflection of my work during the last three years.

I also thank all present and former members of the Ullrich Lab for sharing the last years in a very friendly and diverse working atmosphere. Particularly I would like to mention my ‘PhD student batch’ Khaled Abdallah and Amna Mehmood, furthermore Mariann Árkosi, Anja Kamp, and Daniel Pletzer as well as our lab technicians Nina Böttcher und Maike Last whose effort allows us lab members to fully focus on our work.

The ‘Fritten-Freitag-Gang‘ is thanked for sharing many fried lunches, conversations apart from the usual PhD business, and simply good times on campus.

I further express my gratitude to the ‘Graduate School for Polar and Marine Research’ (POLMAR) which supported me financially over the last three years. In particular I would like to emphasize the travel options I gained due to POLMAR, enabling outstanding trips to a variety of conferences and events. Here I would also like to mention Dr. Claudia Hanfland and Dr. Claudia Sprengel as coordinators of the graduate school as well as their associated colleagues.

Last but highly important I thank my family, friends, and boyfriend for guidance, support, and understanding over the last years.

III

IV Summary

Summary

Bacteria and microalgae conduct close mutualistic interactions that might impact metabolic and physiological features in both partners. As microalgae play major roles in global primary production and carbon cycling in marine systems, knowledge about such interactions is directly linked to an understanding of large-scale carbon and nutrient turnover processes. The impact of bacteria that colonize microalgae has been the focus of a number of studies. However, an understanding of the basis of such interactions at an advanced molecular level, including the identification of ‘traded items’ between the interacting partners, remained scarce. In the present study, we analyzed a specific beneficial bacteria-diatom interaction by application of proteomics approaches. A bilateral model system formerly established for the analysis of such interactions was used, consisting of the marine γ-proteobacterium Marinobacter adhaerens HP15 and the ubiquitously occurring centric diatom Thalassiosira weissflogii. The present study focused on the proteome of M. adhaerens HP15. Alterations in the bacterial proteome during co-cultivation with the diatom in comparison to a reference treatment were identified. Results imply that free amino acids are probable those traded items provided by the diatom, taken up by the bacterium, and potentially used as carbon and nitrogen source. Furthermore, proteomic results indicated a favorable supply with nutrients, particularly with nitrogen. For prospective studies, we suggest including the analysis of the diatom’s proteome and recommend the application of proteomics in combination with metabolomic approaches. For bacteria encountering and attaching to microalgae and other marine particles, the ability of chemotaxis, combining chemosensing and motility is beneficial. Existing mutants of M. adhaerens HP15 lacking chemotaxis features like motility response regulators or flagella were used. As part of the present thesis, we generated transformants that allow stereo-detection via differential fluorescence emission, in order to distinguish wild type and chemotaxis-impaired mutants in comparative studies. This advanced tool allows an easy analysis and quantification of effects of chemotaxis features on particle colonization. As a further topic of this thesis, features of heavy metal resistance in M. adhaerens HP15 and other members of the genus Marinobacter were analyzed. The genus Marinobacter has only recently been suggested as a bacterial taxon inhabiting metal-enriched environments and potentially impacting biogeochemical cycling of metals. M. adhaerens HP15 encodes two CzcCBA efflux pumps assumed to facilitate resistance towards cadmium, zinc, and cobalt. Single and double knock-out mutants of the representative genes were generated and V Summary

phenotypically characterized. Results showed that both pumps play a role in zinc tolerance but not in resistance towards cadmium and cobalt, respectively. In some but not all of the analyzed Marinobacter genomes czcCBA clusters were identified, substantiating the hypothesis of an association with environments containing increased levels of heavy metals. The actual role of the genus Marinobacter in biogeochemical cycling and its feature to inhabit heavy metal enriched environments remains to be determined in future studies.

VI List of Abbreviations

List of Abbreviations

5%-Suc 5% sucrose 5-ALA 5-aminolevulinic acid Ap ampicillin ASW artificial sea water bp base pairs BLAST Basic Local Alignment Search Tool BSA bovine serum albumin V CAZy carbohydrate-active enzymes [database] CDF cation diffusion facilitator CFB Cytophaga -Flavobacterium -Bacteroidetes Cm chloramphenicol Cm R chloramphenicol resistance [cassette] dd H2O tap water, deionized twice DIC diluted inorganic carbon DTT dithiothreitol EPS exopolymeric substances IAA indole-3-acetic acid IPG immobilized pH gradient [strip] kb kilo base pairs KEGG Kyoto Encyclopedia of Genes and Genomes LB lysogeny broth [medium] MALDI MS matrix assisted laser desorption/ ionization mass spectrometry MB marine broth [medium] MFP membrane fusion protein MIC minimal inhibitory concentration NCBI National Center of Biotechnology Information NSW North Sea water OMF outer membrane factor PCR polymerase chain reaction POC particulate organic carbon POM particulate organic matter PP primary production VII List of Abbreviations

ppm parts per million qRT-PCR quantitative reverse-transcription polymerase chain reaction rcf relative centrifugal force RND resistance-nodulation-cell division [type efflux pump] rpm revolutions per minute SDS sodium dodecyl sulphate S/N signal-to-noise PAGE polyacrylamide gel electrophoresis TCA tricarboxylic acid [cycle] TEP transparent exopolymer particles TOF time of flight v/v volume concentration w/v weight concentration

VIII Introduction

1. Introduction

1.1. Oceanic carbon cycling

About 50% of carbon fixation is conducted in surface waters of the World’s oceans (Field et al. , 1998). This incorporation of inorganic carbon into organic structures, also termed primary production (PP), is mainly carried out by photosynthetically active phytoplankton in the open ocean (Smith & Hollibaugh, 1993). The mainly responsible fraction of phytoplankton is represented by microscopically small cyanobacteria (Partensky et al. , 1999), diatoms (Nelson et al. , 1995), dinoflagellates (Verity et al. , 1993; Kopczynska et al. , 2007), or haptophytes (Figure 1). The rate of carbon fixation per area is on average about three times lower in marine surface waters, in comparison to certain terrestrial areas (Field et al. , 1998). Yet, the fact that oceans cover about 70% of our globe shows that marine associated carbon fixation is a major parameter in shaping global PP and thus carbon cycling.

Figure 1: Carbon and organic matter cycling in the marine water column with focus on the biological carbon pump (modified after Azam, 1998 and Ducklow et al ., 2001). Note that scales were moderately respected.

Photosynthetic PP carried out by phytoplankton is the major driving force of marine carbon fixation. However, it is worth mentioning that carbon fixation in marine environments may

1 Introduction

also be carried out by bacteria due to chemosynthesis. In case of chemosynthesis, energy is not obtained from light but from energy rich compounds like hydrogen, hydrogen sulfide, or methane, as found at deep sea vent or cold seep habitats (Dubilier et al. , 2008). A number of different carbon pools exist in the ocean. The fraction of dissolved inorganic carbon (DIC) is a product of dissolved atmospheric carbon dioxide (CO 2) that forms carbonic - + acid (H 2CO 3). H2CO 3 mainly dissociates to bicarbonate (HCO 3 ) and a proton (H ) in a pH dependent manner ( Figure 1) (Zeebe & Wolf-Gladrow, 2001). DIC accounts for the largest fraction of carbon in the ocean (Steele et al. , 2001). The phytoplanktonic fraction fixes DIC and incorporates it into organic carbon structures like proteins, carbohydrates, and DNA. Organic carbon does appear as either particulate organic carbon (POC) or dissolved organic carbon (DOC), the latter representing the largest pool of organic carbon compounds in the ocean (Hansell & Carlson, 2001). POC or, more general, particulate organic matter (POM) is represented by particles, that can be removed via filtration using 0.2 - 1 µm filter cut-offs (Figure 2) (Verdugo et al. , 2004).

Figure 2: Size range of carbon fractions and microorganisms as found in the marine water column (Azam & Malfatti, 2007).

Such particles might represent living organisms like prokaryotes, phytoplankton, or zooplankton, but also detritus, fecal pellets, and diverse types of marine aggregates (1.1.2). One of the important characteristics of POC is its ability to sink, therefore transporting organic matter to deeper ocean layers (1.1.1 & 1.1.2). In contrast to POC, DOC represents truly dissolved molecules ( Figure 2). DOC may be of different origin, being either actively

2 Introduction

released from organisms as exopolymeric substances (EPS) (1.1.3), released from dead organisms after cell lysis, or leak from more complex particulate aggregates, such as so-called marine snow (1.1.2). The molecular dynamics between DOC and POC are transient and characterized by colloidal particles and gel-like arrangements (Figure 2) (Verdugo et al. , 2004). DOC may accumulate thus forming larger coagulations resulting in those gel-like arrangements. These structures further aggregate with other particles, forming accumulations termed transparent exopolymer particles (TEP) (1.1.3). Newly generated organic carbon reaches higher food levels due to the uptake by higher organisms ( Figure 1). However, the majority of organic matter is re-allocated and remineralized by lower trophic levels, where microorganism like bacteria and small representatives of the zooplanktonic fraction play central roles. In addition, also viruses contribute to organic matter cycling (Cho & Azam, 1988; Suttle, 2007) as they are assumed to kill 10 - 20% of the abundant bacterioplankton per day (Suttle, 1994). Such numbers thus give an idea of massive re-allocation processes taking place in the water column caused by viral lysis. The cycling of carbon and nutrients on lower trophic levels due to microbial activity has been manifested in the concept of the ‘microbial loop’ (Azam et al. , 1983; Martin et al. , 1987). Understanding the carbon cycle is of high importance, as global carbon dynamics have been facing great shifts over the last approximately 150 years. The atmospheric carbon pool has strongly been impacted by massive anthropogenic combustion of fossil fuels, adding additional ‘available’ carbon to the global environment. This resulted in a massive increase of the atmospheric CO 2 concentration, which has been rising from 280 ppm to almost 400 ppm over the last decades (Friedlingstein et al. , 2010). Oceans actually have taken up ~30% of the additionally released carbon mainly by solution of CO 2 due to a higher atmospheric gas pressure and, representing a much smaller fraction, due to an increased burial of organic carbon on the sea floor (Raven & Falkowski, 1999). As a result of increased dissolved CO 2, the oceans’ pH has decreased by 0.1 to a pH of 8.1 during industrial times, a scenario that will likely continue (Caldeira & Wickett, 2003).

3 Introduction

1.1.1. Carbon fluxes and the biological carbon pump

The fixation of DIC and its incorporation into POC and DOC by phytoplankton takes place in the euphotic zone, marking the upper marine water column that is still penetrated by light. Photosynthetic carbon fixation in greater depth is limited due to the lack of light. POC, however, tents to sink, therefore generating a material flux towards deeper ocean layers that are thus supplied with carbon and nutrients (Fowler & Knauer, 1986). While sinking, the majority of biomass is remineralized by microorganisms or assimilated by higher organisms. This process mainly takes place in the upper water column. In dependence of the region roughly 1% of organically fixed carbon reaches the deep sea floor (Shanks & Trent, 1980; Suess, 1980; Lee et al. , 1998). Once being at the sea floor, this biomass usually remains widely inert and is thus buried for geological time scales (Stein, 1991). Consequently, a carbon net flux constantly removes biologically fixed carbon from the euphotic zone and exports it to the deep sea, a process hence termed ‘biological carbon pump’ (Ducklow et al. ,

2001). The flux of fixed DIC to the seabed allows new atmospheric CO 2 to dissolve in upper ocean waters. The principle of the biological carbon pump has gained attention in the research field of geo-engineering, which deals with the reduction of climate change effects like increasing CO 2 levels or ocean acidification by application of environmental large scale interventions (Shepherd, 2012). Phytoplankton growth and thus carbon fixation in marine systems is usually limited by the availability of nitrogen (Moore et al. , 2013). However, some particular oceanic regions termed ‘high nutrient low chlorophyll’ regions are rather limited by iron availability (Martin et al. , 1990; Coale et al. , 1996). Fertilization of these regions with iron would increase both PP and POC formation and possibly cause a greater export of material to the seabed. Thus the removal of atmospheric CO 2 can be enhanced, thereby probably lowering its present concentration (Buesseler et al. , 2004). Important in this context appear to be growth promoting effects on diatoms, which potentially lead to a faster and more pronounced sinking and thus carbon sequestration (Martin et al. , 2013).

1.1.2. Features of marine aggregates and marine snow

Marine aggregates larger than 500 µm in diameter are considered as marine snow (Alldredge & Silver, 1988) and can reach up to centimeters in size (Trent et al. , 1978; Shanks, 2002). These aggregates are coagulations of detritus, fecal pellets, and TEP (Figure 3), but also harbor a living community of bacteria and microalgae (Silver et al. , 1978; Alldredge & Silver, 1988; Simon et al. , 2002). Small invertebrates representing the zooplanktonic fraction may also inhabit an aggregate, as there are copepods, nematodes, or polychaete larvae found in or 4 Introduction

on aggregates (Kiørboe, 2000). The appearance of marine snow has been correlated to phytoplankton bloom events (Alldredge & Gotschalk, 1989).

Figure 3: Marine snow obtained from a sediment trap, Kerguelen Plateau, Southern Indian Ocean (Laurenceau- Cornec et al. , 2015).

Being enriched in carbon sources and other nutrients, marine snow offers an attracting food source in the marine water column: Bacterial cell and zooplankton numbers on such aggregates are well above those found in surrounding waters (Kiørboe, 2000; Azam & Malfatti, 2007). Thus, aggregates generate areas of high bacterial activity and turn-over processes in an elsewise oligotrophic surrounding (Smith et al. , 1992; Grossart et al. , 2007), thereby turning the marine water column into a patchy and heterogeneous pattern of microbial and nutritional hot spots (Shanks & Trent, 1979; Stocker, 2012) (1.2.3). Marine snow is a major player in the biological carbon pump, generating the flux of organic matter to deeper ocean layers. As one parameter, its sinking speed is determined by size and weight. Usually larger particles sink faster (Ploug et al. , 2008; Gärdes et al. , 2011), and the sinking speed observed is usually found in a range of ~10 to 150 m day -1 (Shanks & Trent, 1980; Shanks, 2002). However, single velocities of almost 300 m day -1 or more have been observed (Diercks & Asper, 1997; Laurenceau-Cornec et al. , 2015). Also the aggregate composition affects sinking speed. A high mineral ballast represented by silica-based diatom and calcium carbonate-based coccolithophore shells increases sinking speed of aggregates (Armstrong et al. , 2002; Klaas & Archer, 2002). Fast sinking supports the burial of carbon on the sea floor, as the time available for decomposition within the water column is reduced (Alldredge & Gotschalk, 1989).

5 Introduction

1.1.3. Transparent exopolymer particles trigger aggregate formation

The coagulation of marine snow is favored and triggered by the presence of transparent exopolymer particles (TEP) in the water column (Alldredge et al. , 1993; Engel, 2000; Passow, 2002a). These particles increase the size and mass of aggregates thus triggering their sinking and are therefore an important parameter in facilitating the downward flux of POC/ POM to deeper ocean layers. TEP mark a transient status between the dissolved and the particulate fraction of organic matter and form gel-like structures in the water column (Figure 2). They can be sampled by filtration and are traditionally visualized and quantified by alcian blue staining (Figure 4) (Passow & Alldredge, 1995). Their abundance can be up to 5,000 particles mL -1 and higher during (diatom) bloom events (Alldredge et al. , 1993; Mari & Burd, 1998). An important aspect of particle aggregation is the particle’s stickiness: TEP act as a kind of glue, collecting floating particles, sticking them together, and thus supporting their downward flux (Passow et al. , 2001; Engel et al. , 2004). Due to the high presence of sulfate half-ester groups, TEP are quite surface-active (Mopper et al. , 1995; Zhou et al. , 1998). Trace elements and other nutrients can be bound (Geesey et al. , 1988) and are therefore being channeled to the pool of the gel-like/ particulate organic fractions.

100 µm

Figure 4: TEP visualized by alcian blue staining (modified after Passow, 2002a).

TEP are formed from polysaccharides by both biotic or abiotic pathways and are characteristically made up by acidic carbohydrates. Fucose, rhamnose, galactose, glucose, xylose, and arabinose are classical carbohydrates that build up TEP and appear in varying ratios (Mopper et al. , 1995; Zhou et al. , 1998; Bruckner et al. , 2011). Additionally, TEP enter the marine water column by direct release from certain organisms, e.g., from corals (Huettel et al. , 2006). However, the majority of TEP is formed abiotically from dissolved precursors termed EPS as part of the DOC fraction (Chin et al. , 1998; Zhou et al. , 1998; Passow, 2000). EPS are released by both phyto- and bacterioplankton (Stoderegger & Herndl, 1999; Passow et al. , 2001), however, it is assumed that phytoplankton is the main source (Passow, 2002b;

6 Introduction

Pompei et al. , 2003). Phytoplanktonic EPS are mainly composed of carbohydrates and to a minor fraction of peptides or amino acids (Myklestad et al. , 1989). The actual composition and released quantities are highly variable, depending on species, growth state, nutrient supply, and environmental conditions (Obernosterer & Herndl, 1995; Stoderegger & Herndl, 1999; Passow, 2002b; Bhaskar et al. , 2005). TEP concentrations do further correlate with phytoplankton abundances and bloom events (Mari & Burd, 1998; Passow, 2002b). High concentrations were found during late exponential and stationary phase in in vitro studies (Alldredge et al. , 1993), similar to the observed high EPS concentrations found during the collapse of a natural bloom (Lancelot, 1983).

7 Introduction

1.2. Bacteria-diatom interactions

Diatoms are important primary producers in the marine system. Their natural life style includes the presence of bacteria in their nearest surrounding. The interaction of those two groups may be of different quality, varying intensity, and of particular advantage for one or both of them. In the following, mutualistic interactions, in which both partners benefit from the presence of the other, will be mainly focused on.

1.2.1. Diatoms are colonized by bacteria

Diatoms (Bacillariophyceae) are usually colonized by bacteria in their ‘phycosphere’, defined as the area surrounding the algal cell, in which growth of bacteria is impacted by algal exudates (Bell & Mitchell, 1972). Different diatom genera show a distinct bacterial colonization pattern (Schäfer et al. , 2002; Grossart et al. , 2005; Sapp et al. , 2007a) that may even vary on a species level (Kaczmarska et al. , 2005; Guannel et al. , 2011). Also physiological stages of diatoms may trigger varying colonizing bacterial taxa (Grossart et al. , 2005). Specific bacterial taxa can be identified, that are regularly found in association with diatoms (Amin et al. , 2012). These groups predominantly include α-proteobacteria such as members of the Roseobacter clade, γ-proteobacteria and the Cytophaga -Flavobacterium - Bacteroidetes (CFB) clade (Grossart et al. , 2005; Kaczmarska et al. , 2005; Sapp et al. , 2007a & 2007b; Huenken et al. , 2008), to a lesser extend also β-proteobacteria (Schäfer et al. , 2002; Bruckner et al. , 2008; Guannel et al. , 2011; von Scheibner et al. , 2014). Certain bacterial taxa are rather not associated with diatoms which refer to ε-proteobacteria, Actinobacteria, Firmicutes, and Tenericutes (Amin et al. , 2012). However, deviating observations were reported as well (Sapp et al. , 2007b; Baker & Kemp, 2014; von Scheibner et al. , 2014). The relationship between diatoms and bacteria may be of varying quality, including parasitic, commensalistic, or beneficial mutualistic interactions. Various examples describing parasitic relationships have been reported, in which one partner suffers a disadvantage due to present of the other partner (Fukami et al. , 1997; Lovejoy et al. , 1998; Paul & Pohnert, 2011). For example Kordia algicida , a member of the CFB clade, shows an algicidal effect on certain diatom species, likely induced by a secreted protease. Interestingly, the algicidal effect is species-specific as shown in a study where only three of four tested diatoms were sensitive to the particular bacterium (Paul & Pohnert, 2011). In case of commensalistic or beneficial mutualistic relationships, it was postulated, that the consortium of ‘natural bacterial partners’ is mandatory for full fitness of the diatom cell: Diatoms inoculated with a ‘foreign bacterial consortium’ showed to be less fit than with their own microbiota (Sison-Mangus et al. , 2014). 8 Introduction

1.2.2. Bacteria-diatom interactions and the global carbon cycling

Diatoms contribute to about 40% of marine PP, which accounts for 20% of global PP, and are therefore fundamental organisms for shaping global carbon fluxes (Falkowski et al. , 1998). Due to the diatoms’ heavy silicate shells, marine snow tends to sink faster, when being enriched with diatom shells (Takahashi et al. , 2000; McDonnell & Buesseler, 2010). Diatoms are therefore important for downward fluxes and burial of POC. Seasonal phytoplankton dynamics like spring and autumn blooms are accompanied by increases in certain bacterial taxa that rise during the succession of blooms (Sapp et al. , 2007c; Teeling et al. , 2012). The bacterial colonization of diatoms (1.2.1) may impact physiological processes taking place in the algae (Grossart & Simon, 2007). It is evident that in certain cases diatoms depend on their bacterial partners, as they provide molecules or functions, diatoms cannot synthesize or conduct on their own (1.2.4). Certain studies showed that the presence of bacteria is an important parameter in diatom-based aggregate formation, thus enhancing sinking and sequestration of diatom derived biomass (Grossart et al. , 2006; Gärdes et al. , 2011) (1.1.2). The presence of bacteria may induce the release of EPS and the formation of TEP, respectively, in diatoms (Grossart & Simon, 2007; Bruckner et al. , 2008; Gärdes et al. , 2011), thereby releasing valuable carbon sources for bacteria. However, opposing observations were also documented (Bruckner et al. , 2011). TEP in turn have a major impact on marine snow formation (1.1.3), strongly shaping carbon dynamics in the water column and the biological carbon pump (1.1.1). A better knowledge about the interactions of bacteria with diatoms is consequently directly linked to an improved understanding of global carbon cycling and will probably allow the estimation of future changes caused by anthropogenic impact.

1.2.3. Bacterial chemotaxis promotes the encountering of nutritional hot spots

Although being almost invisible for the human’s eye, the marine water column is not a homogeneous, evenly mixed environment but a patchy pattern of nutritional hot spots, represented by marine snow, leaky plumes of DOC, concentration gradients, and niches (Shanks & Trent, 1979). This patchy distribution is not only reflected in the nutrient pattern, but also favors a heterogeneous distribution of bacterioplankton (Alldredge et al. , 1993; Long & Azam, 2001). The combined abilities of sensing a chemoattractant (chemosensing) and the directed movement towards it (motility) result in bacterial chemotaxis. In environments such as marine waters, nutritional hot spots like, e.g., a lysed microalga appear and disappear

9 Introduction

within short time scales of minutes. Microorganisms consequently benefit from chemotaxis allowing fast encountering of such hot spots (Blackburn et al. , 1998; Kiørboe, 2001; Kiørboe & Jackson, 2001). Motile marine bacteria were also found to track and follow moving particles, as there are sinking aggregates, particles agitated by ocean turbulences, or motile organisms (Barbara & Mitchell, 2003). As a probable evolutionary adaption to such environmental conditions, swimming speed and acceleration of marine bacteria were observed to be faster than in other, for example enteric bacteria (Mitchell et al. , 1995b; Johansen et al. , 2002; Stocker et al. , 2008). Marine bacteria further show different motility patterns in comparison to the model organism Escherichia coli : E. coli and other peritrichous bacteria follow a run-and-tumble movement, when moving chemotactically, resulting in turning angles of about ~70° to 90° in dependence of the test medium (Figure 5) (Berg, 2003).

Figure 5: Scheme representing different bacterial motility patterns. a Run-and-tumble movement of peritrichous E. coli , involving turn angles of about 70° to 90° b Run-reverse movement of monotrichous marine bacteria, including turn angles of up to 180°, caused by clockwise and counter-clockwise flagellum movement (green arrow) c Run-reverse-flick movement, combining run-reverse movement with a following flick of the flagellum, causing frequent turn angles of about 90°. Red arrows mark the starting point of a run (modified after Taktikos et al. , 2013).

Due to the presence of usually one single flagellum (monotrichous), a run-reverse movement was observed among ~70% of marine bacteria (Johansen et al. , 2002) generating turning angles of up to 180°. Such angles are obtained as the flagellum can only rotate clockwise or counter-clockwise, resulting in either forward or backward movement. Due to these strategies, it was shown that marine representatives performed much better in the encountering of a nutrient patch than E. coli (Stocker et al. , 2008; Xie et al. , 2011). Recently, the principle of run-reverse migration was extended by an additional flick of the flagellum, allowing a reorientation of the bacterium that includes a smaller angle of ~90° (Figure 5). Thus, exploration of the environment becomes even more efficient (Stocker, 2011; Xie et al. , 2011). 10 Introduction

Just as chemotaxis itself, fast encountering of chemoattractants adapts to a life style in an environment, in which nutrients are scarce and hot spots collapse within minutes. Bacterial motility observed in natural assemblages varies over the time course of a year, being higher in late summer months and lower during winter at a selected sampling site (Grossart et al. , 2001). Although tempting to assume that motility was induced by temperature, this suggestion could not be proven in the respective study. Motility, however, seems to be a function of available carbon and nutrient sources, probably also supporting the assumption that motility may depend on certain energy availabilities (Mitchell et al. , 1995a; Stretton et al. , 1997; Grossart et al. , 2001). The presence of a flagellum facilitates initial contact with a surface and determines the formation of biofilms (Tomihama et al. , 2006; Malamud et al. , 2011). The presence of different types of pili further supports the development of biofilms, as it was shown for E. coli (Pratt & Kolter, 1998). Besides these structural elements, the inner cellular chemosensing and response mechanisms were shown to be important (Kalmokoff et al. , 2006; Sonnenschein et al. , 2012). Lacking the ability of chemotaxis limits the attachment of bacteria towards particles, including microalgae like diatoms and dinoflagellates (Miller & Belas, 2006; Sonnenschein et al. , 2012). Chemotactic bacteria encounter microalgae and marine aggregates more efficiently, causing a more efficient remineralization (Kiørboe et al. , 2002). Therefore, chemotaxis can be considered as a crucial parameter in the dynamics of carbon and nutrient cycling (Grossart et al. , 2001).

1.2.4. Bacteria-diatom interaction on a molecular level

Mutualistic relationships between diatoms and their bacterial hosts are documented, yet the biochemical basis of this mutualism is not well understood. To identify ‘trade-off compounds’ between bacteria and diatoms the application of advanced methods on a molecular level has just started. A number of studies applied transcriptomic, proteomic, or metabolomic approaches to identify driving forces of bacteria-diatom interactions (Bruckner et al. , 2011; Paul et al. , 2013; Amin et al. , 2015). The presence of bacteria might promote the growth of a diatom population during co- cultivation in comparison to axenic diatom cultures (Grossart, 1999; Bruckner et al. , 2008; Amin et al. , 2009; Amin et al. , 2015). These effects, however, can shift from mutualistic to commensalistic or even parasitic in dependence on the environmental nutrient status and may even lead to a stop of the diatom growth (Grossart & Simon, 2007). Vice versa , also diatoms 11 Introduction

may promote bacterial growth as bacteria may feed on diatom exudates as sole carbon source (Figure 6) (Bruckner et al. , 2008; Amin et al. , 2015).

Figure 6: Schematic presentation of compounds traded between bacteria and diatoms during beneficial mutualistic interaction (Note that the allocation of bio-available iron was only shown for a dinoflagellate).

Growth-promoting effects induced by bacteria can be highly specific, are dependent on the co-cultivated partners (Bruckner et al. , 2011), and may even be selective at the strain level: In a co-cultivation experiment conducted with the α-proteobacterium Sulfitobacter sp. SA11, the bacterium increased the growth rate of Pseudo-nitzschia multiseries strain PC9 by 35%, while strain PC4 remained almost unaffected by the presence of S. sp. SA11 (Amin et al. , 2015). Growth promoting effects could also be obtained when Phaeodactylum tricornutum , a benthic brackish water diatom, was merely supplemented with cell-free supernatants of bacterial cultures (Bruckner et al. , 2011). When using 10% supernatant or more, the growth effect was comparable to that obtained by direct bacterial co-cultivation, suggesting the secretion of the growth-promoting factor(s) into the growth medium. In contrast to more recent findings obtained from advanced methodical applications, cobalamin (vitamin B 12 ) had already been identified as a potentially important compound provided by bacteria to diatoms about 40 years ago (Haines & Guillard, 1974). Vitamin B 12 auxotrophy is common in a wide range of diatoms and other microalgae (Croft et al. , 2005). The same study showed that bacteria of the genus Halomonas sp. provide this vitamin to the red alga Porphyridium purpureum and to the dinoflagellate Amphidinium operculatum . Since microalgae including diatoms also frequently show auxotrophy towards other vitamins such as thiamin (B 1) or biotin (B 7), additional mutualistic relationships based on those compounds may be suggested (Tang et al. , 2010). Indeed, it was proven that the Roseobacter clade

12 Introduction

member Dinoroseobacter shibae provides vitamin B 1 to its dinoflagellate host Prorocentrum minimum (Wagner-Doebler et al. , 2010). Paul and colleagues conducted a metabolomics approach for the interaction between D. shibae and Thalassiosira pseudonana (Paul et al. , 2013), with particular emphasis on the cellular metabolites of the diatom. Several amino acids and amino acid derivatives as well as sugars and certain (fatty) acids were found in higher concentrations during co-cultivation. It could not be resolved if those metabolites were of bacterial source and had been taken up by the diatom, or if their synthesis had been stimulated within the diatom by supply of beneficial compounds, for instance vitamins, obtained from the bacterium. Alterations in the synthesis of free amino acids released by the diatoms were documented in another study conducted by Bruckner and colleagues: A proteomics approach including metabolomic data acquisition was applied for the rather unconventional co-cultivation of E. coli and P. tricornutum (Bruckner et al. , 2011). Results showed transporters and protein binding proteins being expressed in the bacterium in the presence of the diatom. The authors further found indications for bacterial biofilm formation based on differentially expressed bacterial proteins. Besides alterations in EPS release by the diatom, changes in both quantity and quality of dissolved free amino acids were observed, likely offering rich carbon and nitrogen sources for associated bacteria. Bioavailable iron, Fe(II), is a mandatory element in, e.g., photosynthesis, however, due to its low solubility it is often a limiting factor in particular oceanic regions (Martin et al. , 1990). Different prokaryotes have developed strategies of gathering bioavailable iron by the synthesis, release, and recapture of chelating agents, so called siderophores. Their siderophores bind bio-unavailable Fe(III) with high affinity and reduce it to bioavailable Fe(II) (Barbeau et al. , 2001). This compound might further be oxidized to the inorganic ferric hydrolysis ion Fe(III)’, which is also bioavailable for an alga cell (Amin et al. , 2009). Interestingly, marine siderophores are sensitive to light; after the light induced oxidation of Fe(III) to Fe(II), the complex disintegrates, followed by iron release. Members of the genus Marinobacter were found to provide bioavailable iron to the dinoflagellate Scrippsiella trochoidea in in vitro studies, releasing the siderophore vibrioferrin (Amin et al. , 2009). In return, the bacteria obtain carbon sources provided by the dinoflagellate allowing the speculation that they might act as mutualistic ‘trade-off products’ provided by the dinoflagellate, to support iron gathering by the bacterium in its close proximity. Earlier field and in vitro studies also suggested bacterial siderophores as crucial for iron acquisition as shown for the diatom P. tricornutum (Soria-Dengg et al. , 2001).

13 Introduction

Only recently, the phytohormone indole-3-acetic acid (IAA) has been identified as a growth promoting factor for P. multiseries , provided by Sulfitobacter sp. SA11 in a co-cultivation experiment (Amin et al. , 2015). This study further showed, that the diatom provides carbon in form of possibly taurine and tryptophan to the bacterium. Interestingly, the authors assumed a feedback loop, in which diatom-provided tryptophan is taken up by the bacterium, converted to IAA, and re-allocated to the algae. Besides this phytohormone, S. sp. SA11 additionally provided ammonium to the diatom, which preferred this reduced nitrogen compound over nitrate that had been offered in the diatom’s growth medium. Certain cyanobacteria were found to highly increase their nitrogen fixation rate when being co-cultivated with diatoms (Foster et al. , 2011). Those authors found a substantial transfer of fixed nitrogen from the cyanobacteria to the diatom host suggesting that mutualistic cyanobacteria may act as important nitrogen providers to algae in nitrogen-limited regions of the ocean.

