Struktur und Dynamik heterotropher Bakteriengemeinschaften im Wattenmeer und der Deutschen Bucht

Structure and dynamics of heterotrophic bacterial communities in the German Wadden Sea and the German Bight

Dissertation

zur Erlangung des akademischen Grades einer

Doktorin der Naturwissenschaften (Dr. rer. nat.)

der Fakultät V Mathematik und Naturwissenschaften der Carl von Ossietzky Universität Oldenburg

vorgelegt von

Beate Rink

geboren am 23.01.1974 in Bremerhaven

Erstgutachter : Prof. Dr. Meinhard Simon Zweitgutachter: Prof. Dr. Heribert Cypionka

Eingereicht am: Disputation am:

Für Rosemarie

Erklärung

Teilergebnisse dieser Arbeit sind als Beiträge bei den genannten Fachzeitschriften eingereicht oder werden eingereicht. Mein Beitrag an der Erstellung der verschiedenen Manuskripte wird im Folgenden erläutert:

Rink, B. , Seeberger, S., Martens, T., Duerselen, C. D., Simon, M., und Brinkhoff, T. (2006) Effects of a phytoplankton bloom in a coastal ecosystem on the composition of bacterial communities (Eingereicht bei Aquat. Microb. Ecol. )

Etablierung und Spezifitätstest der Roseobacter spezifischen PCR durch S. S. unter Anleitung von B. R. und T. B (Diplomarbeit, 2003). Durchführung der spezifischen PCR und DGGE, der Klonierung und Sequenzierung durch B. R. Statistische Auswertung und Erstellung der phylogenetischen Stammbäume durch B. R. Erstellung der ersten Fassung des Manuskripts durch B. R., Überarbeitung durch T. B., B. R. und M. S.

Rink, B., Martens, T., Fischer, D., Lemke, A., Grossart, H. P., Simon, M., und Brinkhoff, T. (2006) Tidal effects on coastal bacterioplankton (In Vorbereitung zum Einreichen bei Limnol. Oceanogr. )

Planung und Durchführung der Probenahme 2005 durch B. R. Durchführung der spezifischen PCR und DGGE sowie der RNA Untersuchungen und CARD-FISH durch B. R. Statistische Auswertung und Erstellung der phylogenetischen Stammbäume durch B. R. Erstellung der ersten Fassung des Manuskripts durch B. R., Überarbeitung durch T. B., B. R. und M. S.

Stevens, H., Brinkhoff, T., Rink, B. , Vollmers, J., und Simon, M. (2006) Diversity and abundance of Gram-positive bacteria in a tidal flat ecosystem (Eingereicht bei Environ. Microbiol. )

Durchführung der spezifischen CARD-FISH und DGGE Untersuchungen von J. V. unter Anleitung von B. R. und T. B (Leistungsnachweis, 2005). Überarbeitung des Manuskriptes von B. R., T. B. und M. S.

Rink, B. , Brinkhoff, T., Ziegelmüller, K., und Simon, M. (2006) High regional variability of bacterial communities in the German Bight, North Sea (Eingereicht bei Aquat. Microb. Ecol. )

Planung und Durchführung der Probenahme 2002 von Mirko Lunau und B. R. Planung und Durchführung der Probenahme 2003 von B. R. Molekularbiologische Untersuchungen (PCR, DGGE, Klonierung, Sequenzierung), statistische Auswertung und Erstellung der phylogenetischen Stammbäume durch B. R. Erstellung der ersten Fassung des Manuskripts durch B. R., Überarbeitung durch T. B., B. R. und M. S.

Tagungsbeiträge

Rink B , Stevens H, Simon M, Brinkhoff T (2006) Stability of Microbial Communities Within Different Time Scales in a Tidal Flat Ecosystem. Posterbeitrag, International Symposium Microbial Ecology (ISME-11), Wien, Österreich, 20-25 August

Rink B , Brinkhoff T, Simon M (2004) Bacterial communities reflect different regional properties of the German Bight. Vortrag, VAAM-Meeting Braunschweig, 28-31 March

Rink B , Kruse M, Seeberger S, Stevens H, Brinkhoff T, Simon M (2004) Seasonal and spatial differences in the composition and abundance of bacterial communities in the German Bight of the North Sea. Posterbeitrag, International Symposium Microbial Ecology (ISME-10), Cancun, Mexico, 22-27 August