1.2.5. Marinobacter adhaerens HP15 – Thalassiosira weissflogii : A model system to understand bacteria-diatom interactions

In recent years, a model system consisting of the γ-proteobacterium Marinobacter adhaerens HP15 and the central diatom Thalassiosira weissflogii was established (Figure 7) (Gärdes et al. , 2011; Sonnenschein et al. , 2011). This system aims to analyze bacteria-diatom interaction at the molecular level under controlled laboratory conditions. M. adhaerens HP15 was isolated from marine aggregates, sampled in the German Wadden Sea (Grossart et al. , 2004). In an attempt to identify diatom-attaching bacteria, M. adhaerens HP15 was recognized as such in co-cultivation with T. weissflogii (Figure 8) (Gärdes et al. , 2011). In addition, an increase in TEP formation was observed exclusively during co-cultivations, suggesting a bacterial induction of TEP release in the diatom (Gärdes et al. , 2011). TEP release may alter the stickiness of marine particles, thus inducing marine snow formation and enhancing particle sinking (1.1.3). Based on the aggregate-forming and TEP release-inducing phenotype of M. adhaerens HP15 in co-cultivation with the diatom, a bilateral model system including both partners has been established, that allows controlled in vitro studies. Their interaction is assumed to be of beneficial mutualism, however, in dependence on nutrient availability a shift from mutualism towards a commensalistic interaction has been proposed to be transient (Gärdes et al. , 2012). The genome of M. adhaerens HP15 is sequenced and fully annotated, the organism is furthermore genetically accessible (Gärdes et al. , 2010; Sonnenschein et al. , 2011). Generation and analysis of gene-specific knock-out mutants have served in

14 Introduction

understanding the importance and impact of bacterial chemotaxis during colonization of T. weissflogii (Sonnenschein et al. , 2012). Further in-depth investigations were conducted that analyzed the characteristics of TEP release during co-cultivation under varying nutrient regimes (Gärdes et al. , 2012) but also physical and chemical parameters like temperature and pH (Seebah et al. , 2014). Results showed that under varying nutrient regimes, both quantity and quality of various organic molecules such as sugars and amino acids but also TEP formation varied (Gärdes et al. , 2012).

a b

5 µm

Figure 7: Partners of the bilateral model system for identification of bacteria-diatom interactions. a Transmission electron micrograph of the γ-proteobacterium M. adhaerens HP15 (Gärdes et al. , 2010) b Diatom T. weissflogii (modified after F. Hinz, www.cmore.soest.hawaii.edu).

The genus Marinobacter comprises about 50 described species and is ubiquitously found in varying marine and saline habitats, including marine waters, marine organisms, sediments, or saline wetlands (Romanenko et al. , 2005; Aguilera et al. , 2009; Handley et al. , 2009; Ng et al. , 2014). Isolates from a set of oil-polluted sampling sites imply an ability of the respective Marinobacter strain to degrade hydrocarbons (Gu et al. , 2007; Overholt et al. , 2013), which has thoroughly been described for M. hydrocarbonoclasticus , the first isolated representative of the genus Marinobacter (Gauthier et al. , 1992). Members of the genus Marinobacter were also isolated from cultures of the diatom species Pseudo-nitzschia pungens and P. australis (Sison-Mangus et al. , 2014) as well as from P. multiseries cultures (Amin et al. , 2015). Interestingly, additional Marinobacter strains were furthermore identified in natural bacterial communities of other microalgae including dinoflagellates (Alavi et al. , 2001; Hold et al. , 2001; Green et al. , 2004; Green et al. , 2006) or coccolithophores (Amin et al. , 2009). Genus representatives therefore appear to be commonly associated with microalgae. This underlines 15 Introduction

their functionality and suitability for acting as bacterial representative in an interaction model system. T. weissflogii on the other hand is ubiquitously found in marine systems. Since diatoms are of high climatic impact due to their wide abundance, quantity, and photosynthetic activity, T. weissflogii has offered to be a suitable candidate for the set-up of a model system that aims to understand the impact of bacteria-diatom interactions in global dimensions. The diatom itself has been used in many previous studies on diverse topics and is a well-established laboratory organism. Full genome information of a close relative of T. weissflogii , T. pseudonana , is available (Armbrust et al. , 2004).

Figure 8: Scanning electron micrograph of M. adhaerens HP15, attaching to the diatom T. weissflogii (Gärdes et al. , 2011).

16 Introduction

1.3. Heavy metal resistance in bacteria

Microorganisms may face and cope with elevated heavy metal concentrations in their natural habitats. Examples for such habitats are submarine hydrothermal (deep sea) vents (Varnavas & Cronan, 2005; Minic et al. , 2006), serpentine soils (Fernandez et al. , 1999), or metal- accumulating plants (Comino et al. , 2005). Anthropogenic impact yet has created heavily contaminated sites that result from mining activity or other industrial impact (Schulze et al. , 1997; Rashed, 2010). Such sites select for and allow easy identification and isolation of bacteria that have well developed mechanisms to manage and survive even high levels of heavy metals. Following this strategy, the β-proteobacterium Cupriavidus metallidurans CH34 was isolated from a zinc decanting tank about 40 years ago, and is likely the most comprehensively studied organisms with reference to heavy metal resistance (Mergeay et al. , 1985; Grass et al. , 2000; Legatzki et al. , 2003; Monchy et al. , 2006; Mikolay & Nies, 2009; Scherer & Nies, 2009).

1.3.1. Bacterial strategies to cope with environmental heavy metal stress

Bacteria do exhibit a set of strategies to detoxify the cell from elevated metal concentrations. Mercury ions for example (Hg 2+ ) are reduced to nontoxic and volatile elemental mercury (Hg 0) that leaves the bacterial cell as vapor (Schottel et al. , 1974; Jobling et al. , 1988). The main mechanism of resistance is, however, active efflux facilitated by membrane-bound efflux pumps. Different types of efflux pumps can be present in a bacterium, all together contributing to a comprehensive broad-range metal resistance. These pumps may complement each other, as shown in a variety of gene knock-out studies, in which the loss of one pump was compensated by the function of a different one (Scherer & Nies, 2009). The major types of pumps found to facilitate resistance are 1) resistance-nodulation-cell division (RND) type efflux pumps belonging to the heavy-metal-resistance-RND family, 2) cation diffusion facilitators (CDFs), and 3) P-type ATPases. Other types of pumps are of rather minor importance. Briefly, they comprise members of the CHR protein family facilitating chromium efflux (Ramirez-Diaz et al. , 2008) and NreB-like proteins involved in nickel resistance as well as resistance to further non-metal compounds (Grass et al. , 2001b; Pini et al. , 2014). CnrT- like proteins are probably also involved in minor nickel resistance (Nies, 2003). Both CDFs and P-type ATPases are membrane-bound protein pumps that are found in both Gram-positive and Gram-negative bacteria (Nies, 2003). The model organism C. metallidurans CH34 harbors at least three different CDFs, in detail CzcD, FieF, and DmeF (Anton et al. , 1999; Munkelt et al. , 2004). Being quite variable within the kingdom of life, 17 Introduction

CDFs in bacteria commonly possess six transmembrane domains (Montanini et al. , 2007). They function as proton antiporter (Chao & Fu, 2004; Grass et al. , 2005) and possibly also as proton/ potassium antiporter (Guffanti et al. , 2002). CDFs mainly facilitate divalent metal ion efflux. In dependence on the individual CDF, substrates range from zinc to cadmium, cobalt, copper, iron, manganese, or nickel (Anton et al. , 1999; Grass et al. , 2001a; Munkelt et al. , 2004; Grass et al. , 2005; Moore et al. , 2005; Cubillas et al. , 2014). A few cases are known, where these pumps conduct metal influx (Jiang et al. , 2012; Albareda et al. , 2015). Yet certain protein pumps with close homology to well-defined CDFs were demonstrated to possess additional substrate spectra and functions besides the role as iron transporters (Fang et al. , 2002; Uebe et al. , 2011). P-type ATPases function due to the hydrolysis of ATP (Sharma et al. , 2000; Nies, 2003). Some of these ATPases may either pump monovalent or divalent metal ions. Only in exceptional cases both ions are transported (Tong et al. , 2002). In dependence on the type and source of organism, substrates are either monovalent copper and silver ions (Kanamaru et al. , 1994; Solioz & Odermatt, 1995) or divalent cadmium, zinc, lead, and cobalt ions (Binet & Poole, 2000; Lee et al., 2001; Gaballa & Helmann, 2003; Legatzki et al. , 2003; Leedjarv et al. , 2008). Cadmium appears to be the central substrate. In C. metallidurans , a set of such transport enzymes have been described as CzcP, CadA, ZntA, and PrbA (Scherer & Nies, 2009). The efflux pump system focused in the present work, in detail the cadmium-zinc-cobalt (Czc) system, belongs to the group of RND type efflux pumps. Such pumps are wide-spread among Gram-negative bacteria. Two of the best studied efflux pumps are the AcrAB-TolC system of E. coli (Ma et al. , 1995; Fralick, 1996) and the MexAB-OprM system of Pseudomonas aeruginosa . Both systems are responsible for multidrug efflux (Li et al. , 1995). RND efflux proteins are often accompanied by two additional membrane-bound proteins. These are an inner membrane adapter protein belonging to the family of membrane fusion proteins (MFPs) (Dinh et al. , 1994) and an outer channel protein, a member of the outer membrane factor (OMF) family (Paulsen et al. , 1997). This combination results in a tripartite efflux pump (Figure 9) (Symmons et al. , 2009). The three proteins are usually encoded in an operon. Due to their specific integration into the cell wall structure, RND pumps are almost exclusively found in Gram-negative bacteria (Figure 9) (Nies, 2003).

18 Introduction

CzcC

Czc A Czc B Czc B

Figure 9: Scheme of the AcrAB-TolC efflux pump of E. coli , an RND type efflux system (modified after Alvarez-Ortega et al. , 2013). The three components of the CzcCBA efflux system are marked with orange letters.

A number of RND efflux systems facilitate the efflux of different heavy metals. The Czc system is likely the most popular and best studied one, comprising the central pump CzcA, accompanied by CzcB and CzcC as the MFP and the OMF, respectively (comprehensively termed CzcCBA) (Figure 9). As implied by its name, this pump has been shown to facilitate the efflux of zinc and additionally cobalt and cadmium (Nies et al. , 1989; Nies & Silver, 1989). Other studies have unraveled different substrate spectra for pump systems, previously annotated as CzcCBA systems, showing resistance towards merely zinc and cadmium or additionally to nickel (Stähler et al. , 2006; Leedjarv et al. , 2008). A number of other metal efflux-associated RND pumps are known, as there is the cobalt/ nickel resistance determinant (CnrCBA) (Liesegang et al. , 1993; Siunova et al. , 2009) or the NccCBA system (Schmidt & Schlegel, 1994), facilitating resistance to nickel, cadmium, and cobalt.

1.3.2. Marinobacter – a genus inhabiting environments, enriched in heavy metals?

Although a number of Marinobacter strains have been associated with marine phytoplankton and hydrogen polluted environments (1.2.5), a recent review article stated another rather new niche. Members of this genus may also be associated with habitats, naturally enriched in heavy metals and may have impacts on geochemical metal cycling (Handley & Lloyd, 2013). Thus far, only a small number of Marinobacter species has been isolated from such habitats. They include M. manganoxydans isolated from a manganese nodule in the deep sea (Wang et al. , 2012) and M. santoriniensis obtained from hydrothermal sediments in the Aegean Sea

19 Introduction

(Handley et al. , 2009). Not being isolated from such environments, M. adhaerens HP15 has become popular mainly due to its association with phytoplankton (Gärdes et al. , 2011). Recently, it was observed that its genome harbors two copies of gene sets encoding for the RND-type efflux system CzcCBA (unpublished data) promoting the assumption of a potential resistance to cadmium, zinc, and cobalt (1.3.1). Due to this finding as well as due to further advantages of this strain like its genetic accessibility (1.2.5) M. adhaerens HP15 may act as a probably useful candidate to initially investigate characteristics of heavy metal resistance in the Marinobacter genus.

20 Aims of the Present Study

2. Aims of the Present Study

The major aim of the present study is to develop an in-depth understanding of bacteria-diatom interaction, using the two model organisms, the bacterium M. adhaerens HP15 and the diatom T. weissflogii . Proteomics approaches will be the applied method of choice: Cellular proteins of M. adhaerens HP15 will be sampled during co-cultivation experiments with the diatom and compared to non-co-cultivated reference treatments. Proteins will be visualized by two- dimensional sodium dodecyl sulfate-polyacrylamide gel electrophoresis (2D-SDS-PAGE), differences in the proteomic pattern further identified via application of matrix assisted laser desorption/ ionization-time of flight mass spectrometry (MALDI-TOF MS).

The second aim is the generation of tools that allow the simultaneous detection of different M. adhaerens HP15 mutants within one experimental set-up. These mutants are deficient in features of chemotaxis as there is chemosensing or motility and were already designed in the past. As the method of choice, a variety of plasmids encoding different fluorescent proteins will be constructed and tested for their feasibility under certain experimental conditions.

The third aim of this study is to analyze features of heavy metal resistance in M. adhaerens HP15. For this purpose, knock-out mutants lacking two czcCBA operons encoding RND-type heavy metal efflux pumps will be generated and phenotypically analyzed. In addition, the genomes of members of the genus Marinobacter will be analyzed in detail for the presence of CzcCBA efflux pumps in order to define their potential associations with heavy metal- enriched environments.

22 Results

3. Results

Results of the present doctoral thesis are presented in the following three manuscripts:

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii (Results/ Part I; 3.1) Antje Stahl & Matthias S. Ullrich Antje Stahl conducted the experimental design and entire practical work. Matthias Ullrich supervised the work, discussed aspects of the writing process, and revised the manuscript. This manuscript will be submitted to the peer-reviewed journal ‘Applied and Environmental Microbiology’ in the near future.

Stereo-tracking of chemosensing-deficient and motility impaired Marinobacter adhaerens HP15 mutants during marine particle colonization – a novel methodical approach (Results/ Part II; 3.2) Antje Stahl & Matthias S. Ullrich Antje Stahl conducted the experimental design and entire practical work. Matthias Ullrich supervised the work, discussed aspects of the writing process, and revised the manuscript. This manuscript will remain unpublished at the present stage. So-far achievements will be applied by following PhD students being involved in this project.

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification (Results/ Part III; 3.3) Antje Stahl, Daniel Pletzer, Amna Mehmood, & Matthias S. Ullrich Daniel Pletzer contributed to this study by theoretical mutant design, design of plasmids used for complementation of mutations, and fruitful discussions. Amna Mehmood contributed to this study by conducting RNA extractions and qRT-PCR. Matthias Ullrich supervised the work, discussed aspects of the writing process, and revised the submitted manuscript. Parts of this study were published in the peer-reviewed journal ‘Antonie van Leeuwenhoek’ (108(3): 649-658).

23 Results

During the period of the doctorate, the author contributed to three further publications and manuscripts, which are not discussed in the present thesis.

The conserved upstream region of lscB /C determines expression of different levansucrase genes in plant pathogen Pseudomonas syringae Shaunak Khandekar, Abhishek Srivastava, Daniel Pletzer, Antje Stahl, & Matthias S. Ullrich BMC Microbiol (2014) 14:79 Antje Stahl contributed to this study by conducting respective protein identification via MALDI-TOF MS.

Role of the cell envelope stress regulators BaeR and CpxR in control of RND-type multidrug efflux pumps and transcriptional cross talk with exopolysaccharide synthesis in Erwinia amylovora Daniel Pletzer, Antje Stahl, Anna E. Oja, & Helge Weingart Arch Microbiol (2015) 197:761–772 Antje Stahl contributed to this study by conducting respective protein identification via MALDI-TOF MS.

Development and application of a fluorescence based supramolecular assay as screening tool for bacterial steroid degradation Antje Stahl, Veronica N. Muchemu, Alexandra I. Lazar, Werner M. Nau, Matthias S. Ullrich, & Andreas Hennig Antje Stahl contributed to this study by discussing practical milestones of the project with special emphasis on environmental and microbiological questions. Furthermore the entire cultivation experiments were prepared and conducted. This publication is in preparation and under discussion.

24 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

3.1. Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Abstract

Diatoms are often faced by bacteria in their closest proximity termed the phycosphere. For some of these potential interactions, mutually beneficial relationships between both partners have been assumed. Substrates determining mutualistic interactions have been identified in this context, yet in-depth studies on bacteria-diatom interactions remain scarce. In the present study, we applied a proteomics approach to obtain a deeper perspective into such interactions. To this end, a bilateral model system was used that had been previously established for investigating bacteria-diatom interactions and marine particle formation, comprising the marine γ-proteobacterium Marinobacter adhaerens HP15 and the diatom Thalassiosira weissflogii . In co-cultivation experiments, the proteome of M. adhaerens HP15 derived from cell lysates was sampled and compared with protein samples from diatom-free bacterial cultures. Proteome alterations were visualized by two-dimensional gel electrophoresis, and differentially expressed proteins were analyzed by mass spectrometry. Our results suggested a differentiated nutrient supply for M. adhaerens HP15 during co-cultivation. The bacterium seemed to benefit from the release of amino acids by the diatom as indicated by an up- regulation of several transporter elements responsible for amino acid uptake. To substantiate these results, a metabolic screen for suitable nutrient substrates of M. adhaerens was conducted, revealing that amino acids appeared to be the preferred carbon source of the bacterium, while various tested sugars were not utilized. Future experiments will focus on an in-depth proteome analysis of the diatom in order to understand the role of potential amino acid uptake systems for the interaction of M. adhaerens HP15 with the diatom.

Keywords bacteria-diatom interaction · mutualism · Marinobacter adhaerens HP15· Thalassiosira weissflogii · carbon cycle

25 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Introduction

Despite their small size and inconspicuous appearance, diatoms contribute about one fifth to the global primary production, mainly proceeding in the World’s oceans (Field et al. , 1998). They do not only fix atmospheric carbon dioxide but also provide carbon and other nutrients to their surrounding due to the release of exudates that may coagulate and form transparent exopolymer particles (TEP) (Passow, 2000; Passow et al. , 2001). In dependence of certain environmental parameters such as nutrient availability, up to 50% of carbon fixed during primary production may be re-allocated in form of exudates (Beauvais et al. , 2006). Formed TEP serve as a valuable nutrient and carbon sources for other microbes thus fueling the degradation and re-allocation of compounds in the upper water column on a micro-scale level (Azam & Malfatti, 2007). Indeed, algal bloom dynamics are closely coupled to bacterial dynamics: In dependence on the succession of a bloom, different bacterial taxa specialized on degradation of specific substrates may peak in abundance (Fandino et al. , 2005; Rooney- Varga et al. , 2005; Mayali et al. , 2011; Teeling et al. , 2012). Therefore, primary production of diatoms and subsequent interactions between both bacteria and diatoms were shown to be important in carbon and nutrient cycling on a global scale (Azam et al. , 1994). Diatoms and other single-celled algae such as dinoflagellates are usually colonized by bacteria in their phycosphere (Baker & Kemp, 2014; Sison-Mangus et al. , 2014). This area marks the direct surrounding of a diatom, in which bacterial growth is impacted due to the release of algae compounds (Bell & Mitchell, 1972). In the phycosphere, bacteria may be attached to the cell surface or found swarming in nearest proximity (Blackburn et al. , 1998). Different diatom genera or even species harbor distinct groups of bacteria (Grossart et al. , 2005; Guannel et al. , 2011; Eigemann et al. , 2013; Sison-Mangus et al. , 2014), prompting to assume that bacterial colonization is diatom species-specific. Individual algal exudates may selectively attract particular bacterial species (Seymour et al. , 2009). An intimate co-evolution of diatoms and colonizing bacteria was suggested based on the finding that several hundred genes in the diatom Phaeodactylum tricornutum are of bacterial origin (Bowler et al. , 2008). Although parasitic or commensalistic relationships have been observed (Bratbak & Thingstad, 1985; Paul & Pohnert, 2011), synergistic relationships are common, in which both partners profit from the presence of the other (Grossart, 1999). A number of benefits are known for bacteria and diatoms or other single-celled algae that are gained by both partners during the interaction (Amin et al. , 2012). Respective studies focused on, e.g., TEP production 26 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

dynamics, release of hydrolytic enzymes, or growth behavior (Grossart, 1999; Gärdes et al. , 2012). However, in-depth studies that use advanced metabolomics, molecular genetics, or proteomics approaches remain scarce. A major benefit for bacteria is the availability of metabolites such as TEP (Grossart, 1999), which are mainly composed of carbohydrate polymers and to a lesser content of nutrient sources such as proteins, peptides, amino acids, or lipids (Myklestad et al. , 1989; Myklestad, 1995; Biddanda & Benner, 1997). Bacteria do not only metabolize but also may alter TEP formation and release resulting in increase or decrease of TEP production (Bruckner et al. , 2011; Gärdes et al. , 2011). As TEP may serve as stickiness-increasing ‘glue’ and therefore trigger particle formation and sinking, bacteria- diatom interactions also impact the biological carbon pump (Alldredge & Gotschalk, 1989; Ducklow et al. , 2001). Diatoms are assumed to benefit from synergistic bacterial interactions due to the allocation of vitamins by bacteria, with cobalamin (vitamin B 12 ) being the most often discussed candidate in this context (Droop, 1970; Cole, 1982; Croft et al. , 2005). A wide range of algae was shown to be auxotrophic for cobalamin (Croft et al. , 2005), whereas auxotrophy towards other vitamins such as thiamine (vitamin B 1) or biotin (vitamin B 7) is less widespread but was also shown to be a potential trigger for synergistic relationships (Croft et al. , 2006; Wagner- Doebler et al. , 2010). Another valuable trade between bacteria and diatoms is iron. Iron is a limiting growth factor for phytoplankton in certain parts of the ocean (Behrenfeld et al. , 1996) and could be provided by algae-associated bacteria via their ability to form siderophores (Amin et al. , 2009). Nitrogen is a further ‘traded item’ provided by bacteria, as shown for cyanobacteria (Foster et al. , 2011) or for a Sulfitobacter strain (Amin et al. , 2015). The latter bacterium did not only provide ammonia, but also allocated the plant hormone indole-3-acetic acid (IAA), thus generating a growth promoting effect in its co-cultivated diatom partner P. multiseries (Amin et al. , 2015). Bruckner et al. (2011) applied a proteomics approach to an in vitro co-cultivation of the diatom P. tricornutum and Escherichia coli K12. These authors focused on the extracellular proteome and found bacterial proteins induced during co- cultivation that were associated with transport, biofilm formation, and carbohydrate turnover. In a co-cultivation experiment of Thalassiosira pseudonana and the member of the Roseobacter clade Dinoroseobacter shibae , Paul et al. (2013) used metabolomic techniques for the analysis of intracellular diatom metabolites. They observed an increase in amino acid concentration in T. pseudonana during co-cultivation. Additionally, long-chain fatty acids and undefined carbohydrates were increased in the diatom in the presence of bacteria. 27 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Aim of the present study was to contribute to an in-depth understanding of bacteria-diatom interactions. A proteomics approach was applied to a model interaction system comprising the diatom T. weissflogii and the marine bacterium M. adhaerens HP15. This model system had been established in the past in order to investigate bacterial colonization of diatoms and formation of marine aggregates (Sonnenschein et al. , 2011). It has since successfully served in a set of studies dissecting this interaction (Gärdes et al. , 2011; Gärdes et al. , 2012; Sonnenschein et al. , 2012). With the current proteomics approach, bacterial proteins involved in the bacteria-diatom interaction were identified. Their potential role during the interaction was substantiated by metabolic profiling of the bacterium.

28 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Methods

Culture conditions for M. adhaerens HP15 and T. weissflogii M. adhaerens HP15 (Grossart et al. , 2004; Kaeppel et al. , 2012) was cultured in liquid marine broth (MB) medium (Sonnenschein et al. , 2011) at 18°C under constant shaking at 250 rpm. Axenic T. weissflogii diatom cultures were grown in f/2 medium prepared from filtered and autoclaved North Sea water (NSW) (15°C; 12/12 h light/ dark period at 150 µm photons m -2 s- 1 -4 -5 ). In detail, f/2 medium was supplemented with 8.82 x 10 M NaNO3, 3.62 x 10 M -4 -1 NaH 2PO 4 · 2 H2O, 1.06 x 10 M Na 2SiO3 · 9 H 2O, 1 ml l trace element and vitamin solution according to Guillard (Guillard, 1975). Bacterial cell numbers during the experiment were determined by serial dilution plaiting in triplicates, diatom numbers were determined with a Sedgwick-Rafter counting chamber, using a Zeiss Axiostar plus microscope (Carl Zeiss, Jena, Germany) at 100x magnification.

Co-cultivation of M. adhaerens HP15 with T. weissflogii M. adhaerens HP15 cells were harvested from liquid culture in exponential growth phase

(OD 600 0.5-1.0) by centrifugation (15 min, 3,000 rcf, 4°C). Cells were washed twice in 75% (v/v) NSW, resuspended and starved for 3 h in 75% NSW at 18°C under constant shaking (250 rpm). Cells were further on pelleted and resuspended in 75% NSW supplemented with 3.4 x 10 -2 M glutamate as sole carbon and nitrogen source and 2.12 x 10 -3 M phosphate

(NaH 2PO 4 · 2H 2O). The phosphate content was chosen according to the nitrogen molarity provided by glutamate and the thereof resulting favorable Redfield Ration of N:P, 16:1, respectively (Hecky et al. , 1993). The initial start OD600 for the bacterial culture was adjusted to 0.05 (~7 x 10 6 cells ml -1). Three individually grown T. weissflogii pre-cultures in equal growth state (TW I-III) were split in two parts (Figure 1). Half of each culture was used for sampling of culture supernatant. For this, cultures were centrifuged under mild conditions (1,250 rcf, 4°C) to avoid cell lysis and the supernatant was filtered afterwards through 0.2 µm pore size filters. The second half was used for direct co-cultivation of T. weissflogii with the bacterium.

29 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Figure 1: Experimental set-up of co-cultivation experiment. Three individual pre-cultures of T. weissflogii were cultivated (TW I - TW III). Each culture was split in two parts; one part was used for sampling of the culture supernatant by centrifugal removal of the diatom cells, the second part was directly used for co-cultivation. Both culture supernatant (REFERENCE) and direct diatom culture (TREATMENT) were filled in dialysis hoses and combined with a M. adhaerens HP15 culture. M. adhaerens HP15 was previously prepared in f/2 medium -2 supplemented with 3.4 x 10 M glutamate (OD 600 = 0.05) and nutrients. Both REFERENCE and TREATMENT were conducted in triplicates (REF I - REF III; TRE I - TRE III). Cultures were gently stirred during incubation to support diffusion of compounds through the dialysis hose.

Two treatments were chosen, in which M. adhaerens HP15 was cultivated either with the supernatant of the diatom culture, serving as reference for the two-dimensional protein gel analysis (REFERENCE, REF I-III) or was co-cultivated with the diatom culture itself, serving as the actual treatment (TREATMENT, TRE I-III, Figure 1 ). For these cultivations, 100 ml of diatom culture supernatant or diatom culture, respectively, were loaded in dialysis hoses (25 mm Zellu Trans hose, MWCO 12,000-14,000; Carl Roth, Karlsruhe, Germany). Filled hoses were placed into a 200 ml suspension of M. adhaerens HP15 cells, previously starved, pelleted, and resuspended in corresponding medium to an OD 600 of 0.05. All treatments were

30 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

further supplemented with trace elements, vitamins, and nutrients as done for f/2 medium. Treatments were carried out in triplicates. Culture conditions used were those applied for diatom cultivation (15°C; 12/12 h light/ dark period at 150 µm photons m -2 s-1). Cultures were gently stirred (50 mm magnetic stirring bar, 60 rpm) to guarantee diffusion of chemical signal molecules through the dialysis hose.

Protein sampling, two-dimensional sodium dodecyl sulfate polyacrylamide gel electrophoresis (2D-SDS-PAGE), and data analysis Bacterial cells were harvested by centrifugation at 3,000 g and resuspended in chilled sonication buffer (100 mM Tris-HCl, pH 7.5; 50 mM NaCl; 0.5 mM dithiothreitol (DTT); Protease Inhibitor Cocktail according to manufacturer’s instruction (Thermo Scientific, Schwerte, Germany)). Cell suspensions were sonicated on ice (10 x 1 sec, 4 sec break, 3 repeats; amplitude 70%; Active Motif ® sonicator, Regensburg, Germany) and cell debris was centrifuged off afterwards (30 min, 16,100 rcf, 4°C). Supernatants were immediately used or kept at -80°C. Protein content of supernatant was determined via bicinchoninic acid test (Thermo Scientific) according to manufacturer’s instruction. For conducting isoelectric focusing, 80 µg of protein were applied to immobilized pH gradient (IPG) strips (7 cm, pH 4-7; Biorad, München, Germany). Corresponding amounts of protein were precipitated and cleaned via acetone precipitation. In detail, 1 volume of protein suspension was mixed with 4 volumes of ice-cold acetone (-20°C). Samples were kept at - 20°C for at least 3 hours. Protein pellet was harvested afterwards by centrifugation (30 min, 16,100 rcf, 4°C). The supernatant was decanted and the protein pellet briefly air-dried. The pellet was resuspended in rehydration buffer (2 M thiourea, 6 M urea, 16.2 x 10 -3 M CHAPS, 25.9 x 10 -3 M DTT) and supplemented with ampholytes according to manufacturer’s specification (Biorad). IPG-strips were soaked with protein suspension for ~14 h. Isoelectric focusing was carried out on a Biorad Protean ® i12 TM IEF Cell (50 V, 70 min; 150 V, 20 min; 300 V, 15 min; gradient to 600 V, 10 min; 600 V, 15 min; gradient to 1,500 V, 10 min; 1,500 V, 30 min; gradient to 3,000 V, 20 min; 3,000 V, 210 min; pause on 50 V). Strips were afterwards either kept at -80°C till further use or used immediately for run of the 2 nd dimension. IPG-strips were equilibrated beforehand, each 15 min in 6.48 x 10 -2 M DTT and 0.216 M iodoacetamide solution, dissolved in equilibration buffer (6 M urea, 30% (w/v) glycerol, 69.2 x 10 -3 M SDS in 0.05 M Tris-HCl buffer, pH 8.8)). Molecular weight separation was conducted on a Biorad Mini-Protean ® Tetra System (50 mV, 10 min; 110 mV 31 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

further on) via a 12.5% acrylamide concentrated polyacrylamide gel. Resulting gel was Coomassie ® stained for 20 min (45% (v/v) methanol, 10% acetic acid, 2.93 x 10 -3 M Coomassie ® Brilliant Blue G-250) and further treated in corresponding destaining solution (10% (v/v) acetic acid, 5% (v/v) 2-propanol). Destained gels were scanned (600 dpi resolution) and analyzed for differential protein patterns using Delta 2D software (Decodon, Greifswald, Germany). Gels resulting from bacterial co-cultivation with diatom supernatant were set as reference gels (REF I-III), those gels resulting from co-cultivation with diatoms itself (treatment; TRE I-III) were compared to the reference gels ( Figure 1). Only significant results ( p<0.05) based on three replicates with at least 2-fold changes were considered (2-fold up-regulation, 0.5 fold down-regulation). All results were manually reviewed with respect to correct spot definition and warping.

Matrix assisted laser desorption/ ionization-time of flight mass spectrometry (MALDI- TOF MS) analysis Protein spots of interest were excised from SDS gels, chopped into small pieces and washed twice each 15 min in 100 µL of 0.05 M ammonium bicarbonate buffer, containing 50% acetonitrile (v/v). Gel pieces were finally dehydrated by addition of 500 µL acetonitrile. After decanting and short air-drying, samples were further supplemented with trypsin digestion buffer (Promega, Mannheim, Germany) according to Shevchenko and colleagues (Shevchenko et al. , 2006). Fast tryptic digest was carried out by incubation of samples at 55°C for 30 min (Shevchenko et al. , 2006). Sample supernatant was directly used for MALDI-TOF MS analysis: Supernatant was mixed in a 1:1 ratio with a saturated α-cyano-4- hydroxycinnamic acid solution beforehand (prepared in 30% acetonitrile, 0.5% trifluoroacetic acid; non-dissolved solids of matrix were removed by centrifugation (15 min, 16,100 rcf)). An Autoflex II TOF/TOF mass spectrometer (Bruker Daltonics, Bremen, Germany) was used according to standard parameters (acquisition range 840-4000 Da; S/N = 6, in specific cases 3; error range 50 ppm; allowed miss cleavages = 1; potential modifications ‘Oxidation (M)’). Peptide masses derived from trypsin auto digestion were used for calibration (842.50940; 1045.56370; 1713.80840; 1774.89750; 2083.00960; 2211.10400; 2283.18020 Da). Obtained mass lists were identified using the MASCOT search engine (Perkins et al. , 1999). All results were reviewed with respect to the identified organism and expected molecular weight as retrieved from SDS-PAGE.

32 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Metabolic profiling with different carbon sources in liquid and on solid medium The use of potential carbon sources of M. adhaerens HP15 was tested both in liquid and on solid medium. 5 g L -1 of a compound were prepared in 75% NSW, containing additions of -2 -3 3.62 x 10 M NaH 2PO 4 · 2 H2O, 1 x 10 M NH 4Cl, and trace elements as used for f/2 medium. The pH was adjusted to 7.6 with 1 M NaOH solution. In case of solid medium, 12 g L-1 agar were supplemented. 21 carbohydrates, 17 amino acids, two protein-associated and 12 further compounds were tested as potential carbon source. In more detail, 10 monosaccharides (L(+)-arabinose, galactose, N-acetyl-D-glucosamine, fructose, D(+)-fucose, D-maltose, D(+)- mannose, N-acetyl-D-mannosamin, L(+)-rhamnose, D(+)-xylose), 3 disaccharides (lactose, D-maltose, and D(+)-sucrose), 1 trisaccharide (raffinose), and 7 polysaccharides (alginic acid, carboxymethyl-cellulose, carrageenan, cellulose, chitin, dextrin, and xylan) were applied. The tested amino acids and protein-associated compounds were: alanine, L-arginine, L-asparagine, cystein, glutamate, glutamine, glycine, isoleucine, histidine, leucine, lysine, methionine, proline, phenylalanine, serine, L-threonine, and valine, in addition bovine serum albumin V (BSA) and a casamino acid mix. Further compounds tested were acetate, butyrate, citrate, lecithin, malate, nicotininc acid, octanoic acid, putrescine, propionate, pyruvate, spermidine, and succinate. In case of amino acids, experiments were additionally conducted under supplementation of another carbon source known to support growth (5 g L -1 glutamate). This additional test aimed to determine any potential toxic effects on the bacterium caused by the provided amino acid.

Databases used for protein and sequence interpretation Information on available metabolic pathways and uptake systems of M. adhaerens HP15 as well as gene cluster analysis was retrieved from the Kyoto Encyclopedia of Genes and Genomes (KEGG) website (Kanehisa & Goto, 2000). Proteins and gene loci of the bacterium were obtained from the National Center of Biotechnology Information (NCBI) website, where the full genome sequence of the bacterium is available (CP001978.1 (chromosome); CP001979.1 (plasmid pHP-42); CP001980.1 (plasmid pHP-187)). Gene and protein similarity searches were conducted, using the Basic Local Alignment Search Tool (BLAST) offered by the NCBI. Further information on proteins of M. adhaerens HP15 involved in carbohydrate utilization was derived from the carbohydrate-active enzymes (CAZy) database (Cantarel et al. , 2009; Lombard et al. , 2014).

33 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Results

Sampling of co-cultivation experiment M. adhaerens HP15 cells were incubated with the supernatant of a diatom culture (REFERENCE) or the diatom culture itself (TREATMENT, Figure 1) for 66.5 h. The obtained cell numbers upon sampling of both treatments did not differ significantly, being 2.6 x 10 9 cells ml -1 (± 7.1 x 10 8 cells ml -1) for the REFERENCE and 3.6 x 10 9 cells ml -1 (± 3.3 x 10 8 cells ml -1) for the TREATMENT. The culture volume of ~200 ml was used for extraction of bacterial proteins. Diatom cultures grew from 7.62 x 10 5 cells/ ml (± 40 cells ml -1) at the beginning of the incubation to 1.92 x 10 6 cells ml -1 (± 74 cells ml -1) after 66.5 h.

Changes in the bacterial protein pattern during co-cultivation The exposure of M. adhaerens HP15 towards T. weissflogii led to a set of significant changes in certain protein abundances. In summary, 10 proteins were found to be at least two-fold up- or down-regulated during co-cultivation of bacteria with diatom cells (TREATMENT, Figure 1) in comparison to the REFERENCE (Table 1). Gene products of M. adhaerens HP15 encoding two alcohol dehydrogenases (ADP98924.1 and ADP98928.1) and one aldehyde dehydrogenase (ADP98908.1) were down-regulated when diatom cultures were present. Notable for these three genes is their close proximity in the genome sequence: The aldehyde dehydrogenase gene is separated from the two loci comprising the alcohol dehydrogenase genes by 15 genes, while the latter two genes are separated by only three additional genes ( Table 1; Supplementary 1). Due to close proximity of these genes, a cluster analysis for this genomic region was conducted to eventually deduce functional genomic embedment of the respective gene products. The cluster analysis function of the KEGG website was used for this purpose. Results showed that the corresponding genomic region is separated into six different clusters. Importantly, the three genes encoding the down-regulated proteins are not encoded in the same cluster ( Supplementary 1). However, a number of genes in this locus encode proteins that are associated with activities and uptake systems for nitrogen-rich compounds. Examples for this are urea carboxylase- associated proteins (ADP98910.1 & ADP98911.1) and a urea amidolyase-like protein (ADP98913.1), a glutamate-ammonia ligase (ADP98920.1), a spermidine/ putrescine ABC transporter ATPase (ADP98923.1) and the homoserine O-acetyltransferase MetX involved in cysteine and methionine metabolism (ADP98921.1) ( Supplementary 2). 34 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Table 1: Proteins of M. adhaerens HP15 cell lysate, found to be at least 2-fold up- or down-regulated during co- cultivation with the diatom T. weissflogii .

Identified protein EC Additional info Accession (annotation according to NCBI) number

Down-regulated during co-cultivation with the diatom:

quinoprotein alcohol dehydrogenase alcohol <-> aldehyde ADP98924.1 1.2.1.8

low quality protein: quinoprotein alcohol <-> aldehyde ADP98928.1 1.2.1.8 alcohol dehydrogenase

NAD-dependent aldehyde aldehyde <-> alcanic acid ADP98908.1 1.2.1.- dehydrogenase

isocitrate lyase isocitrate -> glyoxylate + succinate ADP98998.1 4.1.3.1 (glyoxylate cycle)

3-hydroxyisobutyrate dehydrogenase valine catabolism ADP96674.1 1.1.1.31

urea short-chain amide or branched- UrtA component of urea ADP98645.1 - chain amino acid uptake ABC transporter transporter, periplasmic solute- binding protein

Up-regulated during co-cultivation with the diatom:

phosphonate ABC transporter, PhnD unit of phosphonate ADP97492.1 - periplasmic phosphonate-binding transporter protein

extracellular solute-binding protein, AotJ unit of an arginine/ ornithine ADP98795.1 - family 3 amino acid transporter

TRAP dicarboxylate family - ADP98792.1 - transporter, DctP subunit

amino acids binding protein LivK unit of branched amino acid ADP99862.1 - uptake transporter, periplasmic

Three additional gene products were down-regulated in M. adhaerens HP15 when encountering diatom cells. These ones included an isocitrate lyase potentially involved in the glyoxylate cycle (ADP98998.1), a 3-hydroxyisobutyrate dehydrogenase putatively involved in valine catabolism (ADP98998.1), and a transporter component which may be involved in short chain amide uptake, possibly also branched amino acid uptake (ADP98645.1) (Table 1). By additional BLAST analysis, the latter transporter component was identified as the UrtA component of a urea transporter. Four M. adhaerens HP15 proteins were identified as being up-regulated during co-cultivation with the diatom ( Table 1). Two of these proteins are encoded by genes located in close proximity to each other on the genome suggesting their functional coupling. These genes encode for a DctP subunit of a tripartite ATP-independent periplasmic (TRAP) dicarboxylate family transporter (ADP98792.1) and a protein annotated as extracellular solute-binding

35 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

(ADP98795.1). Analysis of the latter protein using the KEGG database assigned it as the AotJ component of an arginine/ ornithine amino acid transporter. Its potential function is substantiated by presence of surrounding genes, encoding the subunit of a histidine-lysine- arginine-ornithine transporter (HisP; ADP98794.1) and two inner membrane permeases of arginine and histidine associated ABC transporters (ArtQ; ADP98796.1, and ArtM; ADP98797.1). The third bacterial protein up-regulated during co-cultivation with diatoms was annotated as a periplasmic substrate-binding protein of an ABC transporter and might be a homolog of the PhnD unit of an ABC phosphonate transporter (ADP97492.1). This assumption is supported by the presence of genes encoding for the two other transporter components PhnC (ADP97493.1) and PhnE (ADP97494.1) within the same gene cluster. The fourth of the up-regulated M. adhaerens HP15 proteins was annotated as an amino acid- binding protein (ADP98795.1,), likely being a homolog of the periplasmic LivK unit of a branched amino acid uptake transporter. Again, a functional involvement of this protein in branched amino acid transport is supported by the close proximity of genes, encoding the inner membrane translocators LivH and LivG as well as the ATP binding components LivM and LivF (Adams et al. , 1990). In summary, the data of proteomics analysis suggested that genes encoding for proteins involved in amino acid uptake or utilization seemed to be affected in their expression when M. adhaerens HP15 is co-cultivated with diatom cells.

Vitamin synthesis Using the KEGG database, M. adhaerens HP15 biosynthetic pathways were analyzed in order to identify the organism’s ability to synthesize cobalamin (vitamin B 12 ). According to this in silico analysis, the bacterium is not able to synthesize the vitamin. However and interestingly, other Marinobacter species seem to be capable, as there are corresponding genes present in the genomes of M. hydrocarbonoclasticus VT8 and M. sp. BSs20148. Protein sequences of M. hydrocarbonoclasticus VT8 identified to be involved in cobalamin synthesis were blasted against the genome of M. adhaerens HP15 using tblastn. However, this did not yield in any matching information for a potential cobalamin synthesis by M. adhaerens HP15. Although biosynthesis pathways for other vitamins appear to be present in M. adhaerens HP15, as for riboflavin (vitamin B 2), panthothenate (B 5), pyroxidine (B 6), biotin (B 7), or folate (B 9), this organism does not seem to produce cobalamin. 36 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Analysis of carbon source utilization of M. adhaerens HP15 In order to experimentally support some of the proteomics findings, a set of organic compounds such as diverse carbohydrates, amino acids, and other molecules like short- chained fatty acids were tested as potential sole carbon sources of M. adhaerens HP15. Among all tested carbohydrates, fructose was the only one that facilitated growth of the bacterium in both liquid and solid medium ( Table 2).

Table 2: Compounds tested as potential carbon source for M. adhaerens HP15. Compounds that supported growth are shown in bold .

Compounds

Carbohydrates

Monosaccharides

L(+)-arabinose, galactose, N-acetyl-D-glucosamine, fructose , D(+)-fucose, D-maltose, D(+)- mannose, N-acetyl-D-mannosamine, L(+)-rhamnose, D(+)-xylose

Disaccharide

lactose (D-Glucose & D-Galactose; β-1-4), D-maltose (degradation product of starch), D(+)-sucrose (α-D-Glucose & β-D-Fructose)

Trisaccharide

affinose (Galactose, Glucose & Fructose)

Poly-carbohydrates

alginic acid, carboxymethyl-cellulose, carrageenan, cellulose, chitin, dextrin (polymers of D-glucose, α(1-4) or α(1-6) linkage), xylan

Amino acids and protein associated compounds

alanine , L-arginine , L-asparagine , cystein, glutamate , glutamine , glycine *1 , isoleucine , histidine, leucine , lysine , methionine, proline , phenylalanine , serine, L-threonine, and valine

bovine serum albumin fraction V , casamino acids

Further compounds tested

acetate , butyrate , citrate , lecithin , malate , nicotininc acid, octanoic acid, putrescine, propionate , pyruvate , spermidine, and succinate

*1 glycine showed inhibitory effects, as also no growth was observed in presence of a second, growth supporting carbon source

The analysis of data retrieved from KEGG showed that no classical mono- or oligosaccharide transporter could be identified in M. adhaerens HP15 (Supplementary 3 & 4 ). Interestingly, genes encoding for phosphotransferases are only present in the case of a fructose phosphatase system (Supplementary 5) thus supporting the experimental result. Subsequently, a screening for genes associated with carbohydrate utilization was done using the CAZy database. In total, 64 proteins encoded within the M. adhaerens HP15 genome were

37 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

identified to be associated with carbohydrate utilization and sub-classified as glycoside hydrolases, glycosyl transferases, polysaccharide lyases, carbohydrate esterases, and the carbohydrate binding module family ( Table 3). For comparison, we additionally retrieved the information for carbohydrate-utilizing proteins of the marine bacterium Flavobacterium johnsoniae UW101 (McBride et al. , 2009), a representative of the Bacteroidetes group known to degrade algal polysaccharides. For this bacterium, a total of 281 carbohydrate utilization- associated genes were found ( Table 3), suggesting that the high number of carbohydrate- active enzymes in F. johnsoniae UW101 fairly well represents the metabolic profile and capabilities of this bacterium as an algae polysaccharide-associated organism.

Table 3: Amount of carbohydrate-utilizing gene products of M. adhaerens HP15 and F. johnsoniae UW101 . Numbers were retrieved from the CAZy database.

Organism GH GT PLF CE CBM SUM % of total genes

M. adhaerens HP15 15 41 0 1 7 64 1.43

F. johnsoniae UW101 156 64 11 18 30 279 5.41

GH = glycoside hydrolase family; GT = glycosyl transferase family; PLF = polysaccharide lyase family; CE = carbohydrate esterase family, CBM = carbohydrate-binding module family.

Whilst the majority of carbohydrates were apparently not metabolized by our bacterial model organism, certain amino acids and protein-associated compounds turned out to be growth- supporting for M. adhaerens HP15. Such amino acids were alanine, L-arginine, L-asparagine, glutamic acid, glutamine, isoleucine, leucine, lysine, proline, and phenylalanine ( Table 2). The bacterium further grew on BSA and lecithin as sole carbon source. Glycine showed to be cell toxic at the used concentrations, as the bacterium did neither grow in its presence, nor in the presence of both glycine and a second metabolizable carbon source. Other non- metabolized amino acids did not show any inhibitory effects.

38 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Discussion

Previously, only few studies aimed at a better understanding of bacteria-diatom interactions by using in-depth approaches based on in vitro co-cultivations. Paul et al . (2013) conducted co-cultivation studies with D. shibae and the diatom T. pseudonana , focusing on the detection and identification of intracellular metabolites of the diatom in presence of the bacterium. Bruckner et al. (2011) conducted co-cultivations of different fresh water diatom species with a set of bacterial strains. Besides studying the diatom’s growth behavior, TEP formation, and release of amino acids, these authors also identified extracellular proteins of both partners being induced during co-cultivation. Bacterial extracellular protein patterns showed certain protein-binding and transporter proteins as well as carbohydrate modifying proteins in presence of the diatom. Proteins involved in biofilm formation were further identified. Amin and colleagues (Amin et al. , 2015) have thus far conducted the most comprehensive study on bacteria-diatom interactions: They combined information gained from transcriptomics data of both co-cultivated partners with metabolomics data. Genes and metabolites possibly playing a role during a specific diatom-bacteria interaction were identified. A number of environmental meta-transcriptomic and metabolomic data sets were further screened for the presence of identified metabolites like IAA and the expression of genes, involved in pathways probably induced during bacteria-diatom interaction. These results confirmed that genes of the IAA synthesis pathways were actively transcribed in environmental settings. Further, IAA was identified in all analyzed marine environments and correlated with chlorophyll concentrations. Interestingly, none of the studies had dealt with the cytoplasmic proteome of the bacterial interaction partner.

M. adhaerens HP15 is not a carbohydrate degrader In individual cases, up to 80% of algal exudates are made up by carbohydrates (Myklestad et al. , 1989; Myklestad, 1995; Biddanda & Benner, 1997) regarded as readily bacterial attractants (Seymour et al. , 2008). Common sugars identified in such extracellular matrices may be specific for a given algae species and its respective growth state (Bruckner et al. , 2011). Arabinose, fucose, glucose, galactose, mannose, rhamnose, and xylose are those sugars, frequently identified in algal exudates (Biersmith & Benner, 1998). Consequently, the majority of these carbohydrates were considered experimentally in the present study. Certain

39 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

bacterial groups such as members of the Bacteroidetes are known for their ability to degrade carbohydrate compounds resulting from algae exudates (Teeling et al. , 2012; Hahnke et al. , 2013; Mann et al. , 2013). According to our proteomics and metabolite profiling results, except for fructose none of the tested carbohydrates were used as a carbon source by M. adhaerens HP15. In support of this, the proteomics approach did not show any carbohydrate uptake systems or carbohydrate-utilizing proteins being up-regulated in M. adhaerens HP15 in presence of the diatom. The findings were further substantiated by the remarkably low number of carbohydrate utilization-associated proteins encoded by the genome of M. adhaerens HP15 as compared to those of classical degraders of algae exudates such as F. johnsoniae UW101 (McBride et al. , 2009) or other members of the Flavobacteria (Mann et al. , 2013).

Shifts in protein patterns suggest a favorable nutrient state during co-cultivation Expression of one aldehyde and two quinoprotein alcohol dehydrogenases was apparently down-regulate in M. adhaerens HP15 during co-cultivation. Such enzymes usually facilitate the oxidation of alcohols via aldehydes towards alcanoic acids. Interestingly, the respective down-regulated dehydrogenases genes are located in a genomic region that also encodes a number of proteins, associated with the utilization of nitrogen-rich molecules, as there is urea, amino acids, spermidine and putrescine. We tested if these proteins would be organized in a gene cluster. However, this was not the case, as the genomic region of concern is divided in different clusters, from which a functional relationship could not be easily retrieved. A functional hint, however, becomes visible when looking at homologous genes in another marine bacterium: The γ-proteobacterium Alteromonas sp. KE10 expresses the aldehyde dehydrogenase OlgA (BAA24014.1) during low-nutrient stress. During high-nutrient conditions, this dehydrogenase was found to be down-regulated (Maeda et al. , 2000). The aldehyde dehydrogenase ADP98908.1 of M. adhaerens HP15 down-regulated in presence of the diatom shows a 74% protein identity to OlgA of Alteromonas sp. KE10. Therefore the presence of the diatom might represent a favorable nutrient status for the bacterium. Isocitrate lyase is a central enzyme in the glyoxylate cycle that was found to be down- regulated in the bacterium during co-cultivation with the diatom. This enzyme represents a branching point in glyoxylate cycle due to the reaction of isocitrate to succinate and glyoxylate. While glyoxylate continues fueling the glyoxylate cycle, succinate replenishes 40 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

the tricarboxylic acid (TCA) cycle (Vanni et al. , 1990). In case of a suppression of the TCA cycle due to, e.g., unfavorable nutrient supply, the glyoxylate cycle may short-cut the reaction from isocitrate via α-ketoglutarate and succinyl-CoA to succinate (as it would happen in the TCA cycle) in a single step reaction. This reaction is carried out by isocitrate lyase (Betts et al. , 2002). The down-regulation of a central enzyme of the glyoxylate cycle thus indicates a favourable substrate supply of the TCA cycle. Consequently, the bacterial cell metabolism seems to face more favorable nutrient conditions during co-cultivation with the diatom as compared to the growth in the REFERENCE medium. We further observed the down-regulation of the periplasmic substrate-binding compound UrtA of the urea uptake system (Beckers et al. , 2004) during bacteria-diatom interaction. Urea serves as a nitrogen source for a wide range of bacteria (Mobley & Hausinger, 1989). The operon encoding for the urea uptake system, being made up of the five genes urtABCDE was previously shown to be induced by nitrogen-limiting conditions in Corynebacterium glutamicum (Beckers et al. , 2004). Therefore, the down-regulation of the UrtA protein in the current study implies that the nitrogen supply in the TREATMENT seems to be more optimal as compared to the REFERENCE in which the amino acid glutamate is the only nitrogen source provided. The enzyme 3-hydroxyisobutyrase dehydrogenase, less expressed in presence of the diatom, is part of the valine degradation pathway and oxidizes 3-hydroxyisobutyrate to methylmalonate semialdehyde (Robinson & Coon, 1957; Bannerjee et al. , 1970). Reduced expression of a respective gene might suggest that the bacterium was not in need for valine utilization as carbon or nitrogen source during co-cultivation. However, since the corresponding protein was present when glutamate only was provided (Debarbouille et al. , 1991) it is assumed that in the REFERENCE treatment, probably nitrogen and carbon has become more limiting. Alternatively, nitrogen supply might be rather balanced during co- cultivation of M. adhaerens HP15 with the diatom. This hypothesis is in line with the down-regulation of UrtA as part of the urea uptake system, and leads to the conclusion that the diatom might provide an additional source of nitrogen to the bacterium and that this source or signal is furthermore different from glutamate.

41 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Compounds gained by M. adhaerens HP15 during co-cultivation with the diatom T. weissflogii According to the current results, proteins, peptides, or amino acids might be the major carbon sources of the bacterium. Their presence in exopolymeric substances has been shown in previous studies (Myklestad et al. , 1989; Myklestad, 1995; Biddanda & Benner, 1997). During exposure to the diatom, a possible benefit for M. adhaerens HP15 might be the release of specific amino acids by the diatom: Two bacterial proteins up-regulated while being co- cultivated with T. weissflogii can be associated with amino acid uptake and facilitate the binding of amino acids in the periplasm. LivK is part of a branched amino acid uptake system transporting leucine, isoleucine, and valine (Adams et al. , 1990; Ribardo & Hendrixson, 2011). AotJ, another amino acid binding protein, belongs to an arginine/ ornithine uptake system (Wissenbach et al. , 1995; Nishijyo et al. , 1998). A third up-regulated transport- associated protein, a DctP subunit, could not be assigned to a substrate, but is suspected to be part of the above mentioned arginine/ ornithine transporter due to the close proximity of the corresponding genes. Amino acids other than glutamate might serve as possible additional carbon and nitrogen sources when diatoms are present suggesting that several of such amino acids might be secreted by the diatom cells. This finding is supported by the down-regulation of a urea uptake transporter element during co-cultivation with the diatom, a process usually associated with growth under nitrogen-rich conditions (Beckers et al. , 2004). Previous studies support our hypothesis of amino acids playing a major role during bacteria- diatom interactions (Amin et al. , 2015): Intracellular free amino acid concentrations were found to be increased in T. pseudonana during exposure to D. shibae (Paul et al. , 2013). Alterations in the concentration of extracellular free amino acids were also observed in co- cultivations of mainly benthic freshwater diatoms towards different bacterial strains (Bruckner et al. , 2011). Gärdes et al. (2012) confirmed for T. weissflogii cultures that the quantity and quality of both dissolved free and dissolved combined amino acids is highly dependent on the nutrient status of the culture. Further, the axenic or xenic status (being co- cultivated with M. adhaerens HP15) of the culture impacted the net concentration of free and combined amino acids (Gärdes et al. , 2012). Especially a shift in amino acid composition during phosphate-depleted conditions was observed, irrespective of the culture being xenic or axenic. The proportions of the branched amino acids leucine and isoleucine, but not of valine, of the total amino acids pool were found to increase during this stage of phosphate limitation. 42 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

In detail, the ratio of leucine and isoleucine as part of total dissolved free amino acids increased from ~5% in both nutrient-balanced and nitrogen-depleted scenarios to ~30% in the phosphate depleted scenario. Indeed, indications for a phosphate limitation were identified in the present study: An ABC type transporter component, a phosphonate substrate binding unit (PhnD), was up-regulated in M. adhaerens HP15 in the presence of the diatom. The phnD gene was previously shown to be induced during phosphate-limiting conditions (Metcalf & Wanner, 1991; Gebhard et al. , 2006). This finding would further explain and support the up- regulation of a subunit of an uptake transporter for branched amino acids in presence of diatom cells in the current study. However, whether the release of such amino acids by the diatom is indeed due to phosphate limitation, as suggested by Gärdes et al. (2012), remains speculative and deserves further investigation. An obvious and unsolved question is why diatoms actively release high-value metabolites such as amino acids when getting in contact with certain bacteria. One possible explanation is a potential trade-off: The diatom provides a high nutrient investment in order to trigger attachment of certain bacteria. In turn, these bacteria might be favorable for the diatom by providing a metabolite to which the diatom is auxotroph. Such metabolite remains to be identified in further experiments. In case of a different study, such metabolite has already been demonstrated. Sulfitobacter sp. SA11 received tryptophan from its diatom partner and provided IAA to the host organism (Amin et al. , 2015). Assumable, tryptophan was even directly used for the synthesis of IAA in the bacterium.

Potential benefits for the diatom T. weissflogii during co-cultivation with M. adhaerens HP15 Our results suggest that M. adhaerens HP15 might benefit from diatom-borne amino acids. The applied experimental set-up does not allow any conclusions on beneficial compounds, received by the diatom, however, some theoretical assumptions can be made. A potential benefit of this interaction for the diatom could be the allocation of vitamins by the bacterium.

Cobalamin (vitamin B 12 ) has been postulated as such a potential candidate by others (Cole, 1982; Croft et al. , 2005). However, the genome analysis of M. adhaerens HP15 revealed that this organism does obviously not synthesize cobalamin, but might produce a set of other vitamins that may determine synergistic bacteria-diatom interactions (Croft et al. , 2006; Wagner-Doebler et al. , 2010) which deserves future investigations.

43 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Conclusion

Herein, a proteomics approach combined with metabolic profiling was chosen to investigate processes taking place in the bacterial cell during its interaction with diatoms. The approach allowed the interpretation of some characteristics during the interaction. Our results supported the possible importance of amino acids as a ‘traded’ component group serving as a valuable nitrogen and further carbon source for bacteria. However, actual benefits for the diatom remained highly speculative in this study. Therefore future studies should aim to include proteome analysis of the diatom to better understand adaptions taking place in this organism during interaction with bacteria. Corresponding studies should be extended to metabolic analysis of bacteria-diatom interaction in order to develop a mechanistic understanding of the process.

Acknowledgement

This project was funded by the Helmholtz Graduate School for Polar and Marine Research (POLMAR) and the Deutsche Forschungsgemeinschaft (UL 169/6-1).

44 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

References

Adams M, Wagner L, Graddis T, Landick R, Antonucci T, Gibson A , et al. (1990). Nucleotide sequence and genetic characterization reveal six essential genes for the LIV-I and LS transport systems of Escherichia coli . J Biol Chem 265:11436-11443.

Alldredge AL, Gotschalk CC (1989). Direct observations of the mass flocculation of diatom blooms - characteristics, settling velocities and formation of diatom aggregates. Deep-Sea Research Part A-Oceanographic Research Papers 36:159-171.

Amin SA, Hmelo LR, van Tol HM, Durham BP, Carlson LT, Heal KR , et al. (2015). Interaction and signalling between a cosmopolitan phytoplankton and associated bacteria. Nature 522:98-U253.

Amin SA, Parker MS, Armbrust EV (2012). Interactions between diatoms and bacteria. Microbiol Mol Biol Rev 76:667-684.

Amin SA, Green DH, Hart MC, Kuepper FC, Sunda WG, Carrano CJ (2009). Photolysis of iron-siderophore chelates promotes bacterial-algal mutualism. Proc Natl Acad Sci U S A 106:17071-17076.

Azam F, Smith D, Steward G, Hagstrom A (1994). Bacteria-organic matter coupling and its significance for oceanic carbon cycling. Microb Ecol 28:167-179.

Azam F, Malfatti F (2007). Microbial structuring of marine ecosystems. Nat Rev Microbiol 5:782-791.

Baker LJ, Kemp PF (2014). Exploring bacteria-diatom associations using single-cell whole genome amplification. Aquat Microb Ecol 72:73-88.

Bannerjee D, Sanders LE, Sokatch JR (1970). Properties of purified methylmalonate semialdehyde dehydrogenase of Pseudomonas aeruginosa . J Biol Chem 245:1828-1835.

Beauvais S, Pedrotti ML, Egge J, Iversen K, Marrase C (2006). Effects of turbulence on TEP dynamics under contrasting nutrient conditions: Implications for aggregation and sedimentation processes. Mar Ecol Prog Ser 323:47-57.

Beckers G, Bendt A, Kramer R, Burkovski A (2004). Molecular identification of the urea uptake system and transcriptional analysis of urea transporter- and urease-encoding genes in Corynebacterium glutamicum . J Bacteriol 186:7645-7652.

Behrenfeld M, Bale A, Kolber Z, Aiken J, Falkowski P (1996). Confirmation of iron limitation of phytoplankton photosynthesis in the equatorial Pacific Ocean. Nature 383:508-511.

Bell W, Mitchell R (1972). Chemotactic and growth responses of marine bacteria to algal extracellular products. Biol Bull 143:265-277.

Betts J, Lukey P, Robb L, McAdam R, Duncan K (2002). Evaluation of a nutrient starvation model of Mycobacterium tuberculosis persistence by gene and protein expression profiling. Mol Microbiol 43:717-731.

Biddanda B, Benner R (1997). Carbon, nitrogen, and carbohydrate fluxes during the production of particulate and dissolved organic matter by marine phytoplankton. Limnol Oceanogr 42:506-518.

Biersmith A, Benner R (1998). Carbohydrates in phytoplankton and freshly produced dissolved organic matter. Mar Chem 63:131-144.

45 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Blackburn N, Fenchel T, Mitchell J (1998). Microscale nutrient patches in planktonic habitats shown by chemotactic bacteria. Science 282:2254-2256.

Bowler C, Allen AE, Badger JH, Grimwood J, Jabbari K, Kuo A , et al. (2008). The Phaeodactylum genome reveals the evolutionary history of diatom genomes. Nature 456:239-244.

Bratbak G, Thingstad T (1985). Phytoplankton-bacteria interactions - an apparent paradox? Analysis of a model system with both competition and commensalism. Mar Ecol Prog Ser 25:23-30.

Bruckner CG, Rehm C, Grossart H, Kroth PG (2011). Growth and release of extracellular organic compounds by benthic diatoms depend on interactions with bacteria. Environ Microbiol 13:1052-1063.

Cantarel BL, Coutinho PM, Rancurel C, Bernard T, Lombard V, Henrissat B (2009). The Carbohydrate-Active EnZymes database (CAZy): An expert resource for glycogenomics. Nucleic Acids Res 37:D233-D238.

Cole JJ (1982). Interactions between bacteria and algae in aquatic ecosystems. Annu Rev Ecol Syst 13:291-314.

Croft MT, Warren MJ, Smith AG (2006). Algae need their vitamins. Eukaryot Cell 5:1175-1183.

Croft M, Lawrence A, Raux-Deery E, Warren M, Smith A (2005). Algae acquire vitamin B 12 through a symbiotic relationship with bacteria. Nature 438:90-93.

Debarbouille M, Martinverstraete I, Kunst F, Rapoport G (1991). The Bacillus subtilis sigL g ene encodes an equivalent of σ54 from Gram-negative bacteria. Proc Natl Acad Sci U S A 88:9092-9096.

Droop MR (1970). Vitamin B 12 and marine ecology. V. Continuous culture as an approach to nutritional kinetics. Helgoland Wiss Meer 20:629-636.

Ducklow HW, Steinberg DK, Buesseler KO (2001). Upper ocean carbon export and the biological pump. Oceanography 14:50-58.

Eigemann F, Hilt S, Salka I, Grossart H (2013). Bacterial community composition associated with freshwater algae: Species specificity vs. dependency on environmental conditions and source community. FEMS Microbiol Ecol 83:650-663.

Fandino L, Riemann L, Steward G, Azam F (2005). Population dynamics of Cytophaga -Flavobacteria during marine phytoplankton blooms analyzed by real-time quantitative PCR. Aquat Microb Ecol 40:251-257.

Field C, Behrenfeld M, Randerson J, Falkowski P (1998). Primary production of the biosphere: Integrating terrestrial and oceanic components. Science 281:237-240.

Foster RA, Kuypers MMM, Vagner T, Paerl RW, Musat N, Zehr JP (2011). Nitrogen fixation and transfer in open ocean diatom-cyanobacterial symbioses. ISME J 5:1484-1493.

Gärdes A, Ramaye Y, Grossart H, Passow U, Ullrich MS (2012). Effects of Marinobacter adhaerens HP15 on polymer exudation by Thalassiosira weissflogii at different N:P ratios. Mar Ecol Prog Ser 461:1-14.

Gärdes A, Iversen MH, Grossart H, Passow U, Ullrich MS (2011). Diatom-associated bacteria are required for aggregation of Thalassiosira weissflogii . ISME J 5:436-445.

Gebhard S, Tran SL, Cook GM (2006). The Phn system of Mycobacterium smegmatis : A second high-affinity ABC-transporter for phosphate. Microbiology-(UK) 152:3453-3465.

Grossart H, Levold F, Allgaier M, Simon M, Brinkhoff T (2005). Marine diatom species harbour distinct bacterial communities. Environ Microbiol 7:860-873. 46 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Grossart H, Schlingloff A, Bernhard M, Simon M, Brinkhoff T (2004). Antagonistic activity of bacteria isolated from organic aggregates of the German Wadden Sea. FEMS Microbiol Ecol 47:387-396.

Grossart H (1999). Interactions between marine bacteria and axenic diatoms ( Cylindrotheca fusiformis , Nitzschia laevis , and Thalassiosira weissflogii ) incubated under various conditions in the lab. Aquat Microb Ecol 19:1-11.

Guannel ML, Horner-Devine MC, Rocap G (2011). Bacterial community composition differs with species and toxigenicity of the diatom Pseudo -nitzschia . Aquat Microb Ecol 64:117-133.

Guillard R (1975). Culture of phytoplankton for feeding marine invertebrates. Culture of marine invertebrate animals:108-113.

Hahnke S, Sperling M, Langer T, Wichels A, Gerdts G, Beardsley C , et al. (2013). Distinct seasonal growth patterns of the bacterium Planktotalea frisia in the North Sea and specific interaction with phytoplankton algae. FEMS Microbiol Ecol 86:185-199.

Hecky R, Campbell P, Hendzel L (1993). The stoichiometry of carbon, nitrogen, and phosphorus in particulate matter of lakes and oceans. Limnol Oceanogr 38:709-724.

Kaeppel EC, Gärdes A, Seebah S, Grossart H, Ullrich MS (2012). Marinobacter adhaerens sp. nov., isolated from marine aggregates formed with the diatom Thalassiosira weissflogii . Int J Syst Evol Microbiol 62:124-128.

Kanehisa M, Goto S (2000). KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res 28:27-30.

Lombard V, Ramulu HG, Drula E, Coutinho PM, Henrissat B (2014). The carbohydrate-active enzymes database (CAZy) in 2013. Nucleic Acids Res 42:D490-D495.