Simon M, Selje N, Schledjewski R, Rink B , Grossart HP (2004) Diversity and substrate turnover of bacterioplankton communities in the Gulf of Aqaba, Red Sea. Posterbeitrag, International Symposium Microbial Ecology (ISME-10), Cancun, Mexico, 22-27 August

Rink B , Lunau M, Seeberger S, Stevens H, Brinkhoff T, Grossart H-P, Simon M (2003) Diversity patterns of aggregate-associated and free-living bacterial communities in the German Wadden Sea. In Rullkötter J. (ed.), BioGeoChemistry of Tidal Flats - Proceedings of a Workshop held at the Hanse Institute of Advanced Study, Delmenhorst (Germany), 14- 17 May. Forschungszentrum Terramare, Wilhelmshaven, Berichte Nr. 12, 96-98. ISSN 1432-797X.

Lunau M, Rink B , Grossart H-P, Simon M (2003) How to sample marine microaggregates in shallow and turbid environments? - Problems and solutions. In Rullkötter J. (ed.), BioGeoChemistry of Tidal Flats - Proceedings of a Workshop held at the Hanse Institute of Advanced Study, Delmenhorst (Germany), 14-17 May. Forschungszentrum Terramare, Wilhelmshaven, Berichte Nr. 12, 85-88. ISSN 1432-797X.

Rink B , Brinkhoff T, Simon M (2002) Completing the picture of natural habitats: The use of specific Primersets in DGGE. Posterbeitrag, VAAM Meeting, Berlin, 23-26 March

Zusammenfassung

Im Wattenmeer unterliegen die Organismen hochdynamischen Prozessen. Eine flache Wassersäule und der Einfluß der Gezeiten sorgen für starke Strömungen und hohe Resuspensionsraten. Auch der tidale Ein- und Ausstrom von Wassermassen aus der Nordsee in das Rückseitenwatt beeinflusst das System. Während in Herbst- und Wintermonaten sedimentologische Faktoren überwiegen, ist im Frühjahr und Sommer ein deutlicher Einfluss biologischer Größen nachweisbar. Im Rahmen des interdisziplinären Forschungsprojekts „Biogeochemie des Watts“, in das diese Arbeit eingebunden ist, wurden große Varianzen innerhalb des Schwebstoffaufkommens sowie in bakterieller Aktivität und Abundanz auf saisonaler Ebene sowie im Tidenzyklus beschrieben. In der vorliegenden Arbeit wurde untersucht, inwiefern tidale und saisonale Faktoren die Struktur der ansässigen Bakteriengemeinschaften in der Wassersäule beeinflussen. Weiterführend wurde untersucht, ob die im Wattenmeer detektierten Phylotypen standortspezifisch oder auch in anderen Gebieten der Deutschen Bucht nachweisbar sind. Im Wattenmeer fand die Beprobung in der Otzumer Balje im Rückseitenwatt von Spiekeroog statt. Im ersten Teil dieser Arbeit wurden zur Untersuchung des Zusammenhangs von Bakteriengemeinschaften und Phytoplankton wöchentlich Proben genommen und mittels gruppenspezifischer DGGE (Denaturierende Gradienten Gelelektrophorese) und statistischer Methoden untersucht. Im zweiten Teil wurden neben saisonalen auch tidale Vorgänge beleuchtet. Die Probenahme fand im Herbst, Frühjahr und Sommer in einstündigem und dreistündigem Probenahmeraster statt. Die Bakteriengemeinschaften wurden mittels gruppenspezifischer DGGE für alpha-Proteobakterien , Bacteroidetes und Roseobacter sowohl DNA- als auch RNA basiert untersucht. Zusätzlich wurden FISH (Fluoreszens in situ Hybridisierung) und die hoch sensitive CARD-FISH (Catalyzed Reporter Deposition-FISH) eingesetzt und somit erstmalig die Abundanzen einzelner Bakteriengruppen in der Wassersäule des Wattenmeeres dargestellt. In einer vorangegangenen Arbeit wurden im Wattenmeer bemerkenswert viele gram-positive Bakterien isoliert, was zu der Annahme führte, dass diese Bakteriengruppe eine besondere Stellung in diesem Habitat einnimmt. Zur Vervollständigung der Daten wurde im dritten Teil dieser Arbeit die CARD-FISH eine Actinobakterien -spezifische Sonde eingesetzt und zusätzlich eine spezifische DGGE entwickelt, um Abundanz und phylogenetische Vielfalt der Actinobakterien im Watt zu untersuchen. Die Probenahme hierzu wurde an verschiedenen Standorten im Spiekerooger Rückseitenwatt durchgeführt.