Maeda T, Yoshinaga I, Shiba T, Murakami M, Wada A, Ishida Y (2000). Cloning and sequencing of the gene encoding an aldehyde dehydrogenase that is induced by growing Alteromonas sp. strain KE10 in a low concentration of organic nutrients. Appl Environ Microbiol 66:1883-1889.

Mann AJ, Hahnke RL, Huang S, Werner J, Xing P, Barbeyron T , et al. (2013). The genome of the alga- associated marine flavobacterium Formosa agariphila KMM 3901 T reveals a broad potential for degradation of algal polysaccharides. Appl Environ Microbiol 79:6813-6822.

Mayali X, Franks PJS, Burton RS (2011). Temporal attachment dynamics by distinct bacterial taxa during a dinoflagellate bloom. Aquat Microb Ecol 63:111-122.

McBride MJ, Xie G, Martens EC, Lapidus A, Henrissat B, Rhodes RG , et al. (2009). Novel features of the polysaccharide-digesting gliding bacterium Flavobacterium johnsoniae as revealed by genome sequence analysis. Appl Environ Microbiol 75:6864-6875.

Metcalf WW, Wanner BL (1991). Involvement of the Escherichia coli phn (psiD ) gene cluster in assimilation of phosphorus in the form of phosphonates, phosphite, P i esters, and P i. J Bacteriol 173:587-600.

Mobley HL, Hausinger RP (1989). Microbial ureases: Significance, regulation, and molecular characterization. Microbiol Mol Biol R 53:85-108.

Myklestad S, Holmhansen O, Varum KM, Volcani BE (1989). Rate of release of extracellular amino-acids and carbohydrates from the marine diatom Chaetoceros affinis . J Plankton Res 11:763-773.

Myklestad SM (1995). Release of extracellular products by phytoplankton with special emphasis on polysaccharides. Sci Total Environ 165:155-164.

47 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Nishijyo T, Park S, Lu C, Itoh Y, Abdelal A (1998). Molecular characterization and regulation of an operon encoding a system for transport of arginine and ornithine and the ArgR regulatory protein in Pseudomonas aeruginosa . J Bacteriol 180:5559-5566.

Passow U, Shipe R, Murray A, Pak D, Brzezinski M, Alldredge A (2001). The origin of transparent exopolymer particles (TEP) and their role in the sedimentation of particulate matter. Cont Shelf Res 21:327-346.

Passow U (2000). Formation of transparent exopolymer particles, TEP, from dissolved precursor material. Mar Ecol Prog Ser 192:1-11.

Paul C, Mausz MA, Pohnert G (2013). A co-culturing/metabolomics approach to investigate chemically mediated interactions of planktonic organisms reveals influence of bacteria on diatom metabolism. Metabolomics 9:349-359.

Paul C, Pohnert G (2011). Interactions of the algicidal bacterium Kordia algicida with diatoms: Regulated protease excretion for specific algal lysis. PLoS One 6:e21032.

Perkins D, Pappin D, Creasy D, Cottrell J (1999). Probability-based protein identification by searching sequence databases using mass spectrometry data. Electrophoresis 20:3551-3567.

Ribardo DA, Hendrixson DR (2011). Analysis of the LIV system of Campylobacter jejuni reveals alternative roles for LivJ and LivK in commensalism beyond branched-chain amino acid transport. J Bacteriol 193:6233- 6243.

Robinson WG, Coon MJ (1957). The purification and properties of β-hydroxyisobutyric dehydrogenase. J Biol Chem 225:511-521.

Rooney-Varga J, Giewat M, Savin M, Sood S, LeGresley M, Martin J (2005). Links between phytoplankton and bacterial community dynamics in a coastal marine environment. Microb Ecol 49:163-175.

Seymour JR, Ahmed T, Marcos, Stocker R (2008). A microfluidic chemotaxis assay to study microbial behavior in diffusing nutrient patches. Limnol Oceanogr Meth 6:477-488.

Seymour JR, Ahmed T, Stocker R (2009). Bacterial chemotaxis towards the extracellular products of the toxic phytoplankton Heterosigma akashiwo . J Plankton Res 31:1557-1561.

Shevchenko A, Tomas H, Havlis J, Olsen JV, Mann M (2006). In-gel digestion for mass spectrometric characterization of proteins and proteomes. Nat Protoc 1:2856-2860.

Sison-Mangus MP, Jiang S, Tran KN, Kudela RM (2014). Host-specific adaptation governs the interaction of the marine diatom, Pseudo-nitzschia and their microbiota. ISME J 8:63-76.

Sonnenschein EC, Syit DA, Grossart H, Ullrich MS (2012). Chemotaxis of Marinobacter adhaerens and its impact on attachment to the diatom Thalassiosira weissflogii . Appl Environ Microbiol 78:6900-6907.

Sonnenschein EC, Gärdes A, Seebah S, Torres-Monroy I, Grossart H, Ullrich MS (2011). Development of a genetic system for Marinobacter adhaerens HP15 involved in marine aggregate formation by interacting with diatom cells. J Microbiol Methods 87:176-183.

Teeling H, Fuchs BM, Becher D, Klockow C, Gardebrecht A, Bennke CM , et al. (2012). Substrate-controlled succession of marine bacterioplankton populations induced by a phytoplankton bloom. Science 336:608-611.

Vanni P, Giachetti E, Pinzauti G, McFadden BA (1990). Comparative structure, function and regulation of isocitrate lyase, an important assimilatory enzyme. Comp Biochem Physiol 95:431-458.

48 Results/ Part I

Proteomics analysis focusing the interaction of the marine bacterium Marinobacter adhaerens HP15 with the diatom Thalassiosira weissflogii

Wagner-Doebler I, Ballhausen B, Berger M, Brinkhoff T, Buchholz I, Bunk B , et al. (2010). The complete genome sequence of the algal symbiont Dinoroseobacter shibae : A hitchhiker's guide to life in the sea. ISME J 4:61-77.

Wissenbach U, Six S, Bongaerts J, Ternes D, Steinwachs S, Unden G (1995). A 3rd periplasmic transport- system for L-arginine in Escherichia coli - molecular characterization of the artPIQMJ g enes, arginine binding and transport. Mol Microbiol 17:675-686.

50 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

3.2. Stereo-tracking of chemosensing-deficient and motility impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

Abstract

Marine aggregates are nutrient-rich hot spots in the rather oligotrophic marine water column. Thus these aggregates are usually heavily colonized by diverse microorganisms. Microbial chemotaxis comprising of chemosensing and motility is a beneficial feature of microbes utilizing the nutrient pool of marine aggregates. Laboratory-based methods for evaluation of particle colonization and the impact of chemotaxis have improved over the last years. In the present study, we aimed to contribute to broadening the range of methods used for the analysis of marine aggregate colonization. To this end, differentially fluorescence-labeled chemotaxis mutants of the particle-colonizing, marine heterotrophic bacterium Marinobacter adhaerens HP15 were generated. Different variants of fluorescent protein-encoding genes were cloned on plasmids, which were successfully transferred to distinct M. adhaerens HP15 wild type and mutants. The obtained transformants were shown to meet requirements for long term co-incubations with diatoms and signal intensity. Three fluorescent protein variants, eCfp, eYfp, and DsRed, are now available in M. adhaerens HP15 wild type as well as motility-impaired, chemotaxis-defective, and biofilm-negative mutants. These mutants allow stereo-detection of combined pools of mutants and wild type during aggregate colonization experiments, in order to address questions regarding the contribution of individual genes to dynamics and interaction efficiency during particle colonization.

Keywords motility · chemosensing · particle exploitation · fluorescence · ecfp · eyfp · DsRed · marine snow aggregates · bacteria-diatom interaction · microbial loop

51 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

Introduction

The distribution of organic matter in the ocean is assumed to be heterogeneous: Nutrient hot spots in the form of marine snow aggregates, fecal pellets, plumes, sticky polymers, or lysed cells are distributed in a patchy manner (Azam & Long, 2001; Long & Azam, 2001; Blackburn et al. , 1998) thereby generating a ‘microspacial architecture’ (Stocker & Seymour, 2012). This patch-like distribution of nutritional hot spots seems to be an important parameter in shaping biogeochemical cycling in the marine water column which obviously is anything but a homogenous system (Stocker, 2012), as it was observed traditionally. Particles like marine snow aggregates usually harbor higher concentrations of microbes than the ambient waters (Ploug et al. , 1999; Simon et al. , 2002). Heavy microbial colonization turns aggregates into hot spots of metabolic activities (Alldredge & Silver, 1988; Ploug et al. , 1999; Ayo et al. , 2001; Kiørboe, 2001) thereby facilitating direct re-allocation of organic matter to the upper water column (Azam et al. , 1983; Smith et al. , 1992). Thus, the aggregate-associated activities of microorganisms strongly contribute to biomass turnover which impacts element cycling on a global scale (Azam, 1998; Falkowski et al. , 2008). The bacterium’s exploitation of a point source of nutrients demands the capability of chemotaxis combining sensing (chemosensing) of and the directed movement (motility) towards a chemoattractant (Wadhams & Armitage, 2004). Thus chemotaxis allows a bacterium to actively cope with an environment, in which nutrients are distributed in a patchy manner. Once having encountered an aggregate, flagella and pili may govern adhesion on its surface (Dalisay et al. , 2006; Frischkorn et al. , 2013). Usually, motility is combined with the feature of chemosensing (Lux & Shi, 2004). The assumed numbers of motile bacteria in marine systems vary strongly, ranging from less than 10% to 60% of the bacteria observed and further depend on seasonal variations, nutritional status, and the respective oceanic region (Mitchell et al. , 1995; Stretton et al. , 1997; Grossart et al. , 2001). The ability of chemotaxis is assumed to be a trade-off, as on the one hand motility is energy intensive whilst on the other hand the active exploitation of nutrient point sources is a major benefit (Kiørboe & Jackson, 2001; Barbara & Mitchell, 2003; Stocker et al. , 2008). A fast encountering of nutritional hot spots in marine systems is advantageous since such hot spots tend to be of short life time and are easily dispersed within minutes due to diffusion and turbulences (Blackburn et al. , 1998; Kiørboe et al. , 2001).

52 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

In recent years, strategies and devices have been developed that allow the observation of swimming behavior and aggregate exploitation of microorganisms. Information on cell numbers and species of, e.g., bacterioplankton colonizing aggregates is available (Kiørboe et al. , 2003). Mathematical models and use of artificial aggregate systems such as agar spheres (Kiørboe et al. , 2002; Kiørboe et al. , 2003; Tang et al. , 2006) or diatom-derived aggregates (Ploug & Grossart, 2000) have been applied to model the dynamics occurring on aggregates. This involves the analysis of bacterial attachment, growth, detachment, or grazing of and by other microorganisms. More recently, microfluidic devices were developed that allow in- depth observation of swimming speed or tumble behavior of microorganisms in response to attracting or repelling compounds (Seymour et al. , 2008; Stocker et al. , 2008) or particles such as diatoms (Stocker & Seymour, 2012). For investigation of the dynamic processes that lead to marine snow formation, a model system based on bacteria-diatom interaction was established in the past. The marine heterotrophic bacterium Marinobacter adhaerens HP15 was shown to trigger aggregate formation when being co-cultivated with the diatom Thalassiosira weissflogii (Gärdes et al. , 2011). As aggregate formation was stronger with M. adhaerens HP15 than other bacteria, both organisms were chosen for the establishment of a corresponding bilateral system. Since the bacterium is genetically accessible and its genome being fully annotated, mutants of M. adhaerens HP15 were generated that lack features of chemotaxis, as they are deficient in either flagella or pili or in response regulators needed for both pilus- and flagellum-mediated motility (Sonnenschein et al. , 2011; Sonnenschein et al. , 2012). In the present study, we aimed to provide an easily applicable method that allows the stereo- detection of M. adhaerens HP15 wild type with mutants altered in features of chemotactic behavior. To this end, plasmids carrying a variety of fluorescent protein encoding genes were constructed and introduced to M. adhaerens HP15 wild type and its mutants. Parameters of experimental relevance were tested including fluorescence intensity and long-term stability of the fluorescence signal in M. adhaerens HP15 over time.

53 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

Methods

Bacteria, plasmids, and growth conditions Bacterial strains, mutants, and transformants as well as plasmids generated and used in this study are listed in Supplementary 6. M. adhaerens HP15 was cultivated with marine broth (MB) medium at 37°C (Sonnenschein et al. , 2011) unless otherwise stated. The construction and maintenance of plasmids was conducted in Escherichia coli DH5α, routinely grown at 37°C in lysogeny broth (LB) medium. Growth medium of the hemA deficient, auxotropic E. coli ST18 (Thoma & Schobert, 2009) was additionally supplemented with 50 mg L -1 5- aminolevulinic acid (5-ALA). Antibiotics used were 25 mg L -1 chloramphenicol (Cm) and 50 mg L-1 ampicillin (Ap).

DNA manipulations and cloning Used enzymes were purchased from Thermo Fisher Scientific (Schwerte, Germany). Plasmid digests were separated via 1% agarose gel runs. DNA fragments further on ligated into plasmid backbones were purified from agarose gels using a ‘GeneJETGel Extraction Kit’. Plasmid extractions were carried out with a ‘GeneJET Plasmid Miniprep Kit’ (Thermo Fisher Scientific).

Generation of fluorescent protein-encoding plasmids and plasmid conjugation into Marinobacter adhaerens HP15 Plasmid pBBR1MCS-4 (Kovach et al. , 1994) has previously been shown to replicate in M. adhaerens HP15 (Sonnenschein et al. , 2011) and was used as backbone for the aimed fluorescent protein-coding genes ecfp , eyfp , and DsRed. MiniTn7 delivery plasmids as described by Lambertsen et al. (2004) ( miniTn7(Gm) PA1/04/03 ecfp -a, miniTn7(Gm)

PA1/04/03 eyfp -a, miniTn7(Gm) PA1/04/03 DsRedExpress -a, Supplementary 6) served as source for respective fluorescent protein-coding genes. Fragments of ~2 kb were obtained by Not I digestion of plasmids. The fragments contained the respective genes encoding for fluorescent proteins under control of promoter P A1/04/03 and a chloramphenicol resistance cassette ( cat ). As pBBR1MCS-4 harbors two Not I restriction sites, the excised 2-kb fragments were first cloned into the exclusive Not I site of pBluescript resulting in plasmids pBlue-ecfp , pBlue-eyfp , and pBlue-DsRed (Supplementary 6). Next, insert fragments were re-excised at different restriction sites and cloned into the final backbone pBBR1MCS-4. In detail, the ecfp/cat gene 54 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

fragment was re-excised by Pst I/ Cfr 421 ( Sac II) double digest. The resulting fragment of ~2 kb was ligated into Pst I/ Cfr 421 ( Sac II) opened pBBR1MCS-4. eyfp /cat and DsRed /cat gene fragments were excised by Eco RI digestion and correspondingly ligated into Eco RI site of pBBR1MCS-4. Plasmids were routinely transformed into chemically-competent E. coli strains via heat-shock. Plasmids encoding fluorescent proteins (pBBR-4-ecfp, pBBR-4-eyfp , and pBBR-4-DsRed , Supplementary 6) were conjugated into M. adhaerens wild type and mutants by di-parental conjugation using E. coli ST18 (Thoma & Schobert, 2009) . Briefly, M. adhaerens HP15 wild type and mutants were grown for two days on MB agar. E. coli ST18, previously transformed with fluorescent protein encoding plasmids, were grown over night on LB-Cm-5-ALA. Biomass was scraped off agar plates, re-suspended in MB ( M. adhaerens HP15) or LB medium (E. coli ST18), respectively, and the OD 600 was adjusted to 1.0. 200 µL of each cell suspension were combined, spotted on non-selective MB-5-ALA agar, and incubated at 28°C. Cell biomass was harvested after 20 h of incubation, re-suspended, and re-streaked on selective MB-Cm agar. M. adhaerens HP15 transformants appeared within two days of incubation at 37°C. All three plasmids were conjugated into the wild type and five different mutants of HP15. In detail the mutants were deficient in a flagellum (∆fliC and fliG ::Tn5) and a type IV pili mutant (∆mshB ), the histidine kinase CheA needed for chemosensing of flagellum-mediated motility and attachment (∆cheA ) (Foynes et al. , 2000; Tremaroli et al. , 2011), and the histidine kinase ChpA needed for type IV pilus functioning ( chpA ::Tn5, Supplementary 6) (Bertrand et al. , 2010). Successful transformation of plasmids and expression of functional fluorescent proteins were evaluated by fluorescence microscopy in both E. coli and M. adhaerens HP15 at all stages of the transformation and conjugation processes. Unlabeled E. coli and M. adhaerens HP15 cells were used as negative control. Fluorescence emission maxima of transformed bacterial cells were obtained with a Tecan infinite M1000 PRO plate reader (Maennedorf, Switzerland). Emission scans were carried out in a range of 400 to 700 nm.

Testing fluorescence stability and cell survival during starvation under non-selective conditions M. adhaerens HP15 transformants were grown in MB-Ap medium and harvested in exponential growth phase (OD 600 0.5-1.0) via centrifugation (20 min, 3,000 rcf). Cell pellets were washed twice with artificial seawater (ASW) (Seymour et al. , 2008) to remove residual 55 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

8 -1 MB medium. The OD 600 was adjusted to 0.1 (~2 x 10 cells ml ) with ASW without supplementation of nutrient or carbon sources (three biological replicates). Cells were incubated under starvation conditions with constant shaking (250 rpm) at 18°C. Aliquots of cells were further simultaneously incubated under plasmid-selecting (antibiotic-containing) and non-selective (antibiotic free) conditions. This test aimed to determine the possible loss of fluorescence signal due to incubation under non-selective conditions and thus stability of plasmid presence. Fluorescence signals were determined via fluorescence microscopy at time point 0 and after 72 h of incubation. Signals in an area of 3.62 x 10-2 mm² were counted (3-7 technical replicates each). In parallel, the amount of colony forming units was determined at both time points, to test if cell died over the time course of starvation. Fluorescence signal and colony forming units data of time point 0 were set to 100%, data obtained from time point 72 indicated loss or gain in total cell numbers.

Stereo-detection of differentially labeled Marinobacter adhaerens HP15 wild type cells As prove of principle, two differentially labeled wild type transformants ( M. adhaerens HP15 + pBBR-4-eyfp and M. adhaerens HP15 + pBBR-4-DsRed ) were cultivated, washed, and the

OD 600 was adjusted to 0.1 as stated above. In addition, cells were starved for 72 h in ASW (18°C, constant shaking at 250 rpm) in the dark to avoid bleaching of the fluorescence signal. After starvation, equal volumes of both transformants were combined. MB agar spheres were generated by pipette-spotting of melted MB agar onto glass slides (size ~0.1 - 1 mm diameter). 1.5 ml tubes were equipped with a small metal grid in a way to form a second level within the tube, avoiding the sinking of agar spheres onto the ground. 1 ml of combined transformants was added and incubated for 14 h. Colonization of agar spheres was observed by fluorescence microscopy at the end of the incubation period.

56 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

Results

Marinobacter adhaerens HP15 mutants labeled with three different fluorescence plasmids pBBR-4-ecfp , pBBR-4-eyfp , and pBBR-4-DsRed Fluorescence-labeled M. adhaerens HP15 transformants were successfully generated. The labeled bacteria showed satisfactory fluorescence intensity (Figure 1 ). Transformants with any of the three fluorescence plasmids are now available in the wild type, the motility mutants lacking the flagellum or the flagellum hook (HP15_∆fliC ; HP15-fliG ::Tn5), a type IV pilus chemosensory deficient mutant (HP15-chpA ::Tn5) and a flagellum chemosensory mutants (HP15_ ∆cheA ;) as well as a type-IV pilus structural mutant (HP15_ ∆mshB , Supplementary 6). The observed fluorescence emission maxima for each transformant were 476 nm for Ecfp, 528 nm for eYfp, and 586 nm for DsRed, being in line with values stated in literature (Baird et al. , 2000; Patterson et al. , 2001). The fluorescence signal of transformants was sufficiently detected at an exposure time of ~33ms ( Figure 1 ), recommended as an adequate exposure time, at which a signal should be detectable in context of exposure and attachment experiments (S. Smriga, personal communication).

Figure 1: M. adhaerens HP15 labeled with fluorescent protein encoding plasmids. a M. adhaerens HP15 + pBBR-4-ecfp b M. adhaerens HP15 + pBBR-4-eyfp c M. adhaerens HP15 + pBBR-4-DsRed. Exposure time = 32.54 ms.

To initially test the principle of stereo-detection of two fluorescent proteins, a co-cultivation trial of M. adhaerens HP15 wild type labeled with eYfp and DsRed was successfully conducted with MB agar spheres, showing a satisfying detection of the fluorescence signal

57 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

(Figure 2). This test trial further showed, that both transformants attached equally well to the agar surface.

Figure 2: Stereo-detection of M. adhaerens HP15 wild type labeled either with eCfp ( b) or DsRed ( c), attached to an agar sphere. Equal amounts of two differentially labeled M. adhaerens HP15 wild type transformants were starved and further exposed to an MB agar sphere for 14 h. a Overlap of emissions derived from both eYfp and DsRed labeled M. adhaerens HP15 wild type b Detection of eYfp labeled M. adhaerens HP15 c Detection of DsRed labeled M. adhaerens HP15.

The suitability of labeled wild type and mutants for various experimental applications was demonstrated as follows: Differentially labeled transformants of M. adhaerens HP15 wild type, its chemosensing mutant ( ∆cheA ), and its motility mutant ( ∆fliC ) were combined in ratios of 1:1:1 and exposed to T. weissflogii diatom cells (Figure 3 ). The stereo-detection of all three transformants carrying different fluorescence plasmids was successfully shown. A chemotactic behavior was observed for the wild type cells (green (eYfp)) resulting in cell clusters in close vicinity to the diatom cell whilst the motility- or chemotaxis-deficient mutants did not cluster but remained distributed equally in the medium (∆cheA = red (DsRed), and ∆fliC = blue (eCfp)) ( Figure 3 ).

58 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

Figure 3: Exposure of differentially labeled M. adhaerens HP15 wild type and mutants to a T. weissflogii diatom cell . Clustering around the cell is observed for the wild type (green, eYfp), but not for the motility ( ∆cheA ; red, DsRed) and chemosensing deficient mutants ( ∆fliC ; blue, eCfp) of M. adhaerens HP15 (image taken by S. Smriga, Boston, USA).

Fluorescence stability and cell survival over time The survival of Marinobacter cells and the strength of the fluorescence signal under non- selective conditions were tested over a starvation phase of 72 h. The survival of cells was considered important for long-term exposure experiments and particularly when conducted in sea water being low in carbon and nutrient sources. Our data showed that the net number of viable M. adhaerens HP15 cells under starving conditions did not decrease over a period of 72 h ( Table 1). This finding implied that long-term exposure experiments under nutrient limited conditions, i.e. the co-incubation of M. adhaerens HP15 with diatom cells in nutrient- depleted sea water would be possible for at least 72 h without a net loss of viable bacterial cells.

59 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

In addition, the number of cells detected due to their fluorescence signal remained constant under non-selective conditions ( Table 1). This result suggested that long-term fluorescence observations with M. adhaerens HP15 are possible without the use of antibiotics as selective marker, which might interfere with the experimental set-up.

Table 1: Fluorescence and cell recovery after 72 h of starvation under non-selective conditions.

Transformant Cell recovery [%] Fluorescence recovery [%]

0 h 72 h 0 h 72 h

M. adhaerens HP15 + eCfp 100.0 ± 19.2 101.8 ± 4.5 100.0 ± 32.3 109.1 ± 5.8

M. adhaerens HP15 + eYfp 100.0 ± 10.2 102.1 ± 6.5 N/A N/A

M. adhaerens HP15 + DsRed 100.0 ± 8.5 113.8 ± 9.65 100.0 ± 14.2 112.6 ± 16.5

60 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

Discussion

Chemotaxis can be assumed as a particularly advantageous feature in environments such as the marine water column, where nutrient sources are distributed in a patchy manner and tend to be of short life time (Long & Azam, 2001). Chemotaxis comprises the combined features of chemosensing and directed motility. Loss of motility due to lack of flagella might limit the ability of marine bacteria to exploit nutritional hot spots such as marine snow particles or aggregates. Our so-far obtained in vitro data support this assumption. Type IV pili have been further shown to be mandatory for surface adhesion and colonization (Hadi et al. , 2012). As part of a chemosensory system, the histidine kinase CheA is a mandatory element for flagellum-mediated chemotaxis, as shown in Helicobacter pylori 26695 or Pseudomonas pseudoalcaligens KF707 (Tremaroli et al. , 2011). Lack of CheA resulted in a failure of host colonization (Foynes et al. , 2000), not only due to loss of flagellum control but possibly also due to loss of biofilm formation (Foynes et al. , 2000; Tremaroli et al. , 2011). The importance of CheA in attachment to surfaces was also demonstrated for M. adhaerens HP15, where a mutant lacking CheA showed a reduced attachment towards abiotic surfaces (Sonnenschein et al. , 2012). Another histidine kinase is ChpA, yet acting on the functioning of type IV pili. ChpA in M. adhaerens HP15 does not impact chemotactic swimming behavior, however a loss of ChpA resulted in reduced attachment to diatom cells (Sonnenschein et al. , 2012) possibly due to loss of twitching motility, as demonstrated in other studies (Bertrand et al. , 2010). When speculating about chemotaxis as an advantageous feature for bacteria in the marine system, it is worth to mention that certain dominant taxa such as Pelagibacter ubique as a representative of the SAR11 clade are non-motile (Morris et al. , 2002; Giovannoni et al. , 2005). Thus such microorganisms appear to adapt to oligotrophic environments with alternative strategies and inhabit different niches. The present study offers another tool to quantify and define features of aggregate sensing and colonization by simultaneous detection of differential fluorescence signals, thus being able to observe different M. adhaerens HP15 mutants and their wild type. M. adhaerens HP15 was shown to develop chemotactic behavior towards the supernatant of diatom cultures (Sonnenschein et al. , 2012). It further induced aggregate formation during interaction with the diatom T. weissflogii and triggered an increased formation of transparent exopolymeric particles (Gärdes et al. , 2011), known to serve as a glue during marine aggregate formation (Passow, 2002). Thus this bacterium is a suitable and promising candidate for improving our 61 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

understanding of aggregate colonization dynamics. A further major advantage of this model bacterium is found in its genetic accessibility that already led to the generation of several gene-specific mutants lacking important factors responsible for chemotaxis, as outlined above. With the tools generated in the current study, it is now possible to track and observe three different phenotypes simultaneously due to the ability to differentiate between mutants in a mixture. Parameters other than the mutated ones are assumed to remain unaltered, as there is cell size or features of the inner or outer membrane. The herein applied bi-parental conjugation method using E. coli ST18 had previously been successfully applied in other bacteria such as Dinoroseobacter shibae (Endres et al. , 2015), further members of the Roseobacter clade (Piekarski et al. , 2009), or Pseudomonadaceae (Pletzer et al. , 2014). Plasmid pBBR1MCS is commonly used in cloning applications and applicable in a wide range of bacteria, as it is the case for the used lacZ promoter PA1/04/03 . Thus, herein generated plasmids may be applied for the labeling of other bacteria that have relevance in marine aggregate colonization, as there are certain members of α-, γ-, or Flavobacteria (Grossart et al. , 2004). The labeling will allow an easy-to-detect tracking of a simplified heterogeneous mixture of bacteria with so far three different species or mutant variants. This will form an advancement over previous studies which dealt with aggregate colonization and considered the analysis of only one bacterial strain (Yam & Tang, 2007) or natural sea water assemblages (Grossart et al. , 2006) being heterogeneous in their taxonomical composition. With the application of three different variants that are easily detectable, a more complex system may be set up for the understanding of dynamics appearing between species on an aggregate. Such a system would furthermore offer the investigation of bacterial inter-species competition that appears in biofilms of marine aggregate assemblages. A further option resulting from fluorescence labeling is the possibility to separate different mutants after exposure to an aggregate by fluorescence-activated cell sorting (Sekar et al. , 2004). Thus sorted cells could for example be applied to proteomics approaches (Jehmlich et al. , 2010). By use of different -omics techniques, a view on the niches taken into account by different bacteria and an eventual niches shift due to competition could be investigated in detail. With the tool of differentially labeled motility-deficient and chemosensing-lacking mutants in combination with high quality microscopy, it is now possible to stereo-monitor both wild type and mutants within one single experimental set-up. The transformants may be used for exposure experiments towards artificial particles such as agar spheres, natural marine snow 62 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

particles, or microalgae. Colonization rates and the impact of chemosensing and motility on colonization efficiency may be quantified thus giving us more detailed insights into the nature and mechanisms of marine aggregate formation.

63 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

Acknowledgement

The authors would like to thank Steven Smriga and Roman Stocker for useful discussions and especially Steven Smriga for great effort on pushing experimental progress. This project was funded by Jacobs University Bremen, Helmholtz Graduate School for Polar and Marine Research (POLMAR) and Deutsche Forschungsgemeinschaft (UL 169/6-1).

64 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

References

Alldredge A, Silver M (1988). Characteristics, dynamics and significance of marine snow. Prog Oceanogr 20:41- 82.

Ayo B, Unanue M, Azua I, Gorsky G, Turley C, Iriberri J (2001). Kinetics of glucose and amino acid uptake by attached and free-living marine bacteria in oligotrophic waters. Mar Biol 138:1071-1076.

Azam F, Long R (2001). Oceanography - Sea snow microcosms. Nature 414:495-498.

Azam F (1998). Microbial control of oceanic carbon flux: The plot thickens. Science 280:694-696.

Azam F, Fenchel T, Field J, Gray J, Meyer-Reil L, Thingstad F (1983). The ecological role of water-column microbes in the sea. Mar Ecol Prog Ser 10:257-263.

Baird G, Zacharias D, Tsien R (2000). Biochemistry, mutagenesis, and oligomerization of DsRed, a red fluorescent protein from coral. Proc Natl Acad Sci U S A 97:11984-11989.

Barbara G, Mitchell J (2003). Marine bacterial organisation around point-like sources of amino acids. FEMS Microbiol Ecol 43:99-109.

Bertrand JJ, West JT, Engel JN (2010). Genetic analysis of the regulation of type IV pilus function by the Chp chemosensory system of Pseudomonas aeruginosa . J Bacteriol 192:994-1010.

Blackburn N, Fenchel T, Mitchell J (1998). Microscale nutrient patches in planktonic habitats shown by chemotactic bacteria. Science 282:2254-2256.

Dalisay DS, Webb JS, Scheffel A, Svenson C, James S, Holmstroem C , et al. (2006). A mannose-sensitive haemagglutinin (MSHA)-like pilus promotes attachment of Pseudoalteromonas tunicata cells to the surface of the green alga Ulva australis . Microbiology-(UK) 152:2875-2883.

Endres S, Granzin J, Circolone F, Stadler A, Krauss U, Drepper T , et al. (2015). Structure and function of a short LOV protein from the marine phototrophic bacterium Dinoroseobacter shibae . Bmc Microbiology 15:30.

Falkowski PG, Fenchel T, Delong EF (2008). The microbial engines that drive Earth's biogeochemical cycles. Science 320:1034-1039.

Foynes S, Dorrell N, Ward S, Stabler R, McColm A, Rycroft A , et al. (2000). Helicobacter pylori possesses two CheY response regulators and a histidine kinase sensor, CheA, which are essential for chemotaxis and colonization of the gastric mucosa. Infect Immun 68:2016-2023.

Frischkorn KR, Stojanovski A, Paranjpye R (2013). Vibrio parahaemolyticus type IV pili mediate interactions with diatom-derived chitin and point to an unexplored mechanism of environmental persistence. Environ Microbiol 15:1416-1427.

Gärdes A, Iversen MH, Grossart H, Passow U, Ullrich MS (2011). Diatom-associated bacteria are required for aggregation of Thalassiosira weissflogii . ISME J 5:436-445.

Giovannoni S, Tripp H, Givan S, Podar M, Vergin K, Baptista D , et al. (2005). Genome streamlining in a cosmopolitan oceanic bacterium. Science 309:1242-1245.

Grossart H, Czub G, Simon M (2006). Algae-bacteria interactions and their effects on aggregation and organic matter flux in the sea. Environ Microbiol 8:1074-1084.

65 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

Grossart H, Schlingloff A, Bernhard M, Simon M, Brinkhoff T (2004). Antagonistic activity of bacteria isolated from organic aggregates of the German Wadden Sea. FEMS Microbiol Ecol 47:387-396.

Grossart H, Riemann L, Azam F (2001). Bacterial motility in the sea and its ecological implications. Aquat Microb Ecol 25:247-258.

Hadi N, Yang Q, Barnett TC, Tabei SMB, Kirov SM, Shaw JG (2012). Bundle-forming pilus locus of Aeromonas veronii bv. Sobria. Infect Immun 80:1351-1360.

Jehmlich N, Huebschmann T, Salazar MG, Voelker U, Benndorf D, Mueller S , et al. (2010). Advanced tool for characterization of microbial cultures by combining cytomics and proteomics. Appl Microbiol Biotechnol 88:575-584.

Kiørboe T, Tang K, Grossart H, Ploug H (2003). Dynamics of microbial communities on marine snow aggregates: Colonization, growth, detachment, and grazing mortality of attached bacteria. Appl Environ Microbiol 69:3036-3047.

Kiørboe T, Grossart H, Ploug H, Tang K (2002). Mechanisms and rates of bacterial colonization of sinking aggregates. Appl Environ Microbiol 68:3996-4006.

Kiørboe T (2001). Formation and fate of marine snow: Small-scale processes with large-scale implications. Sci Mar 65:57-71.

Kiørboe T, Jackson G (2001). Marine snow, organic solute plumes, and optimal chemosensory behavior of bacteria. Limnol Oceanogr 46:1309-1318.

Kiørboe T, Ploug H, Thygesen U (2001). Fluid motion and solute distribution around sinking aggregates. I. Small-scale fluxes and heterogeneity of nutrients in the pelagic environment. Mar Ecol Prog Ser 211:1-13.

Kovach M, Phillips R, Elzer P, Roop R, Peterson K (1994). pBBR1MCS - a broad-host-range cloning vector. BioTechniques 16:800-802.

Long R, Azam F (2001). Microscale patchiness of bacterioplankton assemblage richness in seawater. Aquat Microb Ecol 26:103-113.

Lux R, Shi W (2004). Chemotaxis-guided movements in bacteria. Crit Rev Oral Biol Med 15:207-220.

Mitchell J, Pearson L, Bonazinga A, Dillon S, Khouri H, Paxinos R (1995). Long lag times and high velocities in the motility of natural assemblages of marine bacteria. Appl Environ Microbiol 61:877-882.

Morris R, Rappe M, Connon S, Vergin K, Siebold W, Carlson C , et al. (2002). SAR11 clade dominates ocean surface bacterioplankton communities. Nature 420:806-810.

Passow U (2002). Production of transparent exopolymer particles (TEP) by phyto- and bacterioplankton. Mar Ecol Prog Ser 236:1-12.

Patterson G, Day R, Piston D (2001). Fluorescent protein spectra. J Cell Sci 114:837-838.

Piekarski T, Buchholz I, Drepper T, Schobert M, Wagner-Doebler I, Tielen P , et al. (2009). Genetic tools for the investigation of Roseobacter clade bacteria. BMC Microbiol 9:265.

Pletzer D, Lafon C, Braun Y, Koehler T, Page MGP, Mourez M , et al. (2014). High-throughput screening of dipeptide utilization mediated by the ABC transporter DppBCDF and its substrate-binding proteins DppA1-A5 in Pseudomonas aeruginosa . Plos One 9:e111311.

66 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

Ploug H, Grossart H (2000). Bacterial growth and grazing on diatom aggregates: Respiratory carbon turnover as a function of aggregate size and sinking velocity. Limnol Oceanogr 45:1467-1475.

Ploug H, Grossart H, Azam F, Jorgensen BB (1999). Photosynthesis, respiration, and carbon turnover in sinking marine snow from surface waters of Southern California Bight: Implications for the carbon cycle in the ocean. Mar Ecol Prog Ser 179:1-11.

Sambrook J, Russel DW (2001). Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory: New York, USA.

Seebah S (2012) Ph.D. thesis. Jacobs University Bremen, Bremen, Germany

Sekar R, Fuchs B, Amann R, Pernthaler J (2004). Flow sorting of marine bacterioplankton after fluorescence in situ hybridization. Appl Environ Microbiol 70:6210-6219.

Seymour JR, Ahmed T, Marcos, Stocker R (2008). A microfluidic chemotaxis assay to study microbial behavior in diffusing nutrient patches. Limnol Oceanogr Meth 6:477-488.

Simon M, Grossart H, Schweitzer B, Ploug H (2002). Microbial ecology of organic aggregates in aquatic ecosystems. Aquat Microb Ecol 28:175-211.

Smith D, Simon M, Alldredge A, Azam F (1992). Intense hydrolytic enzyme-activity on marine aggregates and implications for rapid particle dissolution. Nature 359:139-142.

Stocker R (2012). Marine microbes see a sea of gradients. Science 338:628-633.

Stocker R, Seymour JR (2012). Ecology and physics of bacterial chemotaxis in the ocean. Microbiol Mol Biol Rev 76:792-812.

Stocker R, Seymour JR, Samadani A, Hunt DE, Polz MF (2008). Rapid chemotactic response enables marine bacteria to exploit ephemeral microscale nutrient patches. Proc Natl Acad Sci U S A 105:4209-4214.

Stretton S, Danon S, Kjelleberg S, Goodman A (1997). Changes in cell morphology and motility in the marine Vibrio sp. strain S14 during conditions of starvation and recovery. FEMS Microbiol Lett 146:23-29.

Tang KW, Grossart H, Yam EM, Jackson GA, Ducklowl HW, Kiørboe T (2006). Mesocosm study of particle dynamics and control of particle-associated bacteria by flagellate grazing. Mar Ecol Prog Ser 325:15-27.

Thoma S, Schobert M (2009). An improved Escherichia coli donor strain for diparental mating. FEMS Microbiol Lett 294:127-132.

Tremaroli V, Fedi S, Tamburini S, Viti C, Tatti E, Ceri H , et al. (2011). A histidine-kinase cheA gene of Pseudomonas pseudoalcaligens KF707 not only has a key role in chemotaxis but also affects biofilm formation and cell metabolism. Biofouling 27:33-46.

Wadhams G, Armitage J (2004). Making sense of it all: Bacterial chemotaxis. Nat Rev Mol Cell Biol 5:1024- 1037.

67 Results/ Part II

Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach

Yam EM, Tang KW (2007). Effects of starvation on aggregate colonization and motility of marine bacteria. Aquat Microb Ecol 48:207-215.

68 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

3.3. Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

Abstract

Several members of the ubiquitously found γ-proteobacterial genus Marinobacter were described or assumed to inhabit marine environments naturally enriched in heavy metals. However, direct studies that analyze the ability of this genus to occupy such environments have not been conducted. To cope with heavy metal stress, bacteria possess specific efflux pumps as tools for detoxification, among which the CzcCBA type efflux system is one representative. Previous studies showed that this system plays an important role in resistance towards cadmium, zinc, and cobalt. Up to now, no study had focused on characterization of Czc pumps in Marinobacter sp. or other marine prokaryotes. Herein, we elucidated the function of two CzcCBA pumps encoded by the genome of Marinobacter adhaerens HP15 during exposure to cadmium, zinc, and cobalt. Single and double knock-out mutants lacking the corresponding two czcCBA operons were generated and analyzed in terms of their resistance profiles. Both operons appeared to be important for zinc resistance but had no role in tolerance towards cadmium or cobalt. One of the mutations was genetically complemented, thereby restoring the wild type phenotype. In accordance with the resistance pattern, expression of the genes coding for both CzcCBA pumps was induced by zinc but neither by cadmium nor cobalt.

Keywords

CzcA · multidrug efflux pump · Cupriavidus metallidurans CH34 · zinc · heavy metal resistance · czc operon

69 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

Introduction

The genus Marinobacter comprises a heterogeneous group of more than 40 isolated γ- proteobacterial species that are ubiquitously found in a wide range of marine habitats (Gauthier et al. , 1992; Zhang et al. , 2008; Aguilera et al. , 2009; Handley et al. , 2009; Gärdes et al. , 2010). Handley and Lloyd (2013) postulated a specification in habitats occupied by Marinobacter species. These authors observed that environments these bacterial species were isolated from are habitats enriched in metals or metalloids such as arsenic, manganese, or iron (Handley & Lloyd, 2013). In support of this, some Marinobacter species showed metabolic features favoring colonization of habitats characterized by the presence of certain heavy metals. The potential to oxidize iron II was shown in M. hydrocarbonoclasticus VT8 (former M. aquaeolei ) at the genomic level (Singer et al. , 2011). A nitrate-dependent oxidation of arsenate was observed in M. santoriniensis NKSG1 (Handley et al. , 2009). Wang et al. (2012) isolated M. manganoxydans MnI7-9 from a deep sea vent habitat enriched in heavy metals. They furthermore showed that the genome of this organism encoded for proteins needed for the oxidation of manganese (II) and for proteins required for living in heavy metal-enriched environments (Wang et al. , 2012). A very close relative of M. manganoxydans MnI7-9 is M. adhaerens HP15, which was isolated from marine particles sampled in the German Wadden Sea (Grossart et al. , 2004; Kaeppel et al. , 2012). This organism has served as a model organism for studying bacteria- diatom interactions (Gärdes et al. , 2011; Gärdes et al. , 2012; Sonnenschein et al. , 2012). Being genetically accessible, the strain represents a useful tool for functional analysis of specific genes (Gärdes et al. , 2010; Sonnenschein et al. , 2011). The genome of M. adhaerens HP15 harbors two adjacently located operons both coding for a CzcCBA heavy metal efflux pump. CzcCBA pumps contain the inner membrane efflux pump CzcA that belongs to the family of resistance-nodulation-cell division (RND) transporters (Tseng et al. , 1999). In complex with the membrane fusion protein CzcB and the outer membrane factor (OMF) CzcC, the CzcCBA pump system represents one of three major efflux mechanisms employed by bacteria to survive heavy metal stress (Nies, 2003). CzcCBA facilitates the efflux of certain heavy metal ions via the mechanism of a cation-proton antiporter (Nies, 1995; Goldberg et al. , 1999). Although all three proteins are mandatory for full metal resistance, CzcA was shown to be the most crucial element (Nies et al. , 1989; Legatzki et al. , 2003a; Scherer & Nies, 2009). The complex was first described in Cupriavidus metallidurans CH34

70 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

(former Wautersia metallidurans (Vandamme & Coenye, 2004) , Ralstonia eutropha (Vaneechoutte et al. , 2004), and Alcaligenes eutrophus (Goris et al. , 2001)), which has served as a model system for investigating heavy metal resistance in prokaryotes. In this model organism, CzcCBA was found to facilitate the efflux of divalent cadmium, zinc, and cobalt during cell detoxification (Nies et al. , 1989). Mutants of C. metallidurans CH34 lacking CzcA revealed a drastic decrease of resistance against all three metal ions whereas lack of CzcB and particularly CzcC merely lowered the resistance. A central role of CzcA in metal resistance was also shown in other bacteria such as Gluconacetobacter diazotrophicus PAI 5 (Intorne et al. , 2012), Pseudomonas aeruginosa PT5 (Caille et al. , 2007), or P. putida KT2440 (Leedjarv et al. , 2008). In the present study, we generated czcCBA -deficient mutants of M. adhaerens HP15 lacking either one or both of the two CzcCBA pumps in order to characterize their role in heavy metal resistance. Furthermore, the expression of the two czcCBA operons in dependence of heavy metals was studied by quantitative reverse-transcription polymerase chain reaction (qRT- PCR). Our results thus contribute to improving the understanding of CzcCBA functioning in the genus Marinobacter . Additionally, an initial identification and comparison of CzcCBA- encoding operons in other Marinobacter species was conducted regardless of whether their habitats are naturally enriched in metals or not.

71 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

Methods

Bacterial strains, plasmids, and growth conditions Bacterial strains and plasmids used in this study are summarized in supplementary materials (Supplementary 7). M. adhaerens HP15 was cultured at 28°C in liquid or on solid marine broth (MB) medium (Sonnenschein et al. , 2011). When needed, the MB medium was additionally supplemented with 5% (v/v) sucrose (5%-Suc) (Hoang et al. , 1998). The construction and maintenance of plasmids was conducted in Escherichia coli DH5 α, routinely grown at 37°C in lysogeny broth (LB) medium. Growth medium of the hemA - deficient, auxotrophic E. coli strain ST18 (Thoma & Schobert, 2009) was additionally supplemented with 50 mg L -1 5-aminolevulinic acid (5-ALA). Strains containing the pFLP2 plasmid were cultured at 30°C. Following antibiotics concentrations were used: 25 mg l -1 chloramphenicol (Cm) and 50 mg l -1 ampicillin (Ap).

DNA manipulations, cloning, and transformation techniques All used enzymes and polymerases were purchased from Thermo Fisher Scientific (Schwerte, Germany). PCR screenings for gene deletions were conducted with DreamTaq Polymerase. High Fidelity Phusion DNA Polymerase was applied for amplifications from genomic DNA. Ligations were carried out with T4 DNA Ligase at room temperature. PCR products were separated on 1% agarose gels. DNA fragments were purified from agarose gels using a ‘GeneJET Gel Extraction Kit’. Plasmids were extracted with a ‘GeneJET Plasmid Miniprep Kit’ (Thermo Fisher Scientific).

Design of czcCBA knock-out plasmids and generation of M. adhaerens HP15 deletion mutants Within the gene locus of the operon czcCBA .1, gene HP15_108 codes for protein CzcC (ADP95872.1), HP15_109 codes for CzcB (ADP95873.1), and HP15_110 codes for CzcA (ADP95874.1). The second operon czcCBA .2 contains gene HP15_113 coding for CzcC (ADP95877.1), HP15_114 coding for CzcB (ADP95878.1), and HP15_115 coding for CzcA (ADP95879.1). The complete genome sequence of M. adhaerens HP15 (including the sequences of two extra-chromosomal replicons) is available on the National Center of Biotechnology Information (NCBI) webpage (CP001978.1 (chromosome), CP001979.1 (plasmid pHP-42), and CP001980.1 (plasmid pHP-187)) (Gärdes et al. , 2010).

72 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

In order to delete each of the two operons the following knock-out plasmids were made: Within the czcCBA .1 operon, two regions were PCR-amplified using primer pairs czcCBA.1_A1/ czcCBA.1_A2 (658 bp) and czcCBA.1_B1/ czcCBA.1_B2 (677 bp) (Supplementary 8), thereby generating the DNA regions for homologous recombination. To conduct fusion PCR according to Zumaquero et al. (2010) primers czcCBA.1_A2 and czcCBA.1_B1 were designed to carry a 20 bp homologous overlap at their 5’ end, derived from T7 primer sequences (Zumaquero et al. , 2010). The homologous overlaps contain a Kpn I restriction site for later insertion of a Cm resistance (Cm R) cassette. For fusion PCR, 20 ng of each amplicon were mixed and again PCR amplified using primers czcCBA.1_A1 and czcCBA.1_B2, resulting in a joint fragment of 1.3 kb length. The fusion product was gel purified and ligated into pJET1.2/ blunt cloning vector. The knock-out plasmid for deletion of the czcCBA. 2 operon was obtained by a similar approach. Briefly, regions for homologous recombination were amplified with primer pair czcCBA.2_A1/ czcCBA.2_A2 (694 bp) and czcCBA.2_B1/ czcCBA.2_B2 (612 bp), combined, amplified via fusion PCR, and ligated into pJET1.2. Both inserts were sequenced for verification. Next, a Cm R cassette was removed from plasmid pFCM1 via Kpn I digestion and ligated into the corresponding restriction site of the fused regions. This resulted in plasmids pJET.czcCBA.1_ko and pJET.czcCBA.2_ko. The mutagenesis inserts were further re-excised and ligated into pEX18Ap, resulting in pEX18Ap.czcCBA.1_ko and pEX18Ap.czcCBA.2_ko, respectively. Both knock-out plasmids were introduced to M. adhaerens HP15 wild type cells via bi-parental conjugation using E. coli ST18 (Thoma & Schobert, 2009). First, knock-out plasmids were introduced to electro- competent E. coli ST18 cells via electroporation (Sambrook & Russel, 2001). Next, M. adhaerens HP15 wild type was grown for 48 h on MB and E. coli ST18 carrying the respective knock-out plasmid was grown on LB-5-ALA-Cm overnight. Biomass was scraped off, re-suspended in corresponding medium, and adjusted to an OD 600 of 1.0. 200 µl of each suspension were combined, vigorously mixed, and spotted on MB-5-ALA plates. Plates were incubated for 20 h at 28°C. Biomass was scraped off afterwards, re-suspended, and streaked on selective MB-Cm-5%-Suc plates in order to select for mutagenesis and double crossover events (Hoang et al. , 1998). Homologous recombination and insertion of the Cm R cassette was verified using a set of primer combinations, binding both inside of the cat gene (cat_2- cat_5) and outside of the deleted operon, both up- and down-stream of mutagenesis (primer pairs czcCBA.1_out1-4; czcCBA.2_out1-4, Supplementary 8). For each region two different primer combinations were used. Verification was done via colony PCR. Briefly, one colony

73 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

was re-suspended in 100 µl dd H2O, incubation at 95°C for 5 min under vigorous shaking for DNA release, and pelleted by centrifugation (5 min, 2,300 rcf). The supernatant was directly used for PCR. Amplicons obtained from potential mutants were sequenced for verification of double cross-over events. Cm R cassette in all M. adhaerens HP15 czcCBA deficient mutants was removed at its FRT sites via the site-specific Flp-recombinase (Hoang et al. , 1998). For this, plasmid pFLP2 carrying the FLP recombinase was conjugated into M. adhaerens HP15 mutants. Potential transformants were plated on MB-Ap plates and incubated for 48 h at 28°C. Colonies were re- plated on both MB and MB-Cm plates to screen for Cm-sensitive colonies. Removal of the Cm R cassette in Cm-sensitive colonies was confirmed via PCR using primer pair czcCBA.1_out1/ czcCBA.1_out2 and primer pair czcCBA.2_out1/ czcCBA.2_out2, respectively ( Supplementary 8). The double mutant HP15. ∆czcCBA.1/2 was obtained by repeating mutagenesis with knock-out plasmid pEX18Ap.czcCBA.2_ko in the Cm-sensitive mutant M. adhaerens HP15. ∆czcCBA.1.

Design of czcCBA complementation plasmids pBBR.czcCBA.1 and pBBR.czcCBA.2 Both czcCBA operons were amplified from genomic DNA using primer sets czcCBA .1_for_KpnI/ czcCBA .1_rev_BamHI (5456 bp) and czcCBA.2_for_BamHI/ czcCBA.2_rev_XbaI (5705 bp). Amplified products were cloned in pJET1.2 yielding pJET.czcCBA.1 and pJET.czcCBA.2, respectively, and subsequently sequenced using primers czcCBA.1_seq1-5 and czcCBA.2_seq1-5 ( Supplementary 8). Operons were excised via Kpn I/ Bam HI for pJET.czcCBA.1 and Bam HI/ Xba I for pJET.czcCBA.2 and ligated into corresponding insertions sites of pBBR1MCS-1, yielding plasmids pBBR.czcCBA.1 and pBBR.czcCBA.2 (Supplementary 7).

Minimal inhibitory concentration (MIC) determination of heavy metals for M. adhaerens HP15 Determination of the MIC of heavy metal ions for M. adhaerens HP15 wild type and its mutants was conducted with a 2-fold dilution assay in 96-well plates. The following heavy metal salts were tested: silver nitrate (AgNO 3) (AppliChem, Darmstadt, Germany), cadmium acetate (Cd (CH 3CO 2)2 · 2 H2O) (Sigma Aldrich, Seelze, Germany), cobalt chloride (CoCl 2 ·

H2O) (Sigma Aldrich), cupric sulfate (CuSO 4 · 5 H 2O) (Sigma Aldrich), manganese (II)- sulfate (MnSO 4 · H 2O) (Carl Roth, Karlsruhe, Germany), nickel chloride (NiCl · 6 H 2O)

74 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

(AppliChem), and zinc sulfate (ZnSO 4 · 7 H 2O) (Sigma Aldrich). Tests were done in triplicates. Cells were harvested from exponential growth phase (OD 600 = 0.5-1.0) and the 6 -1 initial OD 600 was adjusted to 0.001 (~2.85 x 10 cells ml ). Highest metal salt concentration used was 0.1 M. Bacteria were incubated at 28°C for 96 h. The MIC was defined as the lowest concentration of a heavy metal salt at which no visible bacterial growth was occurring. For MIC tests on solid agar, M. adhaerens HP15 wild type and mutants were grown and harvested as stated above. Cells were re-suspended in MB medium, the cell density was adjusted to an OD 600 of 0.001. MB agar plates were supplemented with different concentrations of Cd (CH 3CO 2)2 · 2 H 2O, (0.25, 0.26, 0.28, 0.29, and 0.30 mM), ZnSO 4 · 7

H2O (0.40, 0.60, 0.80, 1.00, and 1.20 mM), and CoCl 2 · 6 H 2O (0.65, 0.74, 0.83, 0.91, and 1.00 mM). 10 µl of bacterial suspension were spotted on the agar and left for drying. Growth was analyzed after 96 h of incubation at 28°C.

Determination of czcCBA expression using qRT-PCR

M. adhaerens HP15 cells were grown in MB medium supplemented with Cd (CH 3CO 2)2 · 2

H2O, (0.28 2mM), ZnSO 4 · 7 H 2O (0.30 mM), or CoCl 2 · 6 H 2O (0.28 mM) at 28°C under constant shaking at 250 rpm. 25 ml of culture were sampled in early exponential growth phase

(OD 600 of 0.3) and combined with 15 ml chilled killing buffer (20 mM Tris-HCl, pH 7.5; 20 mM NaN 3 (Schenk et al. , 2006)). Cells were harvested by centrifugation (20 min, 3,000 rcf) and stored at -80°C for further use. RNA was extracted with a ‘GeneJET RNA Purification Kit’ (Thermo Fisher Scientific). DNA contaminations were removed with a DNA free TM -Kit (Ambion Life Technologies, Carlsbad, USA) as recommended by the manufacturer. Absence of DNA contamination was further verified by samples containing no reverse transcriptase during qRT-PCR. The purity of extracted RNA was evaluated spectrophotometrically and confirmed by gel electrophoresis. The following three house-keeping genes were tested for constitutive expression during heavy metal treatment for data normalization: 23 S rRNA (HP15_rRNA8), gyrA (HP15_2519), and recA (HP15_1048) using corresponding primer pairs (Supplementary 9). The final dataset was normalized with gyrA expression data. One-step qRT-PCR was carried out with a QuantiTect ® SYBR ® Green RT-PCR Kit (Qiagen, Venlo, Netherlands). Deviating from to the manufacturer’s recommendation the reaction volume was 25 µl instead of 50 µl. Amplification was conducted via a Mastercycler ep realplex ² gradient S (Eppendorf, Hamburg, Germany) (initial 50°C, 30 min for cDNA synthesis; 95°C, 15 min for activation of HotStarTaq DNA Polymerase; 40 x [95°C, 30 s; 59°C, 30s; 72°C, 30 s]).

75 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

Specificity of amplification was assessed by analyzing the melting curve of the amplification product. Obtained data was quantified by comparative C t approaches as described earlier (Pletzer & Weingart, 2014).

Applied sequence alignment tools and analysis of genomes Nucleotide and protein sequences of both M. adhaerens HP15 and all other bacterial species were obtained from NCBI with the exception of 16S rRNA sequences, obtained from the SILVA database webpage (Pruesse et al. , 2007). Similarity searches were done with the Basic Local Alignment Search Tool (BLAST) as provided by NCBI. Gene or protein sequences were searched against the nucleotide collection (nr/ nt) and the whole genome shotgun contigs (wgs) databases to take both full genome and shot-gun genome sequences into account. The InterPro webpage was used for retrieval of conserved protein domains and functional sites (Jones et al. , 2014).

76 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

Results

Phenotypic characterization of M. adhaerens HP15 mutants lacking CzcCBA efflux pumps Two single mutants lacking either the operon czcCBA. 1 (HP15. ∆czcCBA.1) or czcCBA. 2 (HP15. ∆czcCBA.2) and a double-mutant lacking both operons (HP15. ∆czcCBA.1/2) (Supplementary 7) were constructed and genetically verified. First, the heavy metal tolerance of M. adhaerens HP15 wild type and its CzcCBA-deficient mutants was tested in liquid medium with a 2-fold dilution approach. Although seven different heavy metal salts were tested, none of the mutants differed in MIC values in comparison to the wild type ( Table 1).

Table 1: Minimal inhibitory concentrations of metals for M. adhaerens HP15 wild type and its CzcCBA pump- deficient mutants determined in liquid culture. The MIC determination was conducted with a 2-fold dilution assay. Standard deviations were calculated from at least 3 replicates.

MIC [mM] of M. adhaerens HP15

Metal salt wild type ∆czcCBA.1 ∆czcCBA.2 ∆czcCBA.1/2

AgNO 3 0.020 ± 0.01 0.016 ± 0.01 0.016 ± 0.01 0.024 ± 0.00

Cd (CH 3CO 2)2 · 2 H 2O 0.651 ± 0.23 0.521 ± 0.23 0.521 ± 0.23 0.521 ± 0.23

CoCl 2 · 6 H 2O 3.125 ± 0.00 2.604 ± 0.90 3.125 ± 0.00 3.125 ± 0.00

CuSO 4 · 5 H 2O 6.250 ± 0.00 6.250 ± 0.00 6.250 ± 0.00 6.250 ± 0.00

MnSO 4 · H2O 20.833 ± 7.22 33.33 ± 14.43 16.67 ± 7.22 25.00 ± 0.00

NiCl · 6 H 2O 6.250 ± 0.00 6.250 ± 0.00 6.250 ± 0.00 6.250 ± 0.00

ZnSO 4 · 7 H 2O 3.125 ± 0.00 3.125 ± 0.00 2.604 ± 0.90 2.083 ± 0.90

Next, MIC tests for cadmium, zinc, and cobalt were repeated on solid agar medium as done by others previously (Abou-Shanab et al. , 2007; Intorne et al. , 2012). In case of cadmium- or cobalt-supplemented medium, none of the mutants showed any growth inhibition different from that of the wild type ( Figure 1). When zinc-supplemented agar plates where used, there was no difference in the growth behavior of the wild type and its single mutant HP15. ∆czcCBA.1. However, single mutant HP15. ∆czcCBA.2 appeared to be more sensitive towards zinc as compared to the latter two ( Figure 1). The double mutant HP15. ∆czcCBA.1/2 showed a 2-fold lower MIC as compared to the wild type and was therefore considered to be more sensitive to zinc than single mutant HP15. ∆czcCBA.2. While the absolute MIC values observed showed slight variations between experimental repeats, the general zinc resistance

77 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

patterns were as follows: wild type and single mutant HP15. ∆czcCBA.1 showed indistinguishable resistance patterns while single mutant HP15. ∆czcCBA.2 was more sensitive towards zinc as compared to the wild type. Double mutant HP15. ∆czcCBA.1/2 was the most sensitive mutant. These results suggested that presence of operon czcCBA.1 was more important for zinc tolerance than that of operon czcCBA.2 but that deletion of both operons led to additive effects in terms of loss of zinc resistance.

Figure 1: Minimal inhibitory concentration determination of M. adhaerens HP15 wild type and mutants on cadmium-, zinc-, or cobalt-supplemented agar plates. The lowest heavy metal concentration completely inhibiting bacterial growth is framed. 10 µl of bacterial suspension (OD 600 = 0.001) were spotted on agar plates and left for drying. One side of a square indicates about 10 mm width. WT = wild type.

Expression analysis of czcCBA operons in M. adhaerens HP15 Induced expression of either efflux pump was analyzed by qRT-PCR. The selected primers specifically amplified cDNA of the mRNAs of the respective czcC genes being the correspondingly first gene in the tricystronic mRNA of czcCBA (van der Lelie et al. , 1997). For induction of gene expression, wild type cultures were treated with subinhibitory concentrations of cadmium, zinc, or cobalt. The chosen metal concentrations were 0.28 mM cadmium, 0.30 mM zinc, and 0.28 mM cobalt and slightly delayed the growth in comparison to the untreated sample. The growth delays of the treated cultures were reflected by their generation times which were 1:57 h ± 2:30 min for cadmium, 2:07 h ± 2:30 min for zinc, and 3:07 h ± 2:00 min for cobalt treatment as compared to a generation time of 1:47 h ± 1:00 min for the untreated sample. The gene expression results depicted in Table 2 showed that the presence of zinc strongly induced both czcC genes. The expression of czcC .1 was induced

78 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

about 900-fold while expression of czcC .2 showed a roughly 160-fold induction as compared to the untreated control. Neither of the two czcC genes was differentially expressed during cadmium or cobalt treatment with respect to the untreated controls ( Table 2 ), indicating that expression of both czcCBA operons of M. adhaerens HP15 is zinc-dependent but not responsive to cadmium or cobalt treatments.

Table 2: Relative fold-changes of czcC transcripts of operon czcCBA. 1 and czcCBA. 2 in response to heavy metals.

Relative fold-change

Operon Cadmium [0.28 mM] Zinc [0.30 mM] Cobalt [0.28 mM]

czcCBA. 1 1.6 ± 0.2 903.2 ± 30.2 1.1 ± 0.2

czcCBA. 2 0.5 ± 0.2 159.9 ± 26.8 0.6 ± 0.2

Complementation of the czcCBA .2-encoded efflux pump in mutant HP15. ∆czcCBA.2 Complementation of the phenotype of mutant HP15. ∆czcCBA.2 was attempted by introducing the corresponding czcCBA .2 operon cloned in plasmid pBBR.czcCBA.2 (Supplementary 7). Cloning of this operon was done in such a way that it was under control of the vector-borne lacZ promoter. Expression of the recombinant efflux pump operon was confirmed by introducing the plasmid to the wild type strain and analyzing gene expression by qRT-PCR (data not shown). Next, the czcCBA .2-deficient mutant was transformed with plasmid pBBR.czcCBA.2. Transformants of the wild type and the mutant containing vector pBBR1MCS-1 served as positive and negative controls, respectively. Deletion of operon czcCBA .2 in mutant HP15. ∆czcCBA.2 could be compensated by presence of the plasmid- borne operon czcCBA .2 ( Supplementary 10 ). Thus, plasmid pBBR.czcCBA.2 restored the ability to grow on agar medium supplemented with zinc and thus confirmed that the chromosomal inactivation of the czcCBA .2 encoded efflux pump causes a zinc-sensitive phenotype.

Protein domains and functional sites of the OMF CzcC in different bacterial strains Protein domains and potential functional regions were identified within the amino acid sequences of various CzcC proteins using the Interpro program (Jones et al. , 2014). CzcC proteins of M. adhaerens HP15 (HP15_108 and HP15_113), C. metallidurans CH34 (Rmet_5982), G. diazotrophicus PAI 5 (Gdia_2211), and P. putida KT2440 (PP_0045 and

79 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

PP2408) were analyzed with specific emphasis on differences in specific protein domains which could help explaining differences in substrate specificities. Domains that were identified in all analyzed proteins referred to their membrane association and ability to transport cations. None of the functional groups identified nor any sequence difference gave additional hints on protein specificity (data not shown).

Bioinformatics analysis of czcCBA operons in Marinobacter genomes Within the genome of M. adhaerens HP15, the two czcCBA operons are located close to each other separated only by a region of 3,342 bp that contains two further genes annotated as ‘PII- like nitrogen regulatory protein’ (HP15_111) and ‘TonB-dependent receptor protein’ (HP15_112) ( Figure 2a). The 14,500-bp genomic sequence comprising the two czcCBA operons of HP15 and ranging from gene czcC (HP15_108) to gene czcA (HP15_115) (nucleotide positions 111,522 – 126,022) was aligned and compared with the five sequenced and fully annotated Marinobacter genomes ( Table 3 ).

Table 3: Fully annotated Marinobacter genomes available on NCBI (05/2015) analyzed for presence of czcCBA operons. Strains containing two czcCBA operons located adjacently on the genome are shown in bold . No single operon was identified in any of the strains.

Marinobacter strain Accession Source of isolation Reference

M. adhaerens HP15 CP001978.1 sea water, German Kaeppel et al . 2012 CP001979.1 (plasmid pHP-42) Wadden Sea CP001980.1 (plasmid pHP-187)

M. sp. BSs20148 CP003735.1 sediment, 3800 m Song et al . 2013 depth, Arctic Ocean

M. hydrocarbonoclasticus VT8 CP000514.1 head of an oil- Huu et al. 1999 (former M. aquaeolei ) CP000515.1 (plasmid pMAQU01) producing well, Marquez and Ventosa CP000516.1 (plasmid pMAQU02) Vietnam 2005

M. hydrocarbonoclasticus ST17 FO203363.1 sea water near Gauthier et al. 1992 ATCC 49840 petroleum refinery, Mediterranean Sea

M. salarius R9SW1 CP007152.1 sea water, Tasman Sea Ng et al . 2014

M. similis A3d10 CP007151.1 sea water, Sea of Japan Ng et al . 2014

80 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

81 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

Figure 2: Schematic alignment of czcCBA operons in different Marinobacter species. a Genomic region of M. adhaerens HP15 encoding two CzcCBA heavy metal efflux pumps CzcCBA.1 and czcCBA.2 b and c Genomic regions of other Marinobacter species harboring two adjacent operons encoding CzcCBA efflux pumps ( b = full genome sequences; c = shot-gun genomes). Numbers in genes of czcC , B, and A indicated the percentage [%] of nucelotide sequence identity with the corresponding gene in M. adhaerens HP15 as revealed by BLAST alignments.