Im vierten Teil wurden im Sommer 2002 und 2003 verschiedene Standorte der Deutschen Bucht an der Küstenzone, vor Helgoland und in der offenen Nordsee beprobt. Die Bakteriengemeinschaften wurden mit spezifischer DGGE für alpha-Proteobakterien und Bacteroidetes untersucht. Zur weiteren Beschreibung der Ökologie an den untersuchten Standorten wurden zusätzlich hydrologische, mikrobiologische und partikuläre Parameter bestimmt. Zusammenfassend ergaben sich aus diesen Arbeiten folgende Hauptaussagen: • Im Wattenmeer sind die Bakteriengemeinschaften in der Wassersäule im Wesentlichen aus alpha- und gamma-Proteobakterien sowie Bacteroidetes zusammengesetzt. Darüber hinaus sind beta-Proteobakterien abundant auf Aggregaten. Hierbei bilden frei lebende und Aggregat-assoziierte Bakterien distinkte Gemeinschaften sowohl im Wattenmeer als auch in der Deutschen Bucht. Die Struktur der frei lebender Bakteriengemeinschaften besteht hauptsächlich aus wenigen dominanten Phylotypen der Roseobacter Gruppe. Ihre Zusammensetzung ist saisonal und räumlich stabil. Die Struktur der Aggregat-assoziierten Bakterien zeigt grössere Artenvielfalt als bei frei lebenden Bakterien und unterliegt deutlicher räumlich- zeitlichen Einflussfaktoren. Hier dominieren Phylotypen innerhalb der Bacteroidetes , gamma- und delta-Proteobakterien . • Saisonale Einflüsse auf die Bakteriengemeinschaften sind in den produktiven Frühjahrs- und Sommermonaten erkennbar. Insbesondere Aggregat-assoziierte Bakterien der Roseobacter -Gruppe und Bacteroidetes unterliegen biologischen Einflussfaktoren wie Phytoplanktonblüten. Tidale Einflüsse auf bakterielle Aktivität und Abundanz werden nur geringfügig und nicht systematisch durch Änderungen in der Zusammensetzung der Bakteriengemeinschaften reflektiert. • Actinobakterien stellen knapp 5% des Bakterioplanktons im Wattenmeer. Ihre Abundanz und Zusammensetzung im Süßwasserbereich unterscheidet sich von den marinen Standorten, wobei frei lebende und Aggregat-assoziierte Actinobakterien distinkte Gemeinschaften bilden. Aus dem Wattenmeer isolierte Stämme zeigen hohe Anpassungsfähigkeit anhand breiter Substrat- und Salinitätsspektren. • Insgesamt wird das organische Material im Wattenmeer von wenigen dominanten Bakterienarten umgesetzt, die ganzjährig auftreten und hoch angepasst sind. In produktiven Jahreszeiten treten darüber hinaus weitere, spezialisierte Bakterienarten auf, die in kurzen Zeitskalen von Änderungen der Zusammensetzung des organischen Materials, z. B. durch absterbendes Phytoplanton, profitieren.

Summary In the German Wadden Sea, organisms are influenced by highly dynamic processes. A shallow water column and tidal impact cause strong currents and high resuspension rates. The introduction of North Sea water masses also influences the Wadden Sea System. While sedimentological factors prevail in autumn and winter months, biological processes dominate in spring and summer. Within the research group “Biogeochemistry of tidal flats”, in which this thesis is included, tidal and seasonal variations of suspended matter appearance and bacterial activity and abundance were described. Hence, the focus of this thesis was to investigate the extend of tidal and seasonal impacts on the structure of resident bacterial communities in the water column. Furthermore, we determined if phylotypes detected in the German Wadden Sea are site-specific or detectable at other locations in the German Bight as well. Sampling was performed in the backbarrier tidal flat system of Spiekeroog in the German Wadden Sea. In the first part of this work, samples were taken weekly to investigate correlations of the bacterial communities and phytoplankton by group-specific DGGE (Denaturing gradient gel electrophoresis) and statistical methods. In the second part, in addition to seasonal also tidal processes were focussed. Sampling was performed in autumn, spring and summer hourly and in three hour intervals. The bacterial communities were investigated by group-specific DGGE (Denaturing gradient gel electrophoresis) for alpha- Proteobacteria , Bacteroidetes and the Roseobacter group. In addition, FISH (Fluorescense in situ hybridization) and the highly sensitive CARD-FISH (Catalyzed reporter deposition- FISH) were applied to determine abundances of individual bacterial groups in the water column of the German Wadden Sea. In a former study, remarkably high numbers of different gram-positive Bacteria were isolated which led to the assumption that this bacterial group exhibits an exceptional position in this habitat. To complete these data, CARD-FISH with Actinobacteria -specific probes was applied and a specific DGGE was established to determine abundances and phylogenetic variety of Actinobacteria in the Wadden Sea. Samples were taken at different sites in the backbarrier tidal flat system of Spiekeroog. In the last part of this work, different locations at the coastal line, near Helgoland and offshore were investigated in the German Bight in summer 2002 and 2003. The bacterial communities were analysed by specific DGGE for alpha-Proteobacteria and Bacteroidetes . To describe the ecology of the sampling sites hydrological, microbiological and particulate parameters were determined additionally.