Strains M. salarius R9SW1 and M. similis A3d10 showed to have a similarly arranged region of two adjacent czcCBA operons ( Figure 2b) and showed significant overall sequence similarities ( Supplementary 11 ) suggesting a close phylogenetic proximity. Next, operons czcCBA. 1 (nt 111,522-116,975) and czcCBA. 2, (nt 120,318-126,022) were individually blasted against the five Marinobacter genomes to potentially identify the presence of a single pump operon. However, no additional similarities were found, showing that the genomes of the remaining strains do not harbor any CzcCBA-encoding regions. This finding was confirmed by blasting all six individual czc genes as well as their protein sequences without finding any further hit in the tested genomes (data not shown) thus excluding the option of parts of the operons being scattered across the genome. These results indicated that Marinobacter genomes either harbor two czcCBA operons in close proximity or none at all but never just carry a single czcCBA operon or single components. In order to substantiate this finding, 15 additional shot-gun sequenced Marinobacter genomes without full annotation and probably incomplete genome were tested for the presence of both czcCBA operons (Table 4 ). Four of those representing the strains M. algicola DG893, M. manganoxydans MnI7-9, M. nanhaiticus D15-8W, and M. sp. MCTG268 harbor each two adjacent czcCBA operons just as the genome of M. adhaerens HP15, thus confirming our

82 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

above reported observations ( Figure 2c). Due to the partially incomplete or multiple contigs- comprising nucleotide sequences it cannot be excluded that any of the remaining 11 strains harbors one or several operons. However, when blasting the two czcCBA operon sequences against the entire set of NCBI nucleotide sequence entries, no other bacterial species was found to contain two adjacent czcCBA operons. This obviously makes the observed genomic arrangements specific to representatives of the genus Marinobacter .

Table 4: Not fully annotated and partially incomplete shot-gut sequenced Marinobacter genomes available on NCBI (05/2015) analyzed for presence of czcCBA operons. Strains containing two czcCBA operons located adjacently on the genome are shown in bold . No single operon was identified in any of the strains.

Marinobacter strain Accession Source of isolation Reference

M. algicola DG893 ABCP00000000.1 dinoflagellate Green et al . 2006 laboratory culture

M. daepoensis DSM 16072 ATWI00000000.1 sea water, Yellow Sea Yoon et al . 2004

M. lipolyticus BF04_CF-4 ARCR00000000.1 - unpublished

M. lipolyticus SM19 ASAD00000000.1 saline soil, Spain Martin et al . 2003

M. manganoxydans MnI7-9 AGTR00000000.1 sediment of deep-sea Wang et al . 2012 hydrothermal vent, Indian Ocean

M. nanhaiticus D15-8W APLQ00000000.1 sediments, South China Gao et al . 2013 Sea

M. santoriniensis NKSG1 APAT00000000.1 hydrothermal sediment, Handley et al . 2009 Adrian Sea

M. sp. ELB17 AAXY00000000.1 Lake Bonney, Antarctica

M. sp. ES-1 AXBV00000000.1 beach sand, impacted Overholt et al. 2013 by oil spill, Gulf of Mexico

M. sp. EN3 AXCC00000000.1 as M. sp. ES-1 as M. sp. ES-1

M. sp. C1S70 AXBW00000000.1 as M. sp. ES-1 as M. sp. ES-1

M. sp. EVN1 AXCB00000000.1 as M. sp. ES-1 as M. sp. ES-1

M. sp. HL-58 JMLY00000000.1 - unpublished

M. sp. AK21 ANIE00000000.1 sea water, Bay of unpublished Bengal

M. sp. MCTG268 JQMK00000000.1 - unpublished

The highest nucleotide sequence similarity to the czcCBA operons of M. adhaerens HP15 was found for those present in M. similis A3d10, M. algicola DG893, and M. nanhaiticus D15-8W (Supplementary 11 ) whilst individual czc genes showed identities of 94 to 97% ( Figure 2b). Major differences between strains were observed in the gap between the two operons varying

83 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

in length and gene content ( Figure 2). All seven strains carry a 345-bp gene coding for the PII-like nitrogen regulatory protein always downstream of and directly adjacent to the first operon. In the genomes of M. algicola DG893, M. manganoxydans MnI7-9, and M. nanhaiticus this gene is the only one between both czcCBA operons suggesting a common phylogenetic track. An additional 2,112-bp Ton B-dependent receptor-encoding gene is found between the czcCBA operons of M. adhaerens HP15, M. similis , and Marinobacter sp. MCTG268 the latter of which shares a 730-bp inter-operon gene coding for a hypothetical protein with M. salarius . This mosaic of the presence of similar genes in close proximity continues for the up- and down-stream regions of both operons ( Figure 2) suggesting tight phylogenetic linkages in the composition of this genomic locus irrespective of the ecological habitats of the Marinobacter strains from which they were originally isolated.

84 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

Discussion

Heavy metals such as copper, iron, manganese, or zinc are essential elements for microorganisms acting as enzymatic co-factors (Lemire et al. , 2013) or as electron acceptors during anaerobic respiration (Weber et al. , 2006). However, high metal concentrations harm the organism, e.g., by formation of reactive oxygen species that damage DNA and proteins or by disruption of membrane integrity (Lemire et al. , 2013). To cope with heavy metal stress, different strategies evolved to detoxify the bacterial cell. Among those strategies, efflux pump systems belong to the most widespread cellular tools. The CzcCBA pump of C. metallidurans CH34, a well-studied model for heavy metal resistance, is known for its resistance towards divalent cadmium, zinc, and cobalt ions (Nies et al. , 1989; Scherer & Nies, 2009). Other authors have shown a similar substrate spectrum for the corresponding system in the plant-associated bacterium G. diazotrophicus PAI 5 (Intorne et al. , 2012). Mutants lacking the RND-pump CzcA of Helicobacter pylori 26695 and P. putida KT2440, respectively, showed increased susceptibility towards cadmium and zinc but not to cobalt (Stähler et al. , 2006; Leedjarv et al. , 2008). The CzcA pump of H. pylori 26695 also facilitated resistance towards nickel prompting the respective authors to rename the corresponding gene cznA (Stähler et al. , 2006). The present study demonstrated that the CzcA efflux pump of M. adhaerens HP15 only mediates resistance towards zinc since any involvement of the two CzcA pumps in cobalt or cadmium efflux could be ruled out experimentally. Based on the finding of Stähler and colleagues (2006), Nickel was recently tested in agar spot pre-experiment but could so far be excluded as a substrate of CzcA of strain HP15 (data not shown). In general, zinc seems to be a common substrate of CzcA pumps whereas efflux of other molecules might vary from pump to pump. Additional metal transporters exist that have functional overlaps with the CzcA pumps as they also transport zinc or cadmium as an example (Scherer & Nies, 2009). However, CzcA pumps fulfill the major zinc efflux function as shown for C. metallidurans CH34, where the P-type ATPases ZntA and CadA support the efflux of zinc only to a minor extend as compared to CzcA (Legatzki et al. , 2003b). Earlier studies with C. metallidurans CH34 aimed to identify the individual roles of CzcC, CzcB, and CzcA (Nies et al. , 1989; Rensing et al. , 1997). It was shown that the OMF CzcC is of minor importance for zinc resistance since mutagenesis of czcC resulted in only a small decrease in sensitivity towards this cation. In contrast, resistance towards cobalt and

85 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

especially cadmium was more efficiently impacted in a czcC -deficient mutant. Different mutations in czcA or czcB almost completely abolished efflux of zinc (Nies et al. , 1989). Herein, we speculate that the two CzcCBA complexes of M. adhaerens HP15 are altered so that efflux of cadmium and cobalt is not conducted anymore by these proteins, which instead are restricted to the transport of zinc. Analysis of potential functional domains in CzcCBA proteins revealed no helpful hints regarding differences in substrate specificity. Expression of both czcCBA operons of M. adhaerens HP15 was clearly induced by zinc but neither by cobalt nor cadmium thus substantiating the hypothesis that zinc is a major substrate these pumps facilitate resistance to. However, a direct link between inducibility of gene expression and actual substrate specificity of Czc efflux pumps requires caution: In P. aeruginosa PT5, expression of the CzcA pump was shown to be induced by copper, but a czcA -deficient mutant did not show a copper sensitive phenotype (Caille et al. , 2007). A set of data of czc gene expression was provided for M. xanthus , which harbors three czcCBA operons (Moraleda-Munoz et al. , 2010). One heavy metal pump was not induced by any tested metal whereas the other two pumps showed an induction in response to cadmium, cobalt, copper, iron, manganese, nickel, and zinc. Due to a lack of efflux mutants of M. xanthus , this pattern of gene expression was not correlated with the corresponding transport functions. We observed a more than 5-fold higher expression of gene czcC .1 over gene czcC .2. However, the single mutant HP15. ∆czcCBA.1 did not show a phenotype different from that of the wild type suggesting additional, not-yet-known function(s) of the corresponding pump. In contrast, operon czcCBA .2 showed a clear affiliation to zinc efflux as evidenced by an elevated sensitivity of mutant HP15. ∆czcCBA.2. The HP15. ∆czcCBA.1/2 double mutant showed a furthermore lowered MIC value implying that both efflux pumps contribute synergistically to zinc resistance. A similar observation was made in P. putida KT2440 which harbors two czcCBA operons (Leedjarv et al. , 2008). During exposure to zinc, only loss of one of the czcA genes resulted in a decrease in MIC while the other gene seemed to be dispensable. However, their double mutant was more sensitive towards zinc than the single mutant. In consequence, both P. putida KT2440 and M. adhaerens HP15 seem to harbor CzcCBA pumps that contribute in tandem to zinc resistance in a yet-to-be determined fashion. Interestingly, pairs of adjacently located czcCBA operons could be found in a subset of Marinobacter genomes. Both operons may originate from gene duplication events as supported by their proximate location within the genome (Taylor & Raes, 2004). Surprisingly,

86 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

we could not identify any czcCBA operon which occurred in singular form in any of the Marinobacter species. Thus, it is tempting to speculate about a common ancestor organism in which gene duplication may have occurred and might have provided a selective ecological advantage. Among the czcCBA -possessing Marinobacter species only M. manganoxydans MnI7-9 was actually isolated from a metal-associated environment and was demonstrated to oxidize manganese (II) (Wang et al. , 2012). The other strains carrying czcCBA operons were isolated from the marine water column, marine sediments, or laboratory cultures (Green et al. , 2006; Kaeppel et al. , 2012; Gao et al. , 2013; Ng et al. , 2014). The heterogeneous ecological features of the places of isolation for these organisms indicate that presence of CzcCBA efflux pumps cannot be linked with heavy metal-enriched habitats. Instead future studies should consider the micro-scale distribution and availability of heavy metals in the water column. Interestingly, although M. santoriniensis NKSG1 originated from metal-containing sediments of the Adriatic Sea (Handley et al. , 2009) its genome sequence did not reveal any czcCBA operons. This paradox situation is somewhat mirrored by previously described terrestrial systems: czcCBA operons were described in soil bacteria such as P. putida KT2440 (Leedjarv et al. , 2008) or M. xanthus (Moraleda-Munoz et al. , 2010) which were not isolated from metal-enriched environments. It may be concluded that CzcCBA efflux pumps are likely features of heavy metal-tolerant bacteria but their presence is not necessarily correlated with the occupation of heavy metal-enriched environments.

87 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

Conclusion

The present study provides first insights into heavy metal resistance features within the genus Marinobacter , which was described to occupy very diverse marine ecosystems but also habitats naturally enriched in heavy metals. While CzcCBA pumps were shown to facilitate resistance towards cadmium, zinc, and cobalt in other bacterial species, the corresponding pumps in M. adhaerens HP15 seemed to be specific for zinc. We furthermore conclude that presence of czcCBA operons within various genomes of Marinobacter species could not be directly correlated with the potential metal enrichment of the actual habitat from which strains were isolated. It therefore remains an open question why marine organism carry CzcCBA operons in duplicates and what their actual functions might be in context of the diverse habitats of Marinobacter species.

Acknowledgement

The authors would like to thank Dr. Helge Weingart for excellent technical suggestions on mutant generation and Desalegne Abebew Syit for help with cloning of knock-out plasmids. This project was funded by the Helmholtz Graduate School for Polar and Marine Research (POLMAR) and Deutsche Forschungsgemeinschaft (UL 169/6-1).

88 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

References

Abou-Shanab RAI, van Berkum P, Angle JS (2007). Heavy metal resistance and genotypic analysis of metal resistance genes in Gram-positive and Gram-negative bacteria present in Ni-rich serpentine soil and in the rhizosphere of Alyssum murale . Chemosphere 68:360-367.

Aguilera M, Jimenez-Pranteda ML, Kharroub K, Gonzalez-Paredes A, Durban JJ, Russell NJ , et al. (2009). Marinobacter lacisalsi sp. nov., a moderately halophilic bacterium isolated from the saline-wetland wildfowl reserve "Fuente de Piedra" in southern Spain. Int J Syst Evol Microbiol 59:1691-1695.

Caille O, Rossier C, Perron K (2007). A copper-activated two-component system interacts with zinc and imipenem resistance in Pseudomonas aeruginosa . J Bacteriol 189:4561-4568.

Gao W, Cui Z, Li Q, Xu G, Jia X, Zheng L (2013). Marinobacter nanhaiticus sp. nov., polycyclic aromatic hydrocarbon-degrading bacterium isolated from the sediment of the South China Sea. Antonie Van Leeuwenhoek 103:485-491.

Gärdes A, Iversen MH, Grossart H, Passow U, Ullrich MS (2011). Diatom-associated bacteria are required for aggregation of Thalassiosira weissflogii . ISME J 5:436-445.

Gärdes A, Kaeppel E, Shehzad A, Seebah S, Teeling H, Yarza P , et al. (2010). Complete genome sequence of Marinobacter adhaerens type strain (HP15), a diatom-interacting marine microorganism. Standards in Genomic Sciences 3:97-107.

Gauthier M, Lafay B, Christen R, Fernandez L, Acquaviva M, Bonin P , et al. (1992). Marinobacter hydrocarbonoclasticus gen. nov., sp. nov., a new, extremely halotolerant, hydrocarbon-degrading marine bacterium. Int J Syst Bacteriol 42:568-576.

Goldberg M, Pribyl T, Juhnke S, Nies D (1999). Energetics and topology of CzcA, a cation/ proton antiporter of the resistance-nodulation-cell division protein family. J Biol Chem 274:26065-26070.

Goris J, De Vos P, Coenye T, Hoste B, Janssens D, Brim H , et al. (2001). Classification of metal-resistant bacteria from industrial biotopes as Ralstonia campinensis sp. nov., Ralstonia metallidurans sp. nov. and Ralstonia basilensis Steinle et al. 1998 emend. Int J Syst Evol Microbiol 51:1773-1782.

Green D, Bowman J, Smith E, Gutierrez T, Bolch C (2006). Marinobacter algicola sp. nov., isolated from laboratory cultures of paralytic shellfish toxin-producing dinoflagellates. Int J Syst Evol Microbiol 56:523-527.

Grossart H, Schlingloff A, Bernhard M, Simon M, Brinkhoff T (2004). Antagonistic activity of bacteria isolated from organic aggregates of the German Wadden Sea. FEMS Microbiol Ecol 47:387-396.

Handley KM, Lloyd JR (2013). Biogeochemical implications of the ubiquitous colonization of marine habitats and redox gradients by Marinobacter species. Front Microbiol 4:136.

Handley KM, Hery M, Lloyd JR (2009). Marinobacter santoriniensis sp. nov., an arsenate-respiring and arsenite-oxidizing bacterium isolated from hydrothermal sediment. Int J Syst Evol Microbiol 59:1850-1850.

Hoang T, Karkhoff-Schweizer R, Kutchma A, Schweizer H (1998). A broad-host-range Flp-FRT recombination system for site-specific excision of chromosomally-located DNA sequences: Application for isolation of unmarked Pseudomonas aeruginosa mutants. Gene 212:77-86.

89 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

Intorne AC, de Oliveira MVV, Pereira LdM, de Souza Filho GA (2012). Essential role of the czc determinant for cadmium, cobalt and zinc resistance in Gluconacetobacter diazotrophicus PAI 5. Int Microbiol 15:69-78.

Jones P, Binns D, Chang H, Fraser M, Li W, McAnulla C , et al. (2014). InterProScan 5: Genome-scale protein function classification. Bioinformatics 30:1236-1240.

Leedjarv A, Ivask A, Virta M (2008). Interplay of different transporters in the mediation of divalent heavy metal resistance in Pseudomonas putida KT2440. J Bacteriol 190:2680-2689.

Legatzki A, Franke S, Lucke S, Hoffmann T, Anton A, Neumann D , et al. (2003a). First step towards a quantitative model describing Czc-mediated heavy metal resistance in Ralstonia metallidurans . Biodegradation 14:153-168.

Legatzki A, Grass G, Anton A, Rensing C, Nies D (2003b). Interplay of the Czc system and two P-type ATPases in conferring metal resistance to Ralstonia metallidurans . J Bacteriol 185:4354-4361.

Lemire JA, Harrison JJ, Turner RJ (2013). Antimicrobial activity of metals: Mechanisms, molecular targets and applications. Nat Rev Microbiol 11:371-384.

Moraleda-Munoz A, Perez J, Luis Extremera A, Munoz-Dorado J (2010). Differential regulation of six heavy metal efflux systems in the response of Myxococcus xanthus to copper. Appl Environ Microbiol 76:6069-6076.

Ng HJ, Lopez-Perez M, Webb HK, Gomez D, Sawabe T, Ryan J , et al. (2014). Marinobacter salarius sp. nov. and Marinobacter similis sp. nov., isolated from sea water. PLoS One 9:e106514.

Nies D (2003). Efflux-mediated heavy metal resistance in prokaryotes. FEMS Microbiol Rev 27:313-339.

Nies D (1995). The cobalt, zinc, and cadmium efflux system CzcABC from Alcaligenes eutrophus functions as a cation-proton antiporter in Escherichia coli . J Bacteriol 177:2707-2712.

Nies D, Nies A, Chu L, Silver S (1989). Expression and nucleotide sequence of a plasmid-determined divalent cation efflux system from Alcaligenes eutrophus . Proc Natl Acad Sci U S A 86:7351-7355.

Pletzer D, Weingart H (2014). Characterization and regulation of the Resistance-Nodulation-Cell Division-type multidrug efflux pumps MdtABC and MdtUVW from the fire blight pathogen Erwinia amylovora . BMC Microbiol 14:185.

Pruesse E, Quast C, Knittel K, Fuchs BM, Ludwig W, Peplies J , et al. (2007). SILVA: A comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB. Nucleic Acids Res 35:7188-7196.

Rensing C, Pribyl T, Nies D (1997). New functions for the three subunits of the CzcCBA cation-proton antiporter. J Bacteriol 179:6871-6879.

Sambrook J, Russel DW (2001). Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory: New York, USA.

Schenk A, Berger M, Keith LM, Bender CL, Muskhelishvili G, Ullrich MS (2006). The algT gene of Pseudomonas syringae pv. glycinea and new insights into the transcriptional organization of the algT-muc gene cluster. J Bacteriol 188:8013-8021.

Scherer J, Nies DH (2009). CzcP is a novel efflux system contributing to transition metal resistance in Cupriavidus metallidurans CH34. Mol Microbiol 73:601-621.

90 Results/ Part III

Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification

Singer E, Webb EA, Nelson WC, Heidelberg JF, Ivanova N, Pati A , et al. (2011). Genomic Potential of Marinobacter aquaeolei , a biogeochemical "opportunitroph". Appl Environ Microbiol 77:2763-2771.

Stähler F, Odenbreit S, Haas R, Wilrich J, Van Vliet A, Kusters J , et al. (2006). The novel Helicobacter pylori CznABC metal efflux pump is required for cadmium, zinc, and nickel resistance, urease modulation, and gastric colonization. Infect Immun 74:3845-3852.

Taylor J, Raes J (2004). Duplication and divergence: The evolution of new genes and old ideas. Annu Rev Genet 38:615-643.

Thoma S, Schobert M (2009). An improved Escherichia coli donor strain for diparental mating. FEMS Microbiol Lett 294:127-132.

Tseng T, Gratwick KS, Kollman J, Park D, Nies DH, Goffeau A , et al. (1999). The RND permease superfamily: An ancient, ubiquitous and diverse family that includes human disease and development proteins. J Mol Microbiol Biotechnol 1:107-125. van der Lelie D, Schwuchow T, Schwidetzky U, Wuertz S, Baeyens W, Mergeay M , et al. (1997). Two- component regulatory system involved in transcriptional control of heavy-metal homoeostasis in Alcaligenes eutrophus . Mol Microbiol 23:493-503.

Vandamme P, Coenye T (2004). Taxonomy of the genus Cupriavidus : A tale of lost and found. Int J Syst Evol Microbiol 54:2285-2289.

Vaneechoutte M, Kampfer P, De Baere T, Falsen E, Verschraegen G (2004). Wautersia gen. nov., a novel genus accommodating the phylogenetic lineage including Ralstonia eutropha and related species, and proposal of Ralstonia [Pseudomonas ] syzygii (Roberts et al. 1990) comb. nov. Int J Syst Evol Microbiol 54:317-327.

Wang H, Li H, Shao Z, Liao S, Johnstone L, Rensing C , et al. (2012). Genome sequence of deep-sea manganese- oxidizing bacterium Marinobacter manganoxydans MnI7-9. J Bacteriol 194:899-900.

Weber KA, Achenbach LA, Coates JD (2006). Microorganisms pumping iron: Anaerobic microbial iron oxidation and reduction. Nat Rev Microbiol 4:752-764.

Zhang D, Li H, Xin Y, Chi Z, Zhou P, Yu Y (2008). Marinobacter psychrophilus sp. nov., a psychrophilic bacterium isolated from the Arctic. Int J Syst Evol Microbiol 58:1463-1466.

Zumaquero A, Macho AP, Rufian JS, Beuzon CR (2010). Analysis of the role of the type III effector inventory of Pseudomonas syringae pv. phaseolicola 1448a in interaction with the plant. J Bacteriol 192:4474-4488.

92 Discussion and Future Scope

4. Discussion and Future Scope

4.1. Understanding Marinobacter adhaerens HP15 – Thalassiosira weissflogii interaction and suggestions on methodical improvements

In the present study, the intracellular proteomics profile of M. adhaerens HP15 during diatom-interaction with T. weissflogii was analyzed. Results indicated that amino acids, possibly released by the diatom, might be beneficial for the bacterium and therefore represent a beneficial mutualistic trade item of the diatom (Figure 10 ).

Figure 10: Traded compounds between M. adhaerens HP15 and T. weissflogii as identified in the present study. Cytoplasmic bacterial proteins up- ( ) and down-regulated ( ) during co-cultivation with the diatom are listed, indications derived from protein alterations are further given ( ).

The fact that amino acids may play a role during bacteria-diatom interaction was postulated in previous studies as, for example, an increase of intracellular amino acids in the diatom was determined when bacteria were present (Paul et al. , 2013). In another study, during interaction of Sulfitobacter sp. SA11 with P. multiseries specifically tryptophan was synthesized and

93 Discussion and Future Scope

released by the diatom (Amin et al. , 2015). In turn, tryptophan was further assumed to be taken up by the co-cultivated bacterium and used for the synthesis of the phytohormone IAA. This compound was re-allocated afterwards, causing growth enhancement in the diatom (Amin et al. , 2015). Amino acids could serve as carbon and nitrogen source as well as complete building blocks for the bacterium’s protein synthesis. In this context, it is worthwhile mentioning that nitrogen supply appeared to be more optimal for M. adhaerens HP15 in presence of the diatom as opposed to sole presence of the artificially supplemented amino acid, glutamate (Figure 10 ). It could thus be assumed that the apparent favorable supply with nitrogen was provided due to the release of amino acids by the diatom. The phenomenon that the algae might provide nitrogen favorable conditions for bacteria has not been stated so far. Amin and colleagues (2015) did show a transfer of an amino acid to the co-cultivated bacterium, which could be assumed as a nitrogen transfer. However, the amino acid was further suggested to be re- allocated, as explained above. In addition, the bacterium also supplied ammonium, which was taken up by the diatom. Thus a net transfer of nitrogen from the bacterium to the alga can likely be assumed here. Also other cases are known, where diatoms receive nitrogen, in that particular case from cyanobacteria (Foster et al. , 2011). The findings that diatoms are provided with nitrogen make sense in that way, as natural environments of these algae are often nitrogen-limited (Moore et al. , 2013). Thus, I have to assume that the given in vitro scenario of my work, in which diatoms facilitate nitrogen-favorable conditions for the bacterium, add a new insight to bacteria-diatom interactions but are likely rather rare under natural conditions. Conditions used in the present study need to be considered as artificial. To explain the allocation of nitrogen from the diatom to the bacterium, one should consider the presence of glutamate which was offered in the growth medium. We cannot rule out the possibility that the diatom benefitted from glutamate as a nitrogen source. Such ‘nitrogen rich’ conditions would enable the diatom to be rather ‘sloppy’ with its nitrogen budget, potentially releasing amino acids, thereby not competing for nitrogen with the bacterium but providing excess nitrogen to the prokaryote. In this context, the use of glutamate needs to be discussed as a factor which might have complicated the analysis of the present study. Glutamate was used for two main reasons: Primary, M. adhaerens HP15 cultivated without a carbon source would not have resulted in growth of the bacterium and synthesis of sufficient amounts of protein in response to applied experimental conditions (particularly in the reference). With the given instrumental boundary conditions, the harvest of a rather large amount of proteins was required. Secondly, the comparison of bacteria grown with a carbon

94 Discussion and Future Scope

source and a carbon source plus diatom, respectively, appeared reasonable since it made a direct comparison easier. In contrast, analyzing the interaction under conditions where no external carbon source is provided in comparison to the treatment with diatoms (where carbon might be available as EPS/ TEP) would likely have created a number of artefacts, based on, for example, starvation conditions in the reference. However, I am aware of the problem, that the addition of glutamate might have decoupled the intimate interaction of both partners probably to a certain extend. To face the problem of low protein concentration and to conduct a co-cultivation experiment without the addition of a carbon source, it is necessary to shift to more sensitive protein detection methods, as briefly discussed further below. Data analysis would then have to dissect artefacts based on, e.g., starvation, potentially found in the non- carbon containing control reference. The question still remains, why valuable compounds such as amino acids should be released by the diatom. A rather advanced application would be that these amino acids are used by the bacterium for the synthesis of a compound that is then re-allocated to the alga cell. An example for this is the synthesis of IAA, stated above (Amin et al. , 2015). An initial possible explanation in the context of the present study is that M. adhaerens HP15 stimulates the release of EPS, TEP respectively in T. weissflogii as shown in previous studies (Gärdes et al. , 2011). Amino acids may simply be part of this fraction, as they were indeed found to be released by diatoms as part of EPS (Myklestad et al. , 1989). As a consumer of amino acids, M. adhaerens HP15 might benefit from this fraction and consequently up-regulate amino acid uptake systems, as observed in the present study. It therefore remains questionable, if the release of those compounds by the diatom is actually a specific response to the presence of M. adhaerens HP15. It could also simply be a matter of fact, that amino acids as part of the EPS fraction are an attractant for the bacterium, which would question the ‘intimate and individual’ interaction of both partners. It thus should be the aim of future studies, to deeper ask for the quality of this specific interaction. To better understand the interaction of M. adhaerens HP15 and T. weissflogii , additional follow-up studies based on the present one will benefit from the analysis of additional protein fractions. Thus, a more complete picture of the bacterial proteome could be represented (Becher et al. , 2011). A representative study would include not only cytoplasmic, but also membrane and extracellular protein fractions. Ideally, a methodical add-on will combine the proteomics approaches with metabolomics, as conducted by Paul and colleagues (2013). By this, enzymatic evidence on the proteomics level can be complemented with the detection of actual substrates on the metabolic level (Paul et al. , 2013). Further, hints on possible signaling

95 Discussion and Future Scope

molecules can be retrieved. Analysis of membrane-bound proteins would give additional support for the release or uptake of compounds, due to the altered abundances of compound- specific transporters. Transport-associated proteins were identified in this study as being up- or down-regulated during co-cultivation. However, the proteins were not truly membrane bound but usually located in the periplasm. An up-regulation of those transporter-associated proteins should be in line with the up-regulation of the corresponding membrane embedded transporter proteins and consequently support present results. However, as a possible draw back the analysis of membrane embedded proteins is rather advanced (Rabilloud, 2009). For a full proteomic picture of the interaction, it almost goes without saying to also consider and include proteome data of T. weissflogii . Knowledge about alterations in the diatom’s proteome would probably answer the question, whether amino acid synthesis and release is specifically stimulated by the bacterium or a certain amino acid fraction is just released along with EPS. Results of the herein applied proteomics approach give a number of hints on the actual interaction between M. adhaerens HP15 and T. weissflogii . In the future, a number of possible technical improvements should be considered: In the current study, we used small protein gels (7 cm width), which in comparison to bigger gels of up to 24 cm resulted in lower resolution and, due to a lower protein load, less proteins being visualized. Furthermore, Coomassie ® staining was the method of choice used for protein visualization, which is usually considered as a low sensitivity staining method (Rehm, 2006). The application of fluorescence-dye staining might lead to a reasonable improvement, as those dyes have a number of advantages, particularly a higher sensitivity but also a (more) linear staining over orders of magnitudes of protein concentrations (Rehm, 2006; J. Bernhardt, personal communication). Possible instrumental improvements like the application of high performance electrophoresis would require certain investments but likely lead to quality improvement and deeper ranging information on altered protein patterns (Moche et al. , 2013). Importantly, such an instrument would also safe time and chemicals. Gel-free or semi-gel approaches combined with sensitive mass spectrometry would further improve the amount of recovered, detectable protein fractions (Wolff et al. , 2006). The financial investments needed for the set-up of the stated methodical options would be high and possibly desire skilled staff. As an option, collaborations with suitable partner laboratories might circumvent this bottle neck. A common problem of in vitro tests is their already stated rather artificial character generating artefacts that disturbs the interpretation of data in light of their environmental implications. Therefore, a ‘long term’ step that follows methodical improvements is to further move away

96 Discussion and Future Scope

from the principle of model systems like the one used in the present study. From my point of view, a fundamental and environmentally representative understanding of bacteria-diatom interaction needs to become independent of lab-based cultivation techniques at some point. It is of course the case that an initial understanding of a (scientific) issue desires the application of a first simple set-up with an ideally small amount of (adjustable and controllable) parameters, underlining that also model systems as the one used in the present study are important. Thus such lab-based in vitro approaches like the one conducted are mandatory for pushing the borders of knowledge and gaining new hypotheses, supporting their importance. This way, valuable information on bacteria-diatom interaction has been reported, and I would again like to emphasize the study of Amin and colleagues (2015) as being of high and complex quality with reference to present available studies. However, neither our, nor the study of Amin et al . (2015) can give an answer, how the two co-cultivated partners will interact in their natural environment. In nature such ‘model systems’ are entrapped in a plot with many other characters, different bacterial and algal species, higher organisms, physical and chemical parameters under constantly changing conditions (Figure 11 ). Therefore at some point it becomes necessary to push scientific effort towards the context of higher environmental complexity. This point might be determined by onward improvement in knowledge but also methodical capabilities.

Figure 11: Model system versus actual environmental conditions and complexity (‘nature’). In nature, the ‘protagonists’ of a laboratory based in vitro model system face a multitude of other organisms and chemical conditions, impacting their behaviors (the scheme of ‘nature’ is likely still highly underestimated and merely gives a rough idea of versatility under environmental conditions). Note that scales were moderately respected.

97 Discussion and Future Scope

The combination of different methodical perspectives, e.g. on all three levels of (meta-) proteomics, (meta-) genomics and (meta-) metabolomics, will be a first step to fully illuminate complex aspects. First studies that go into that direction are known (Teeling et al. , 2012). The challenges of this demand are obvious, but appear as the next logic step towards understanding complex diverse systems (which is in fact nature) that are ideally less biased by laboratory artifacts. Laboratory-based approaches will of course not become irrelevant in this context: Due to the need of fundamental functional information behind genome and protein sequences, lab-based efforts are actually highly desired. The importance of lab-based efforts for the development of new hypotheses has been stated above.

98 Discussion and Future Scope

4.2. Chemotaxis and the colonization of marine particles

In the present study we demonstrated that up to three different M. adhaerens HP15 mutants could be visualized and thus detected in-stereo via differential fluorescence signaling. The analyzed mutants were deficient in chemotaxis capabilities, as there is motility due to the lack of the flagellum, a chemosensing mutant due to the lack of the histidine kinase CheA resulting in a lack of flagellum response towards environmental stimuli, as well as pilus-impaired mutants. Initial ‘proof of principle’ for this stereo tracking approach has been conducted successfully. A next step now is to use these mutants for, e.g., quantification experiments of particle colonization. Initially this could be conducted with artificial particles, in detail agar spheres as applied in previous studies (Kiørboe et al. , 2003; Ploug et al. , 2010). By this the quantity of marine particle colonization due to, e.g., random collision can be estimated in comparison to active recognition and adhesion due to chemosensing and motility. An initial test trial for particle colonization was conducted as part of the present study, in which both fluorescently labeled wild type and mutants were exposed to artificial agar spheres. Surprisingly, no obvious difference in sphere colonization was observed between different mutants and the wild type (Supplementary 12 ), possibly due to inappropriate high cell numbers used. Optimization of this aspect is needed and will become the subject of prospective studies. As suggested in the previous section, colonization experiments also need to be shifted towards set-ups that better reflect the natural environmental conditions. This could be done by colonization experiments conducted with marine aggregates, generated in and carried out as roller tank experiments. The latter have been established for attachment experiments of M. adhaerens HP15 with T. weissflogii in the past (Gärdes et al. , 2010). I further suggest that our previously generated molecular and cellular tools may be used to identify the impact and importance of chemotactic abilities with special emphasis on actual environmental conditions: During algal blooms nutritional hot spots are more abundant in the water column than in a non-bloom event, as more algae are present that create nutrient enriched areas in their phycosphere. Based on this it can be assumed that during bloom events the ability of chemotaxis may be less important for bacteria simply because the chance of random collision with a hot spot is higher and nutrient gradients are less pronounced. In turn, chemotaxis might become more beneficial when nutritional hot spots become rare and thus a selective targeting is needed in order to encounter a hot spot. However, the maintenance of chemotaxis is cost intensive for bacteria (Martinez-Garcia et al. , 2014) suggesting that chemotaxis will – for long – be not sustainable in oligotrophic environments. Thus we have to assume that there is a maximum benefit of chemotaxis at a certain ‘concentration of

99 Discussion and Future Scope

patchiness and nutrient availability’ in the water column. The bacterium’s ability to switch between motile and non-motile stages is likely a useful form of adaption in this context, as it has been reported in laboratory experiments for a Vibrio representative (Stretton et al. , 1997). The quantification of attached M. adhaerens HP15 wild type versus chemotaxis mutants at different marine aggregate concentrations will probably give answers to this hypothesis. With a broader view on climate change, the benefits of chemotaxis for bacteria may even shift as a result of anthropogenic impact. A recent study observed a reduced aggregate formation in the scenario of both increased ocean acidification and temperature rise (Seebah et al. , 2014). Reduced aggregate formation implies that the water column would be less patchy and hot spots more evenly distributed. Consequently the benefit of bacterial chemotaxis will be reduced and the former disadvantage of being ‘non-chemotactic’ might vanish under those conditions. However, this scenario is just one idea to discuss chemotaxis in light of upcoming anthropogenic alterations, as a wide array of additional factors might impact aggregate formation under changing conditions in the future ocean. How the formation of aggregates and the functioning of the biological carbon pump will change is not fully understood (Passow & Carlson, 2012) but deserves experimental approaches in the future.

100 Discussion and Future Scope

4.3. Heavy metals resistance in Marinobacter adhaerens HP15 and further genus members

In a previous review article on the genus Marinobacter it was proposed that this genus might play a so-far underestimated or neglected role in biogeochemical metal cycling, carried out in its natural habitats (Handley & Lloyd, 2013). Evidence was based on a number of observations, as members of this genus show metabolic features like the oxidation of iron, manganese, and arsenite (Handley et al. , 2009; Singer et al. , 2011; Wang et al. , 2012) or are according to genomic information able to deal with the presence of certain metal ions (Wang et al. , 2012; Stahl et al. , 2015). Indeed, isolates of Marinobacter species were found in, e.g., hydrothermal vent environments (Kaye et al. , 2011; Wang et al. , 2012), often characterized by elevated heavy metal concentrations (Kadar et al. , 2005). It is therefore interesting to look at possible genomic proof represented by genes encoding heavy metal efflux pumps which allow the bacterium to inhabit such environments. This was actually done, for example, in the case of M. manganoxydans where genes involved in resistance towards certain metals were identified in the genome (Wang et al. , 2012). In the present study, the presence of two encoded CzcCBA efflux pumps was considered as proof for M. adhaerens HP15 being an interesting candidate to analyze the bacterium’s functions with respect to inhabited environments of the genus (Stahl et al. , 2015). In case of M. adhaerens HP15 which was not isolated from environments enriched in heavy metals, the two CzcCBA pumps previously described as facilitating resistance towards cadmium, zinc, and cobalt merely conferred resistance to elevated zinc concentration. Resistance towards cadmium or cobalt could not be identified. I assume, we also need to raise the question to what extend CzcCBA pumps and other metal-detoxifying cellular tools actually imply a life style adapted to heavy metal- enriched environments. CzcCBA pumps might be encoded in a wide range of bacteria as a classical tool to cope with appearing metal stress: Indeed a keyword search for CzcA on NBCI reveals almost 19,000 hits among the protein database. This result thus questions the uniqueness of genomic regions, annotated as czcA . To fully estimate the distribution among prokaryotes, a bioinformatics survey among all available genomes will allow a non- or better to say less-biased estimation. This data could then be correlated to additional contextual information like sampling site and, if available, site specific heavy metal concentrations. Ideally, this information will further be brought into context of laboratory-based protein characterization that gives evidence on the actual exported substrates and efflux pump- associated features. Not only the presence of genes encoding for CzcCBA pumps determines their potential feature in heavy metal detoxification but also its functionality and efficiency.

101 Discussion and Future Scope

Therefore, work of this study emphasizes the ongoing importance of identifying the role of gene products in wet lab approaches. Although gene and genome annotations already give a good idea on the metabolic potential of a bacterium, gaps in knowledge on the function of a gene product (‘proteins with unknown function’) are common and complicated by the potentially wrong-leading annotations of less conserved homologues that have been functionally analyzed in the past. Handley and Lloyd (2013) have assumed a so-far underestimated role of the genus Marinobacter in biogeochemical metal cycling. They based this idea on certain features found in some representatives of this genus. I assume that more evidence is needed to prove this hypothesis and to clearly show that features of particularly this genus play a role in geochemical metal cycling, also on a quantitative level. Members of the γ-proteobacterial genus Alteromonas for example are found in the marine water column (Yoon et al. , 2003; Ivanova et al. , 2005) but were also isolated from heavy metal enriched environments such as hydrothermal vents (Raguenes et al. , 1997). Signs of heavy metal resistance have been observed in this genus that is closely related to Marinobacter (Jeanthon & Prieur, 1990; Chiu et al. , 2007). Interestingly, a number of Alteromonas species were isolated from a marine biofilm grown on a stainless steel cathode that had been set under current (‘electroactive biofilm’) (Vandecandelaere et al. , 2008). Whilst M. manganoxydans was isolated from manganese nodules in the deep sea, a number of Alteromonas species were found on ferromanganese nodules (Zhang et al. , 2015). The latter information gives the impression that a potentially important role of the genus Marinobacter in geochemical metal cycling is not restricted to this genus but might be more widespread in the family of Alteromonadaceae.

102 References

5. References

Alavi M, Miller T, Erlandson K, Schneider R, Belas R (2001). Bacterial community associated with Pfiesteria - like dinoflagellate cultures. Environ Microbiol 3:380-396.

Albareda M, Rodrigue A, Brito B, Ruiz-Argueso T, Imperial J, Mandrand-Berthelot M , et al. (2015). Rhizobium leguminosarum HupE is a highly-specific diffusion facilitator for nickel uptake. Metallomics 7:691-701.

Alldredge A, Passow U, Logan B (1993). The abundance and significance of a class of large, transparent organic particles in the ocean. Deep-Sea Res Part I-Oceanogr Res Pap 40:1131-1140.

Alldredge A, Silver M (1988). Characteristics, dynamics and significance of marine snow. Prog Oceanogr 20:41- 82.

Alvarez-Ortega C, Olivares J, Martinez JL (2013). RND multidrug efflux pumps: What are they good for? Front Microbiol 4:7.

Amin SA, Hmelo LR, van Tol HM, Durham BP, Carlson LT, Heal KR , et al. (2015). Interaction and signalling between a cosmopolitan phytoplankton and associated bacteria. Nature 522:98-U253.

Amin SA, Parker MS, Armbrust EV (2012). Interactions between diatoms and bacteria. Microbiol Mol Biol Rev 76:667-684.

Amin SA, Green DH, Hart MC, Kuepper FC, Sunda WG, Carrano CJ (2009). Photolysis of iron-siderophore chelates promotes bacterial-algal mutualism. Proc Natl Acad Sci U S A 106:17071-17076.

Anton A, Grosse C, Reissmann J, Pribyl T, Nies D (1999). CzcD is a heavy metal ion transporter involved in regulation of heavy metal resistance in Ralstonia sp. strain CH34. J Bacteriol 181:6876-6881.

Armbrust E, Berges J, Bowler C, Green B, Martinez D, Putnam N , et al. (2004). The genome of the diatom Thalassiosira pseudonana : Ecology, evolution, and metabolism. Science 306:79-86.

Armstrong R, Lee C, Hedges J, Honjo S, Wakeham S (2002). A new, mechanistic model for organic carbon fluxes in the ocean based on the quantitative association of POC with ballast minerals. Deep-Sea Res Part II-Top Stud Oceanogr 49:219-236.

Azam F (1998). Microbial control of oceanic carbon flux: The plot thickens. Science 280:694-696.

Azam F, Fenchel T, Field J, Gray J, Meyer-Reil L, Thingstad F (1983). The ecological role of water-column microbes in the sea. Mar Ecol Prog Ser 10:257-263.

Azam F, Malfatti F (2007). Microbial structuring of marine ecosystems. Nat Rev Microbiol 5:782-791.

Baker LJ, Kemp PF (2014). Exploring bacteria-diatom associations using single-cell whole genome amplification. Aquat Microb Ecol 72:73-88.

Barbara G, Mitchell J (2003). Bacterial tracking of motile algae. FEMS Microbiol Ecol 44:79-87.

103 References

Barbeau K, Rue EL, Bruland KW, Butler A (2001). Photochemical cycling of iron in the surface ocean mediated by microbial iron(III)-binding ligands. Nature 413:409-413.

Becher D, Buettner K, Moche M, Hessling B, Hecker M (2011). From the genome sequence to the protein inventory of Bacillus subtilis . Proteomics 11:2971-2980.

Bell W, Mitchell R (1972). Chemotactic and growth responses of marine bacteria to algal extracellular products. Biol Bull 143:265-277.

Berg HC (2003). E. Coli in Motion. Springer: New York.

Bhaskar P, Grossart H, Bhosle N, Simon M (2005). Production of macroaggregates from dissolved exopolymeric substances (EPS) of bacterial and diatom origin. FEMS Microbiol Ecol 53:255-264.

Binet M, Poole R (2000). Cd(II), Pb(II) and Zn(II) ions regulate expression of the metal-transporting P-type ATPase ZntA in Escherichia coli . FEBS Lett 473:67-70.

Blackburn N, Fenchel T, Mitchell J (1998). Microscale nutrient patches in planktonic habitats shown by chemotactic bacteria. Science 282:2254-2256.

Bruckner CG, Rehm C, Grossart H, Kroth PG (2011). Growth and release of extracellular organic compounds by benthic diatoms depend on interactions with bacteria. Environ Microbiol 13:1052-1063.

Bruckner CG, Bahulikar R, Rahalkar M, Schink B, Kroth PG (2008). Bacteria associated with benthic diatoms from Lake Constance: Phylogeny and influences on diatom growth and secretion of extracellular polymeric substances. Appl Environ Microbiol 74:7740-7749.

Buesseler K, Andrews J, Pike S, Charette M (2004). The effects of iron fertilization on carbon sequestration in the Southern Ocean. Science 304:414-417.

Caldeira K, Wickett M (2003). Anthropogenic carbon and ocean pH. Nature 425:365-365.

Chao Y, Fu D (2004). Kinetic study of the antiport mechanism of an Escherichia coli zinc transporter, ZitB. J Biol Chem 279:12043-12050.

Chin W, Orellana M, Verdugo P (1998). Spontaneous assembly of marine dissolved organic matter into polymer gels. Nature 391:568-572.

Chiu H, Shieh WY, Lin SY, Tseng C, Chiang P, Wagner-Doebler I (2007). Alteromonas tagae sp. nov. and Alteromonas simiduii sp. nov., mercury-resistant bacteria isolated from a Taiwanese estuary. Int J Syst Evol Microbiol 57:1209-1216.

Cho B, Azam F (1988). Major role of bacteria in biogeochemical fluxes in the oceans interior. Nature 332:441- 443.

Coale K, Johnson K, Fitzwater S, Gordon R, Tanner S, Chavez F , et al. (1996). A massive phytoplankton bloom induced by an ecosystem-scale iron fertilization experiment in the equatorial Pacific Ocean. Nature 383:495-501.

Comino E, Whiting S, Neumann P, Baker A (2005). Salt (NaCl) tolerance in the Ni hyperaccumulator Alyssum murale and the Zn hyperaccumulator Thlaspi caerulescens . Plant Soil 270:91-99.

Croft M, Lawrence A, Raux-Deery E, Warren M, Smith A (2005). Algae acquire vitamin B 12 through a symbiotic relationship with bacteria. Nature 438:90-93.

Cubillas C, Vinuesa P, Tabche ML, Davalos A, Vazquez A, Hernandez-Lucas I , et al. (2014). The cation diffusion facilitator protein EmfA of Rhizobium etli belongs to a novel subfamily of Mn2 +/Fe2 + transporters conserved in α-proteobacteria. Metallomics 6:1808-1815.

104 References

Diercks A, Asper V (1997). In situ settling speeds of marine snow aggregates below the mixed layer: Black Sea and Gulf of Mexico. Deep-Sea Res Part I-Oceanogr Res Pap 44:385-&.

Dinh T, Paulsen I, Saier M (1994). A family of extracytoplasmic proteins that allow transport of large molecules across the outer membranes of Gram-negative bacteria. J Bacteriol 176:3825-3831.

Dubilier N, Bergin C, Lott C (2008). Symbiotic diversity in marine animals: The art of harnessing chemosynthesis. Nat Rev Microbiol 6:725-740.

Ducklow HW, Steinberg DK, Buesseler KO (2001). Upper ocean carbon export and the biological pump. Oceanography 14:50-58.

Engel A, Thoms S, Riebesell U, Rochelle-Newall E, Zondervan I (2004). Polysaccharide aggregation as a potential sink of marine dissolved organic carbon. Nature 428:929-932.

Engel A (2000). The role of transparent exopolymer particles (TEP) in the increase in apparent particle stickiness (alpha) during the decline of a diatom bloom. J Plankton Res 22:485-497.

Falkowski P, Barber R, Smetacek V (1998). Biogeochemical controls and feedbacks on ocean primary production. Science 281:200-206.

Fang C, Chen H, Chuang Y, Chang S, Wang J (2002). Cloning of a cation efflux pump gene associated with chlorhexidine resistance in Klebsiella pneumoniae . Antimicrob Agents Chemother 46:2024-2028.

Fernandez S, Seoane S, Merino A (1999). Plant heavy metal concentrations and soil biological properties in agricultural serpentine soils. Commun Soil Sci Plant Anal 30:1867-1884.

Field C, Behrenfeld M, Randerson J, Falkowski P (1998). Primary production of the biosphere: Integrating terrestrial and oceanic components. Science 281:237-240.

Foster RA, Kuypers MMM, Vagner T, Paerl RW, Musat N, Zehr JP (2011). Nitrogen fixation and transfer in open ocean diatom-cyanobacterial symbioses. ISME J 5:1484-1493.

Fowler SW, Knauer GA (1986). Role of large particles in the transport of elements and organic-compounds through the oceanic water column. Prog Oceanogr 16:147-194.

Fralick J (1996). Evidence that TolC is required for functioning of the Mar/AcrAB efflux pump of Escherichia coli . J Bacteriol 178:5803-5805.

Friedlingstein P, Houghton RA, Marland G, Hackler J, Boden TA, Conway TJ , et al. (2010). Update on CO 2 emissions. Nat Geosci 3:811-812.

Fukami K, Nishijima T, Ishida Y (1997). Stimulative and inhibitory effects of bacteria on the growth of microalgae. Hydrobiologia 358:185-191.

Gaballa A, Helmann J (2003). Bacillus subtilis CPx-type ATPases: Characterization of Cd, Zn, Co and Cu efflux systems. Biometals 16:497-505.

Gärdes A, Iversen MH, Grossart H, Passow U, Ullrich MS (2011). Diatom-associated bacteria are required for aggregation of Thalassiosira weissflogii . ISME J 5:436-445.

105 References

Geesey G, Jang L, Jolley J, Hankins M, Iwaoka T, Griffiths P (1988). Binding of metal-ions by extracellular polymers of biofilm bacteria. Water Sci Technol 20:161-165.

Grass G, Otto M, Fricke B, Haney C, Rensing C, Nies D , et al. (2005). FieF (YiiP) from Escherichia coli mediates decreased cellular accumulation of iron and relieves iron stress. Arch Microbiol 183:9-18.

Grass G, Fan B, Rosen B, Franke S, Nies D, Rensing C (2001a). ZitB (YbgR), a member of the cation diffusion facilitator family, is an additional zinc transporter in Escherichia coli . J Bacteriol 183:4664-4667.

Grass G, Fan B, Rosen B, Lemke K, Schlegel H, Rensing C (2001b). NreB from Achromobacter xylosoxidans 31A is a nickel-induced transporter conferring nickel resistance. J Bacteriol 183:2803-2807.

Grass G, Grosse C, Nies D (2000). Regulation of the cnr cobalt and nickel resistance determinant from Ralstonia sp. strain CH34. J Bacteriol 182:1390-1398.

Green D, Llewellyn L, Negri A, Blackburn S, Bolch C (2004). Phylogenetic and functional diversity of the cultivable bacterial community associated with the paralytic shellfish poisoning dinoflagellate Gymnodinium catenatum . FEMS Microbiol Ecol 47:345-357.

Grossart H, Simon M (2007). Interactions of planktonic algae and bacteria: Effects on algal growth and organic matter dynamics. Aquat Microb Ecol 47:163-176.

Grossart H, Tang KW, Kiørboe T, Ploug H (2007). Comparison of cell-specific activity between free-living and attached bacteria using isolates and natural assemblages. FEMS Microbiol Lett 266:194-200.

Grossart H, Kiørboe T, Tang K, Allgaier M, Yam E, Ploug H (2006). Interactions between marine snow and heterotrophic bacteria: Aggregate formation and microbial dynamics. Aquat Microb Ecol 42:19-26.

Grossart H, Levold F, Allgaier M, Simon M, Brinkhoff T (2005). Marine diatom species harbour distinct bacterial communities. Environ Microbiol 7:860-873.

Grossart H, Schlingloff A, Bernhard M, Simon M, Brinkhoff T (2004). Antagonistic activity of bacteria isolated from organic aggregates of the German Wadden Sea. FEMS Microbiol Ecol 47:387-396.

Grossart H, Riemann L, Azam F (2001). Bacterial motility in the sea and its ecological implications. Aquat Microb Ecol 25:247-258.

Gu J, Cai H, Yu S, Qu R, Yin B, Guo Y , et al. (2007). Marinobacter gudaonensis sp. nov., isolated from an oil- polluted saline soil in a Chinese oilfield. Int J Syst Evol Microbiol 57:250-254.

Guannel ML, Horner-Devine MC, Rocap G (2011). Bacterial community composition differs with species and toxigenicity of the diatom Pseudo -nitzschia . Aquat Microb Ecol 64:117-133.

Guffanti A, Wei Y, Rood S, Krulwich T (2002). An antiport mechanism for a member of the cation diffusion facilitator family: Divalent cations efflux in exchange for K + and H +. Mol Microbiol 45:145-153.

Haines K, Guillard R (1974). Growth of vitamin B 12 -requiring marine diatoms in mixed laboratory cultures with vitamin B 12 -producing marine bacteria. J Phycol 10:245-252. 106 References

Handley KM, Lloyd JR (2013). Biogeochemical implications of the ubiquitous colonization of marine habitats and redox gradients by Marinobacter species. Front Microbiol 4:136.

Hansell DA, Carlson CA (2001). Marine dissolved organic matter and the carbon cycle. Oceanography 14(4):41- 49.

Hold G, Smith E, Rappe M, Maas E, Moore E, Stroempl C , et al. (2001). Characterisation of bacterial communities associated with toxic and non-toxic dinoflagellates: Alexandrium spp. and Scrippsiella trochoidea . FEMS Microbiol Ecol 37:161-173.

Huenken M, Harder J, Kirst GO (2008). Epiphytic bacteria on the Antarctic ice diatom Amphiprora kufferathii Manguin cleave hydrogen peroxide produced during algal photosynthesis. Plant Biol 10:519-526.

Huettel M, Wild C, Gonelli S (2006). Mucus trap in coral reefs: Formation and temporal evolution of particle aggregates caused by coral mucus. Mar Ecol Prog Ser 307:69-84.

Ivanova E, Bowman J, Lysenko A, Zhukova N, Gorshkova N, Sergeev A , et al. (2005). Alteromonas addita sp. nov. Int J Syst Evol Microbiol 55:1065-1068.

Jeanthon C, Prieur D (1990). Susceptibility to heavy metals and characterization of heterotrophic bacteria isolated from two hydrothermal vent polychaete Annelids, Alvinella pompejana and Alvinella caudata . Appl Environ Microbiol 56:3308-3314.

Jiang H, Lou W, Du H, Price NM, Qiu B (2012). Sll1263, a unique cation diffusion facilitator protein that promotes iron uptake in the cyanobacterium Synechocystis sp. strain PCC 6803. Plant Cell Physiol 53:1404- 1417.

Jobling M, Peters S, Ritchie D (1988). Plasmid-borne mercury resistance in aquatic bacteria. FEMS Microbiol Lett 49:31-37.

Johansen J, Pinhassi J, Blackburn N, Zweifel U, Hagstrom A (2002). Variability in motility characteristics among marine bacteria. Aquat Microb Ecol 28:229-237.

Kaczmarska I, Ehrman J, Bates S, Green D, Leger C, Harris J (2005). Diversity and distribution of epibiotic bacteria on Pseudo-nitzschia multiseries (Bacillariophyceae) in culture, and comparison with those on diatoms in native seawater. Harmful Algae 4:725-741.

Kadar E, Costa V, Martins I, Santos RS, Powell JJ (2005). Enrichment in trace metals (Al, Mn, Co, Cu, Mo, Cd, Fe, Zn, Pb and Hg) of macro-invertebrate habitats at hydrothermal vents along the Mid-Atlantic Ridge. Hydrobiologia 548:191-205.

Kalmokoff M, Lanthier P, Tremblay T, Foss M, Lau P, Sanders G , et al. (2006). Proteomic analysis of Campylobacter jejuni 11168 biofilms reveals a role for the motility complex in biofilm formation. J Bacteriol 188:4312-4320.

Kanamaru K, Kashiwagi S, Mizuno T (1994). A copper-transporting P-type ATPase found in the thylakoid membrane of the cyanobacterium Synechococcus species PCC7942. Mol Microbiol 13:369-377.

Kaye JZ, Sylvan JB, Edwards KJ, Baross JA (2011). Halomonas and Marinobacter ecotypes from hydrothermal vent, subseafloor and deep-sea environments. FEMS Microbiol Ecol 75:123-133.

107 References

Kiørboe T, Grossart H, Ploug H, Tang K (2002). Mechanisms and rates of bacterial colonization of sinking aggregates. Appl Environ Microbiol 68:3996-4006.

Kiørboe T (2001). Formation and fate of marine snow: Small-scale processes with large-scale implications. Sci Mar 65:57-71.

Kiørboe T, Jackson G (2001). Marine snow, organic solute plumes, and optimal chemosensory behavior of bacteria. Limnol Oceanogr 46:1309-1318.

Kiørboe T (2000). Colonization of marine snow aggregates by invertebrate zooplankton: Abundance, scaling, and possible role. Limnol Oceanogr 45:479-484.

Klaas C, Archer D (2002). Association of sinking organic matter with various types of mineral ballast in the deep sea: Implications for the rain ratio. Global Biogeochem Cycles 16:1116.

Kopczynska EE, Savoye N, Dehairs F, Cardinal D, Elskens M (2007). Spring phytoplankton assemblages in the Southern Ocean between Australia and Antarctica. Polar Biol 31:77-88.

Lancelot C (1983). Factors affecting phytoplankton extracellular release in the southern bight of the North Sea. Mar Ecol Prog Ser 12:115-121.

Laurenceau-Cornec EC, Trull TW, Davies DM, De la Rocha CL, Blain S (2015). Phytoplankton morphology controls on marine snow sinking velocity. Mar Ecol Prog Ser 520:35-56.

Lee C, Murray D, Barber R, Buesseler K, Dymond J, Hedges J , et al. (1998). Particulate organic carbon fluxes: Compilation of results from the 1995 US JGOFS Arabian Sea Process Study. Deep-Sea Res Part II-Top Stud Oceanogr 45:2489-2501.

Lee S, Glickmann E, Cooksey D (2001). Chromosomal locus for cadmium resistance in Pseudomonas putida consisting of a cadmium-transporting ATPase and a MerR family response regulator. Appl Environ Microbiol 67:1437-1444.

Leedjarv A, Ivask A, Virta M (2008). Interplay of different transporters in the mediation of divalent heavy metal resistance in Pseudomonas putida KT2440. J Bacteriol 190:2680-2689.

Legatzki A, Grass G, Anton A, Rensing C, Nies D (2003). Interplay of the Czc system and two P-type ATPases in conferring metal resistance to Ralstonia metallidurans . J Bacteriol 185:4354-4361.

Li X, Nikaido H, Poole K (1995). Role of MexA-MexB-OprM in antibiotic efflux in Pseudomonas aeruginosa . Antimicrob Agents Chemother 39:1948-1953.

Liesegang H, Lemke K, Siddiqui R, Schlegel H (1993). Characterization of the inducible nickel and cobalt resistance determinant cnr from pMOL28 of Alcaligenes eutrophus CH34. J Bacteriol 175:767-778.

Long R, Azam F (2001). Microscale patchiness of bacterioplankton assemblage richness in seawater. Aquat Microb Ecol 26:103-113.

Lovejoy C, Bowman J, Hallegraeff G (1998). Algicidal effects of a novel marine Pseudoalteromonas isolate (class Proteobacteria, gamma subdivision) on harmful algal bloom species of the genera Chattonella , Gymnodinium , and Heterosigma . Appl Environ Microbiol 64:2806-2813.

Ma D, Cook D, Alberti M, Pon N, Nikaido H, Hearst J (1995). Genes acrA and acrB encode a stress-Induced efflux system of Escherichia coli . Mol Microbiol 16:45-55.

Malamud F, Torres PS, Roeschlin R, Rigano LA, Enrique R, Bonomi HR , et al. (2011). The Xanthomonas axonopodis pv. citri flagellum is required for mature biofilm and canker development. Microbiology-(UK) 157:819-829.

108 References

Mari X, Burd A (1998). Seasonal size spectra of transparent exopolymeric particles (TEP) in a coastal sea and comparison with those predicted using coagulation theory. Mar Ecol Prog Ser 163:63-76.

Martin JH, Fitzwater SE, Gordon RM (1990). Iron deficiency limits phytoplankton growth in antarctic waters. Global Biogeochem Cycles 4:5-12.

Martin J, Knauer G, Karl D, Broenkow W (1987). Vertex - carbon cycling in the Northeast Pacific. Deep-Sea Research Part A-Oceanographic Research Papers 34:267-285.

Martin P, van der Loeff MR, Cassar N, Vandromme P, d'Ovidio F, Stemmann L , et al. (2013). Iron fertilization enhanced net community production but not downward particle flux during the Southern Ocean iron fertilization experiment LOHAFEX. Global Biogeochem Cycles 27:871-881.

Martinez-Garcia E, Nikel PI, Chavarria M, de Lorenzo V (2014). The metabolic cost of flagellar motion in Pseudomonas putida KT2440. Environ Microbiol 16 (1):291-303.

McDonnell AMP, Buesseler KO (2010). Variability in the average sinking velocity of marine particles. Limnol Oceanogr 55 (5):2085-2096.

Mergeay M, Nies D, Schlegel H, Gerits J, Charles P, Vangijsegem F (1985). Alcaligenes eutrophus CH34 is a facultative chemolithotroph with plasmid-bound resistance to heavy metals. J Bacteriol 162:328-334.

Mikolay A, Nies DH (2009). The ABC-transporter AtmA is involved in nickel and cobalt resistance of Cupriavidus metallidurans strain CH34. Antonie Van Leeuwenhoek 96:183-191.

Miller TR, Belas R (2006). Motility is involved in Silicibacter sp. TM1040 interaction with dinoflagellates. Environ Microbiol 8:1648-1659.

Minic Z, Serre V, Herve G (2006). Adaptation of organisms to extreme conditions of deep-sea hydrothermal vents. C R Biol 329:527-540.

Mitchell J, Pearson L, Bonazinga A, Dillon S, Khouri H, Paxinos R (1995a). Long lag times and high velocities in the motility of natural assemblages of marine bacteria. Appl Environ Microbiol 61:877-882.

Mitchell J, Pearson L, Dillon S, Kantalis K (1995b). Natural Aasemblages of marine bacteria exhibiting high speed motility and large accelerations. Appl Environ Microbiol 61:4436-4440.

Moche M, Albrecht D, Maass S, Hecker M, Westermeier R, Buettner K (2013). The new horizon in 2D electrophoresis: New technology to increase resolution and sensitivity. Electrophoresis 34:1510-1518.

Monchy S, Vallaeys T, Bossus A, Mergeay M (2006). Metal transport ATPase genes from Cupriavidus metallidurans CH34: A transcriptomic approach. Int J Environ Anal Chem 86:677-692.

Montanini B, Blaudez D, Jeandroz S, Sanders D, Chalot M (2007). Phylogenetic and functional analysis of the cation diffusion facilitator (CDF) family: Improved signature and prediction of substrate specificity. BMC Genomics 8:107.

Moore CM, Mills MM, Arrigo KR, Berman-Frank I, Bopp L, Boyd PW , et al. (2013). Processes and patterns of oceanic nutrient limitation. Nat Geosci 6:701-710.

Moore C, Gaballa A, Hui M, Ye R, Helmann J (2005). Genetic and physiological responses of Bacillus subtilis to metal ion stress. Mol Microbiol 57:27-40.

Mopper K, Zhou J, Ramana K, Passow U, Dam H, Drapeau D (1995). The role of surface-active carbohydrates in the flocculation of a diatom bloom in a mesocosm. Deep-Sea Res Part II-Top Stud Oceanogr 42:47-73.

Munkelt D, Grass G, Nies D (2004). The chromosomally encoded cation diffusion facilitator proteins DmeF and FieF from Wautersia metallidurans CH34 are transporters of broad metal specificity. J Bacteriol 186:8036-8043. 109 References

Myklestad S, Holmhansen O, Varum KM, Volcani BE (1989). Rate of release of extracellular amino-acids and carbohydrates from the marine diatom Chaetoceros affinis . J Plankton Res 11:763-773.

Nelson DM, Treguer P, Brzezinski MA, Leynaert A, Queguiner B (1995). Production and dissolution of biogenic silica in the ocean: Revised global estimates, comparison with regional data and relationship to biogenic sedimentation. Global Biogeochem Cycles 9:359-372.

Ng HJ, Lopez-Perez M, Webb HK, Gomez D, Sawabe T, Ryan J , et al. (2014). Marinobacter salarius sp. nov. and Marinobacter similis sp. nov., isolated from sea water. PLoS One 9:e106514.

Nies D (2003). Efflux-mediated heavy metal resistance in prokaryotes. FEMS Microbiol Rev 27:313-339.

Nies D, Nies A, Chu L, Silver S (1989). Expression and nucleotide sequence of a plasmid-determined divalent cation efflux system from Alcaligenes eutrophus . Proc Natl Acad Sci U S A 86:7351-7355.

Nies D, Silver S (1989). Plasmid-determined inducible efflux is responsible for resistance to cadmium, zinc, and cobalt in Alcaligenes eutrophus . J Bacteriol 171:896-900.

Obernosterer I, Herndl G (1995). Phytoplankton extracellular release and bacterial growth - dependence on the inorganic N:P ratio. Mar Ecol Prog Ser 116:247-257.

Overholt WA, Green SJ, Marks KP, Venkatraman R, Prakash O, Kostka JE (2013). Draft genome sequences for oil-degrading bacterial strains from beach sands impacted by the deepwater horizon oil spill. Genome announcements 1.

Partensky F, Blanchot J, Vaulot D (1999). Differential distribution and ecology of Prochlorococcus and Synechococcus in oceanic waters: A review. Bulletin de l'Institut Oceanographique (Monaco) 19:457-475.

Passow U (2002a). Transparent exopolymer particles (TEP) in aquatic environments. Prog Oceanogr 55:287- 333.

Passow U (2002b). Production of transparent exopolymer particles (TEP) by phyto- and bacterioplankton. Mar Ecol Prog Ser 236:1-12.

Passow U (2000). Formation of transparent exopolymer particles, TEP, from dissolved precursor material. Mar Ecol Prog Ser 192:1-11.

Passow U, Alldredge A (1995). A dye-binding assay for the spectrophotometric measurement of transparent exopolymer particles (TEP). Limnol Oceanogr 40:1326-1335.

Passow U, Carlson CA (2012). The biological pump in a high CO 2 world. Mar Ecol Prog Ser 470:249-271.

Paul C, Pohnert G (2011). Interactions of the algicidal bacterium Kordia algicida with diatoms: Regulated protease excretion for specific algal lysis. PLoS One 6:e21032.

Paulsen I, Park J, Choi P, Saier M (1997). A family of Gram-negative bacterial outer membrane factors that function in the export of proteins, carbohydrates, drugs and heavy metals from Gram-negative bacteria. FEMS Microbiol Lett 156:1-8.

110 References

Pini F, Spini G, Galardini M, Bazzicalupo M, Benedetti A, Chiancianesi M , et al. (2014). Molecular phylogeny of the nickel-resistance gene nreB and functional role in the nickel sensitive symbiotic nitrogen fixing bacterium Sinorhizobium meliloti . Plant Soil 377:189-201.

Ploug H, Terbrueggen A, Kaufmann A, Wolf-Gladrow D, Passow U (2010). A novel method to measure particle sinking velocity in vitro, and its comparison to three other in vitro methods. Limnol Oceanogr Meth 8:386-393.

Ploug H, Iversen MH, Fischer G (2008). Ballast, sinking velocity, and apparent diffusivity within marine snow and zooplankton fecal pellets: Implications for substrate turnover by attached bacteria. Limnol Oceanogr 53:1878-1886.

Pompei M, Mazziotti C, Guerrini F, Cangini M, Pigozzi S, Benzi M , et al. (2003). Correlation between the presence of Gonyaulax fragilis (Dinophyceae) and the mucilage phenomena of the Emilia-Romagna coast (northern Adriatic Sea). Harmful Algae 2:301-316.

Pratt L, Kolter R (1998). Genetic analysis of Escherichia coli biofilm formation: Roles of flagella, motility, chemotaxis and type I pili. Mol Microbiol 30:285-293.

Rabilloud T (2009). Membrane proteins and proteomics: Love is possible, but so difficult. Electrophoresis 30:S174-S180.

Raguenes G, Peres A, Ruimy R, Pignet P, Christen R, Loaec M , et al. (1997). Alteromonas infernus sp. nov., a new polysaccharide-producing bacterium isolated from a deep-sea hydrothermal vent. J Appl Microbiol 82:422- 430.

Ramirez-Diaz MI, Diaz-Perez C, Vargas E, Riveros-Rosas H, Campos-Garcia J, Cervantes C (2008). Mechanisms of bacterial resistance to chromium compounds. Biometals 21:321-332.

Rashed MN (2010). Monitoring of contaminated toxic and heavy metals, from mine tailings through age accumulation, in soil and some wild plants at Southeast Egypt. J Hazard Mater 178:739-746.

Raven J, Falkowski P (1999). Oceanic sinks for atmospheric CO 2. Plant Cell Environ 22:741-755.

Rehm H (2006). Der Experimentator Proteinbiochemie/ Proteomics. Elsevier: München.

Romanenko L, Schumann P, Rohde M, Zhukova N, Mkhailov V, Stackebrandt E (2005). Marinobacter bryozoorum sp. nov. and Marinobacter sediminum sp. nov., novel bacteria from the marine environment. Int J Syst Evol Microbiol 55:143-148.

Sapp M, Schwaderer AS, Wiltshire KH, Hoppe H, Gerdts G, Wichels A (2007a). Species-specific bacterial communities in the phycosphere of microalgae? Microb Ecol 53:683-699.

Sapp M, Wichels A, Gerdts G (2007b). Impacts of cultivation of marine diatoms on the associated bacterial community. Appl Environ Microbiol 73:3117-3120.

Sapp M, Wichels A, Wiltshire KH, Gerdts G (2007c). Bacterial community dynamics during the winter-spring transition in the North Sea. FEMS Microbiol Ecol 59:622-637.

Schäfer H, Abbas B, Witte H, Muyzer G (2002). Genetic diversity of 'satellite' bacteria present in cultures of marine diatoms. FEMS Microbiol Ecol 42:25-35.

Scherer J, Nies DH (2009). CzcP is a novel efflux system contributing to transition metal resistance in Cupriavidus metallidurans CH34. Mol Microbiol 73:601-621.

Schmidt T, Schlegel H (1994). Combined nickel-cobalt-cadmium resistance encoded by the Ncc locus of Alcaligenes xylosoxidans 31a. J Bacteriol 176:7045-7054.

111 References

Schottel J, Mandal A, Clark D, Silver S (1974). Volatilization of mercury and organomercurials determined by inducible R-factor systems in enteric bacteria. Nature 251:335-337.

Schulze D, Kruger A, Kupsch H, Segebade C, Gawlik D (1997). Enrichment and distribution of elements in flood-plain soils of the Bitterfeld industrial area - studied by neutron activation analysis. Sci Total Environ 206:227-248.

Seebah S, Fairfield C, Ullrich MS, Passow U (2014). Aggregation and sedimentation of Thalassiosira weissflogii (diatom) in a warmer and more acidified future ocean. PLoS One 9:e112379.

Shanks A (2002). The abundance, vertical flux, and still-water and apparent sinking rates of marine snow in a shallow coastal water column. Cont Shelf Res 22:2045-2064.

Shanks A, Trent J (1980). Marine snow - sinking rates and potential role in vertical flux. Deep-Sea Research Part A-Oceanographic Research Papers 27:137-143.

Shanks A, Trent J (1979). Marine snow - microscale nutrient patches. Limnol Oceanogr 24:850-854.

Sharma R, Rensing C, Rosen B, Mitra B (2000). The ATP hydrolytic activity of purified ZntA, a Pb(II)/Cd(II)/Zn(II)-translocating ATPase from Escherichia coli . J Biol Chem 275:3873-3878.

Shepherd JG (2012). Geoengineering the climate: An overview and update. Philos Trans A Math Phys Eng Sci 370:4166-4175.

Silver M, Shanks A, Trent J (1978). Marine snow: Microplankton habitat and source of small-scale patchiness in pelagic populations. Science 201:371-373.

Simon M, Grossart H, Schweitzer B, Ploug H (2002). Microbial ecology of organic aggregates in aquatic ecosystems. Aquat Microb Ecol 28:175-211.

Singer E, Webb EA, Nelson WC, Heidelberg JF, Ivanova N, Pati A , et al. (2011). Genomic Potential of Marinobacter aquaeolei , a biogeochemical "opportunitroph". Appl Environ Microbiol 77:2763-2771.

Sison-Mangus MP, Jiang S, Tran KN, Kudela RM (2014). Host-specific adaptation governs the interaction of the marine diatom, Pseudo-nitzschia and their microbiota. ISME J 8:63-76.

Siunova TV, Siunov AV, Kochetkov VV, Boronin AM (2009). The cnr-like operon in strain Comamonas sp. encoding resistance to cobalt and nickel. Russ J Genet 45:292-297.

Smith D, Simon M, Alldredge A, Azam F (1992). Intense hydrolytic enzyme-activity on marine aggregates and implications for rapid particle dissolution. Nature 359:139-142.

Smith S, Hollibaugh J (1993). Coastal metabolism and the oceanic organic carbon balance. Rev Geophys 31:75- 89.

Solioz M, Odermatt A (1995). Copper and silver transport by CopB-ATPase in membrane vesicles of Enterococcus hirae . J Biol Chem 270:9217-9221.

Soria-Dengg S, Reissbrodt R, Horstmann U (2001). Siderophores in marine, coastal waters and their relevance for iron uptake by phytoplankton: Experiments with the diatom Phaeodactylum tricornutum . Mar Ecol Prog Ser 220:73-82. 112 References

Stahl A, Pletzer D, Mehmood A, Ullrich MS (2015). Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification. Antonie Van Leeuwenhoek 108:649-658.

Steele J, Thorpe S, Turekian K (2001). Nutrient Cycling: The Carbon Cycle. Encyclopedia of Ocean Science Academic Press .

Stein R (1991). Accumulation of Organic Carbon in Marine Sediments. Springer: Berlin.

Stocker R (2012). Marine microbes see a sea of gradients. Science 338:628-633.

Stocker R (2011). Reverse and flick: Hybrid locomotion in bacteria. Proc Natl Acad Sci U S A 108:2635-2636.

Stoderegger K, Herndl G (1999). Production of exopolymer particles by marine bacterioplankton under contrasting turbulence conditions. Mar Ecol Prog Ser 189:9-16.

Suess E (1980). Particulate organic carbon flux in the oceans - surface productivity and oxygen utilization. Nature 288:260-263.

Suttle C (1994). The significance of viruses to mortality in aquatic microbial communities. Microb Ecol 28:237- 243.

Suttle CA (2007). Marine viruses - major players in the global ecosystem. Nat Rev Microbiol 5:801-812.

Symmons MF, Bokma E, Koronakis E, Hughes C, Koronakis V (2009). The assembled structure of a complete tripartite bacterial multidrug efflux pump. Proc Natl Acad Sci U S A 106:7173-7178.

Takahashi K, Fujitani N, Yanada M, Maita Y (2000). Long-term biogenic particle fluxes in the Bering Sea and the central subarctic Pacific Ocean, 1990-1995. Deep Sea Res Part 1 Oceanogr Res Pap 47 (9):1723-1759.

Taktikos J, Stark H, Zaburdaev V (2014). How the motility pattern of bacteria affects their dispersal and chemotaxis. PLoS One 9:e92348.

Tang YZ, Koch F, Gobler CJ (2010). Most harmful algal bloom species are vitamin B 1 and B 12 auxotrophs. Proc Natl Acad Sci U S A 107:20756-20761.

Tomihama T, Nishi Y, Arai K (2006). The effects of exopolysaccharide production and flagellum on motility, biofilm formation, survival and virulence of Pseudomonas syringae pv. theae. Japanese Journal of Phytopathology 72:3-13.

Tong L, Nakashima S, Shibasaka M, Katsuhara M, Kasamo K (2002). A novel histidine-rich CPx-ATPase from the filamentous cyanobacterium Oscillatoria brevis related to multiple-heavy-metal cotolerance. J Bacteriol 184:5027-5035.

113 References

Trent J, Shanks A, Silver M (1978). In situ and laboratory measurements on macroscopic aggregates in Monterey Bay, California. Limnol Oceanogr 23:626-635.

Uebe R, Junge K, Henn V, Poxleitner G, Katzmann E, Plitzko JM , et al. (2011). The cation diffusion facilitator proteins MamB and MamM of Magnetospirillum gryphiswaldense have distinct and complex functions, and are involved in magnetite biomineralization and magnetosome membrane assembly. Mol Microbiol 82:818-835.

Vandecandelaere I, Nercessian O, Segaert E, Achouak W, Mollica A, Faimali M , et al. (2008). Alteromonas genovensis sp. nov., isolated from a marine electroactive biofilm and emended description of Alteromonas macleodii Baumann et al . 1972 (Approved Lists 1980). Int J Syst Evol Microbiol 58:2589-2596.

Varnavas S, Cronan D (2005). Submarine hydrothermal activity off Santorini and Milos in the Central Hellenic Volcanic Arc: A synthesis. Chem Geol 224:40-54.

Verdugo P, Alldredge A, Azam F, Kirchman D, Passow U, Santschi P (2004). The oceanic gel phase: A bridge in the DOM-POM continuum. Mar Chem 92:67-85.

Verity P, Stoecker D, Sieracki M, Burkill P, Edwards E, Tronzo C (1993). Abundance, biomass and distribution of heterotrophic dinoflagellates during the North Atlantic spring bloom. Deep-Sea Res Part II-Top Stud Oceanogr 40:227-244. von Scheibner M, Doerge P, Biermann A, Sommer U, Hoppe H, Juergens K (2014). Impact of warming on phyto-bacterioplankton coupling and bacterial community composition in experimental mesocosms. Environ Microbiol 16:718-733.

Wang H, Li H, Shao Z, Liao S, Johnstone L, Rensing C , et al. (2012). Genome sequence of deep-sea manganese- oxidizing bacterium Marinobacter manganoxydans MnI7-9. J Bacteriol 194:899-900.

Wolff S, Otto A, Albrecht D, Zeng J, Buttner K, Gluckmann M , et al. (2006). Gel-free and gel-based proteomics in Bacillus subtilis - A comparative study. Mol Cell Proteomics 5:1183-1192.

Xie L, Altindal T, Chattopadhyay S, Wu X (2011). Bacterial flagellum as a propeller and as a rudder for efficient chemotaxis. Proc Natl Acad Sci U S A 108:2246-2251.

Yoon J, Kim I, Kang K, Oh T, Park Y (2003). Alteromonas marina sp. nov., isolated from sea water of the East Sea in Korea. Int J Syst Evol Microbiol 53:1625-1630.

Zeebe RE, Wolf-Gladrow A (2001). CO 2 in Seawater: Equilibrium, kinetics and isotopes. Elsevier: Amsterdam.

Zhang D, Liu Y, Li X (2015). Characterization of bacterial diversity associated with deep sea ferromanganese nodules from the South China Sea. J Microbiol 53:598-605.

Zhou J, Mopper K, Passow U (1998). The role of surface-active carbohydrates in the formation of transparent exopolymer particles by bubble adsorption of seawater. Limnol Oceanogr 43:1860-1871.

114 Supplementary – Tables and Figures

6. Supplementary – Tables and Figures

Supplementary 1: Cluster Analysis of genomic region encoding down-regulated alcohol and aldehyde dehydrogenases. Grey gene products were found to be down-regulated in M. adhaerens HP15 during co- cultivation with diatom T. weissflogii . 4: aldehyde dehydrogenase (ADP98908.1) 20 and 24 : alcohol dehydrogenases (ADP98924.1 and ADP98928.1). See details on further genes in Supplementary 2.

115 Supplementary – Tables and Figures

Supplementary 2: Genes and gene clusters in close proximity of the down-regulated alcohol and aldehyde dehydrogenases. Down regulated gene products are shown in grey (see also Supplementary 1 ).

Gene Gene annotation Accession Cluster number

1 transcriptional regulator, TetR family ADP98905.1 I

2 quinone oxidoreductase, YhdH/YhfP family ADP98906.1 I

3 uncharacterized peroxidase-related protein ADP98907.1 I

4 NAD-dependent aldehyde dehydrogenase ADP98908.1 I

5 hypothetical protein ADP98909.1 -

6 urea carboxylase-associated protein 2 ADP98910.1 II

7 urea carboxylase-associated protein 1 ADP98911.1 II

8 hypothetical protein ADP98912.1 -

9 urea amidolyase-like protein/ urea carboxylase ADP98913.1 II

10 ABC transporter, periplasmic substrate-binding protein / NitT/TauT ADP98914.1 II family transport system substrate-binding protein

11 transcriptional regulator, TetR family ADP98915.1 II

12 regulatory protein ada ADP98916.1 -

13 protein containing cyclic nucleotide-binding domain ADP98917.1 III

14 class II aldolase/adducin family protein ADP98918.1 III

15 metal dependent phosphohydrolase ADP98919.1 III

16 glutamate-ammonia ligase/ glutamine synthetase ADP98920.1 IV

17 metX, homoserine O-acetyltransferase ADP98921.1 IV

18 binding-protein-dependent transport systems inner membrane ADP98922.1 IV component

19 spermidine/putrescine ABC transporter ATPase ADP98923.1 IV

20 quinoprotein alcohol dehydrogenase ADP98924.1 V

21 extracellular solute-binding protein, family 3 ADP98925.1 V

22 cytochrome c550 ADP98926.1 V

23 hypothetical protein ADP98927.1 -

24 LOW QUALITY PROTEIN: quinoprotein alcohol dehydrogenase ADP98928.1 VI

25 pentapeptide repeat family protein ADP98929.1 VI

26 two component transcriptional regulator, LuxR family ADP98930.1 VI

27 integral membrane sensor signal transduction histidine kinase ADP98931.1 VI

28 conserved hypothetical protein, secreted ADP98932.1 VI

29 40-residue YVTN family beta-propeller repeat-containing protein ADP98933.1 VI

30 ABC efflux transporter, ATP-binding protein ADP98934.1 VI

31 ABC efflux transporter, permease protein ADP98935.1 VI

116 Supplementary – Tables and Figures

Supplementary 3: Oligosaccharide and polyol transporters [section ABC transporter] in M. adhaerens HP15. Uptake systems identified in M. adhaerens HP15 according to KEGG algorithms are shown in bold (none of the corresponding transporter involved proteins shown in this table was found).

Substrate Corresponding proteins involved in substrate uptake

Maltose/ Maltodextrin MalE, MalF, MalG, MalK

Maltoologosaccharide MalE, MalF, MalG, MalK

Multibel sugar MsmE, MsmF, MsmG, MsmK

Cyclodextrin GanO, GanP, GanQ, MsmX

Sn-Glycerol 3-phosphate UgpB, UgpA, UgpE, UgpC

Lactose/ L-arabinose LacE, LacF, LacG, LacK

Sorbitol/ Mannitol SmoE, SmoF, SmoG, SmoK

α-Glucoside AglE, AglF, AglG, AglK

Oligogalacturonide TogB, TogM, TogN, TogA

Glucose/ Arabinose GlcS, GlcU, GlcT, GlcV

Trehalose/ Maltose ThuE, ThuF, ThuG, ThuK

N-Acetylglucosamin NgcE, NgcF, NgcG, [?]

Cellobiose CebE, VebF, CebG, MsiK

Multibel Sugar? ChvE, GguB, GguA

117 Supplementary – Tables and Figures

Supplementary 4: Monosaccharide transporters [section ABC transporter] in M. adhaerens HP15. Uptake systems identified in M. adhaerens HP15 according to KEGG algorithms are shown in bold (none of the corresponding transporter involved proteins shown in this table was found).

Substrate Corresponding proteins involved in substrate uptake

Ribose RbsB, RbsC, RbsD, RbsA

L-Arabinose AraF, AraH, AraG

Methyl-galactoside MglB, MglC, MglA

D-Xylose XylF, XylH, XylG

D-Allose AlsB, AlsC, AlsA

Fructose FrcB, FrcC, FrcA

Autoinducer 2 LsrB, LsrC, LsrD, LsrA

Rhamnose RhaS, RhaP, RhaQ, Rhat

118 Supplementary – Tables and Figures

Supplementary 5: Phosphotransferase systems (PTS) identified in M. adhaerens HP15 according to KEGG. Those systems identified in M. adhaerens HP15 according to KEGG algorithms are shown in bold .

Substrate Corresponding proteins involved in substrate utilization

Glc family

Glucose → Glucose 6-phosphate PtsG, Crr

N-Acetyl-D-glucosamin → N-Acetyl-D-glucosamin 6-phosphate NagE

Maltose → Maltose 6-phosphate MalX, Crr

D-Glucosamin → D-Glucosamin 6-phosphate GamP

Sucrose → Sucrose 6-phosphate ScrA

β-Glucosides → Phospho-β-glucosides BglF

Arbutin/ Salicin → Arbutin/ Salicin 6-phosphate AscF, Crr

Trehalose →Trehalose 6-phosphate TreB, TreP/Crr

N-acetyl-muramic acid → N-acetyl-muramic acid 6-phosphate MurP, Crr

Arbutin →Arbutin 6-phosphate GlvC, Crr, GlvB

Lac family

Lactose -> Lactose 6-phosphate LacE, LacF

Cellobiose -> […] CelB, CelC, CelA

Fru family

Mannitol ->Mannitol 1-phosphate MtlA

2-O-α-Mannosyl-D-glycerate -> 2-O-(6-Phosphate-α-mannosyl)-D- MngA glycerate

Fructose -> Fructose 1-phosphate FruA , FruB

Man family

Mannose -> Mannose 6-phosphate ManY, ManZ, ManX

Fructose -> Fructose 1-phosphate LevF, LevG, LevD, LevE

Sorbose ->Sorbose 1-phosphate SorA, SorM, SorF, SorB

N-Acetyl-galactosamine -> N-Acetyl-galactosamine 6- phosphate AgaW, AgaE, AgaF, AgaV

Galactosamine ->Galactosamin 6-phosphate AgaC, AgaE, AgaF, AgaB

Other family

Sorbitol -> Sorbitol 6-phosphate SrlA/SrlE, SrlB, SrlE

Galactilol ->Galactilol 1-phosphate GatC, GatA, GatB

L-Ascorbate -> L-Ascorbate 6-phosphate UlaA, UlaC, UlaB

Nitrogen regulation

Nitrogen -> […] […], PtsN

Phosphpenol-pyruvate -> pyruvate (inner cell) PtsP

119 Supplementary – Tables and Figures

Supplementary 6: Bacterial strains and plasmids used in the study ‘Stereo-tracking of chemosensing-deficient and motility-impaired Marinobacter adhaerens HP15 strains during marine particle colonization – a novel methodical approach’ (Results/ Part II).

Strain or plasmid Characterization Reference

Bacteria

Escherichia coli

DH5 α subE44 ∆lacU169 (ɸlacZ ∆M15) hsdR17 Sambrook & Russel, 2001 recA1 endA1 gyrA96 thi-1 relA1

ST18 λpir ∆hemA pro thihsd R+ TprSmr Thoma & Schobert, 2009 chromosome::RP4-2 Tc::Mu-Kan::Tn7

Marinobacter adhaerens

HP15 wild type Grossart et al. , 2004

HP15_ ∆cheA deletion mutagenesis lacking histidine Sonnenschein et al. , 2012 kinase CheA, involved in chemosensing system of flagellum control, Cm R

HP15-chpA ::Tn5 Tn5 insertion mutagenesis lacking Sonnenschein et al. , 2012 histidine kinase ChpA, involved in chemosensing system of type VI pilus function (twitching), Kan R

HP15_ ∆fliC deletion mutagenesis lacking flagellum Sonnenschein et al ., 2011 protein FliC (motility deficient), Cm R

HP15-fliG ::Tn5 Tn5 insertion mutagenesis lacking Sonnenschein et al. , 2011 flagellum protein FliG (motility deficient), & 2012 Kan R

HP15_ ∆mshB deletion mutagenesis lacking type IV pilus Seebah, 2012 protein MshB (attachment deficient), Cm R

Plasmids

pBBR1MCS-4 broad host range cloning vector, Ap R Kowach et al. , 1994

pBBR-4-ecfp ecfp under control of constitutive This study R R LacZ promoter PA1/04/03 , Ap , Cm

pBBR-4-eyfp eyfp under control of constitutive This study R R LacZ promoter P A1/04/03 , Ap , Cm

pBBR-4-DsRed DsRed under control of constitutive This study R R LacZ promoter P A1/04/03 , Ap , Cm

pBluescript II SK broad host range cloning vector, Ap R Stratagene

pBlue-ecfp intermediate plasmid, ecfp under control This study of constitutive LacZ promoter P A1/04/03 , Ap R, Cm R

pBlue-eyfp intermediate plasmid, eyfp under control This study of constitutive LacZ promoter P A1/04/03 , Ap R, Cm R

pBlue-DsRed intermediate plasmid, DsRed under control This study of constitutive LacZ promoter P A1/04/03 , Ap R, Cm R

R R R miniTn7(Gm) PA1/04/03 ecfp -a source of ecfp gene, Ap , Cm , Gm Lambertsen et al ., 2004

R R R miniTn7(Gm) PA1/04/03 eyfp -a source of eyfp gene, Ap , Cm , Gm Lambertsen et al ., 2004

R R miniTn7(Gm) PA1/04/03 source of DsRedExpress gene, Ap , Cm , Lambertsen et al ., 2004 DsRedExpress -a Gm R

Ap = Ampicillin; Cm = Chloramphenicol, Gm = Gentamycin; Sm = Streptomycin; Tc = Tetracycline; Tp = Trimethoprim

120 Supplementary – Tables and Figures

Supplementary 7: Bacterial strains and plasmids used in the study ‘Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification’ (Results/ Part III).

Strain or plasmid Characterization Reference or source

Bacteria

Escherichia coli

DH5 α EndA I hsdR I [r K-mK+] glnV44 thi- I recA I gyrA relA Piekarski et al. 2009 ∆[lacZYA-argF )U I 6 9 deoR [ɸ80 dlac ∆[lacZ ]M I 5)

ST18 pro thi hsdR + Tp r Sm r ; chromosome::RP4-2 Tc::Mu- Thoma and Schobert Kan::Tn7/ λpir ∆hemA 2009

Marinobacter adhaerens

HP15 wild type Grossart et al. 2004

HP15. ∆czcCBA.1 deletion mutant of gene loci HP15_108-110 (operon This study czcCBA. 1)

HP15. ∆czcCBA.2 deletion mutant of gene loci HP15_113-115 (operon This study czcCBA. 2)

HP15. ∆czcCBA.1/2 deletion mutant of gene loci HP15_108-110 and This study HP15_113-115 ( czcCBA. 1/2 double mutant)

Plasmids

pBBR1MCS-1 cat (Cm R), broad host range cloning vector Kovach et al. 1994

pBBR.czcCBA.1 cat (Cm R) , carries czcCBA. 1 operon under control of This study LacZ promoter, used for complementation of operon deletion

pBBR.czcCBA.2 cat (Cm R), carries czcCBA. 2 operon under control of This study LacZ promoter, used for complementation of operon deletion

pEX18Ap bla (Ap R), oriT +, sacB + Hoang et al ., 1998

pEX18Ap.czcCBA.1_ko bla (Ap R), cat (Cm R), carries mutagenic construct This study cassette for mutagenesis of operon czcCBA. 1 of M. adhaerens HP15

pEX18Ap.czcCBA.2_ko bla (Ap R), cat (Cm R), carries mutagenic construct This study cassette for mutagenesis of operon czcCBA. 2 of M. adhaerens HP15

pFCM1 source of Cm R cassette Hoang et al. , 1998

pFLP2 bla (Ap R), FLP +, sacB +, used for removal of Cm R Hoang et al. , 1998 cassette

R pJET1.2/blunt bla (Ap ), promoter P lacUV5 , blunt end multiple cloning Thermo Fisher vector Scientific

pJET.czcCBA.1 bla (Ap R), carries czcCBA. 1 operon under control of This study promoter P lacUV5 , intermediate plasmid for generation of pBBR.czcCBA.1

pJET.czcCBA.2 bla (Ap R), carries czcCBA. 2 operon under control of This study promoter P lacUV5 , intermediate plasmid for generation of pBBR.czcCBA.2

pJET.czcCBA.1_ko bla (Ap R), cat (Cm R), carries mutagenic construct This study cassette for mutagenesis of czcCBA. 1 operon in M. adhaerens HP15, intermediate plasmid for generation of pEX18Ap.czcCBA.1_ko

pJET.czcCBA.2_ko bla (Ap R), cat (Cm R), carries mutagenic construct This study cassette for mutagenesis of czcCBA. 2 operon in M. adhaerens HP15, intermediate plasmid for generation of pEX18Ap. czcCBA.2_ko

Ap = Ampicillin; Cm = Chloramphenicol, Kan = Kanamycin; Sm = Streptomycin; Tc = Tetracycline; Tp = Trimethoprim

121 Supplementary – Tables and Figures

Supplementary 8: Oligonucleotide primers for mutant generation and verification used in the study ‘Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification’ (Results/ Part III).

Primer notation Sequence 5‘ - 3‘

Primers for generation of mutagenic constructs of deletion mutagenesis of operon czcCBA.1 and czcCBA.2

czcCBA.1_A1 GAGCTTGAGCTGAGTGGC czcCBA.1_A2 CCCTATAGTGAGTCGGTACC czcCBA.1_B1 GGTACCGACTCACTATAGGG czcCBA.1_B2 AACACCACCAGCGTCAGC

czcCBA.2_A1 AGGCAAGCAAGCACTCGC czcCBA.2_A2 CCCTATAGTGAGTCGGTACC czcCBA.2_B1 GGTACCGACTCACTATAGGG czcCBA.2_B2 GAAGCGCGTGAATAGAGCC

Primers for confirmation of mutagenesis

cat_2 CTTACGTGCCGATCAACG cat_3 AGCATTCATCAGGCGGGC cat_4 ACAAGGTGCTGATGCCGC cat_5 GTGATGGCTTCCATGTCG

czcCBA.1_out1 TGTCCATGCAGGAAGCTG czcCBA.1_out2 CCTTCTCCTGCCTATCAG czcCBA.1_out3 AGTACCAGCAAGCCAAGG czcCBA.1_out4 ATCAACCACATGCTCCGC

czcCBA.2_out1 ACTTCCTACCTCGTCACG czcCBA.2_out2 TATGACGGAGGGACTACC czcCBA.2_out3 AAGCTTATGTCGCGCAGG czcCBA.2_out4 TCCTTTCCCAGAGTAGAGC

Primers for construction and verification of complementation plasmids pBBR.czcCBA.1 and pBBR.czcCBA.2

czcCBA.1_for_KpnI CTAGGTACCATGCAGGAAGCTGACCTCGAT czcCBA.1_rev_BamHI TCAGGATCCTCAGACTTCAACGTCACGACG

czcCBA.2_for_BamHI TCAGGATCCATGATGAAGGCAAGCAAGCAC czcCBA.2_rev_XbaI GCTCTAGACTACCGTTGCATATCATCGAGTTTAGC

czcCBA.1_seq1 GCCTCGGCTCAGTTGAAAAT czcCBA.1_seq2 CGTTGAAAGCTCGGAAAGAAG czcCBA.1_seq3 GGAGCATGATCAACGCAATTGT czcCBA.1_seq4 AAGGATGTGGCGGAGGTTGG czcCBA.1_seq5 CGTTGCTTTTGCAGGCAGT

czcCBA.2_seq1 ACGCAAAGCCACAGCATCG czcCBA.2_seq2 TGAAATTTACGCTCCCGG czcCBA.2_seq3 AACTGGGCCGCTCAATTGG czcCBA.2_seq4 TTGGGGGCTGGTATATGGG czcCBA.2_seq5 GTTTTCATGCCTCTGTTCAGCT

122 Supplementary – Tables and Figures

Supplementary 9: Oligonucleotide primers for mRNA quantification for czcCBA. 1 and czcCBA. 2 (Study ‘Marinobacter adhaerens HP15 harbors two CzcCBA efflux pumps involved in zinc detoxification’ (Results/ Part III)).

Primer notation Sequence 5‘ - 3‘

Primer pairs for potential constantly expressed genes used for data normalization

rRNA8_qRT_for AGTGCGAATGCTGACATGAG rRNA8_qRT_rev TAGCCTTCTCCGTCTCTCCA

gyrA_qRT_for GTGAAGGGTTTGCACCAGAT gyrA_qRT_rev CCTTGGGTTTGTCTTCGGTA

recA_qRT_for CAAACACGCTAACTGCCTGA recA_qRT_rev GACCACCTCGTCACCATCTT

Primer pairs for gene operon czcCBA .1 and czcCBA .2

czcC.1_qRT_for CCAGGTCGTGTCTGCTGTAG czcC.1_qRT_rev GTCGTGTTGCACTTGGACAT

czcC.2_qRT_for GTTGGCCTGACATTCTCGAT czcC.2_qRT_rev GCTTGTCGTATCGCTTCCAG

123 Supplementary – Tables and Figures

Supplementary 10 : Complementation of M. adhaerens HP15. ∆czcCBA.2 with plasmid pBBR.czcCBA.2. a M. adhaerens HP15 wild type + pBBR1MCS-1 (reference) b HP15. ∆czcCBA.2 + pBBR1MCS-1 (negative control) c HP15. ∆czcCBA.2 + pBBR.czcCBA.2 (complementation). 10 µl of bacterial suspension (OD 600 = 0.001) were spotted on agar plates supplemented with 1.0 mM zinc and left for drying. One side of a square indicates about 15 mm width.

124 Supplementary – Tables and Figures

Supplementary 11 : Nucleotide BLAST results for Marinobacter strains harboring two czcCBA operons. BLAST search results of the full czcCBA nucleotide region (including the genes located between the operons, ranging from gene locus czcC _108 to czcA _115, nucleotide position 111,522 – 126,022) encoded by M. adhaerens HP15 and the 16S rRNA gene BLAST results are given. The genome of M. manganoxydans MnI7-9 could not be used for the full region BLAST, as alignments were located on different contigs of the shot-gun genome sequence.

Marinobacter strain czcCBA region BLAST 16S rRNA BLAST

query query identity [%] identity [%] coverage [%] coverage [%]

M. adhaerens HP15 100 100 100 100 (positive control)

M. salarius R9SW1 95 89 100 98

M. similis A3d10 98 95 100 98

M. algicola DG893 a 80 95 100 98

M. manganoxydans a on separate contigs 94 99

M. nanhaiticus D15-8W a 80 95 100 97

M. sp. MCTG268 a 95 85 99 99

a shot-gun genomes

125 Supplementary – Tables and Figures

Supplementary 12 : Stereo-attachment of M. adhaerens HP15 wild type and chemotaxis mutants to agar spheres. Bacterial cells were grown to exponential phase in liquid MB medium, harvested by mild centrifugation, and washed with sterile North Sea water (NSW). Cells were further resuspended in NSW and starved over night. Equal amounts of cells were mixed (10 6 cells mL -1) and exposed to MB agar spheres. Attachment of cells was documented after 2 h of incubation. M. adhaerens HP15 wild type was labeled with plasmid pBBR-4-DsRed (shown in red, lane 1 ), chemotaxis mutants were labeled with plasmid pBBR-4-eyfp (shown in green, lane 2 ). The overlap of both emissions is given in lane 3 . a M. adhaerens HP15 wild type + M. adhaerens HP15 wild type (control) b M. adhaerens HP15 wild type + motility mutant ∆fliC , lacking the main part of the flagellum c M. adhaerens HP15 wild type + chemosensing mutant ∆cheA, lacking the histidine kinase CheA needed for response regulation of the flagellum.

126 Statutory Declaration

Statutory Declaration on authorship of the present dissertation

I, Antje Stahl hereby declare that I have written this PhD thesis independently, unless where clearly stated otherwise. I have used only the sources, the data, and the support that I have clearly mentioned. This PhD thesis has not been submitted for conferral of degree elsewhere.

I confirm that no rights of third parties will be infringed by the publication of this thesis.

Place & Date Signature

127

128