The major findings of this thesis can be summarized as follows:

• The bacterial communities in the water column of the German Wadden Sea are mainly composed of alpha- and gamma-Proteobacteria and Bacteroidetes . In addition, beta- Proteobacteria are abundant on aggregates. Free-living and aggregate-associated bacteria form distinct communities in the German Wadden Sea and in the German Bight as well. The structure of free-living bacterial communities is mainly composed of few dominant phylotypes affiliated to the Roseobacter group. Their composition is stable on seasonal and spatial scales. The structure of aggregate-associated bacteria shows higher richness compared to free-living bacteria and is influenced by spatial- temporal impacts to a greater extend. Aggregate-associated bacteria are dominated by bacteria affiliated to the Bacteroidetes phylum, gamma- and delta-Proteobacteria . • Seasonal influences on the bacterial communities are detectable in the highly productive spring and summer months. Especially aggregate-associated Roseobacter and the Bacteroidetes follow biological impacts e. g. phytoplankton blooms. Tidal influences on bacterial activities and abundances are only marginally and not systematically reflected by changes of the bacterial community composition. • Actinobacteria represent about 5% of the Wadden Sea bacterioplankton. Their abundance and composition differs between the fresh water and marine sites, and free- living and aggregate-associated bacteria form distinct communities. Strains isolated from the Wadden Sea show high adaptation qualities on the basis of broad substrate and salinity ranges. • The organic matter in the Wadden Sea is mediated by few dominant bacterial species which are present throughout the year and are highly adapted. In productive seasons, specialised bacteria appear additionally which benefit from the changes of the organic matter composition, e. g. decaying phytoplankton, on small time-scales.

Inhaltsverzeichnis

Zusammenfassung Summary

I. Einleitung 1 I.1 Kleine Lebewesen, große Wirkung – Marine heterotrophe Bakterien im globalen Stoffkreislauf 2 I.2 Who´s who – Die Zusammensetzung der Bakteriengemeinschaften 5 I. 3 Geographie und Ökologie der Untersuchungsgebiete 8 I.3.1 Die Nordsee und die Deutsche Bucht 8 I.3.2 Das Wattenmeer 11 I.4 Zielsetzungen der Arbeit 15 I.5 Literatur 16

II. Effects of a phytoplankton bloom in a coastal ecosystem on the composition of bacterial communities 20 Abstract 22 Introduction 23 Materials and Methods 24 Results 28 Discussion 32 Literature cited 36

III. Tidal effects on coastal bacterioplankton 48 Abstract 51 Introduction 52 Materials and Methods 54 Results 56 Discussion 60 References 66

IV. Diversity and abundance of Gram-positive bacteria in a tidal flat ecosystem 81 Abstract 83 Introduction 84 Results 85 Discussion 88 Experimental procedures 93 References 98

V. High regional variability of bacterial communities in the German Bight, North Sea 114 Abstract 116 Introduction 117 Materials and Methods 118 Results 121 Discussion 124 References 130

VI. Schlussbetrachtung und Ausblick 144

Danksagung Kurzbiographie

Abkürzungsverzeichnis

CARD-FISH catalyzed reporter deposition-FISH Chl a Chlorophyll a

CO 2 Kohlendioxid DAPI 4´,6´-Diamidino-2-phenylindol hydrochlorid DGGE Denaturierende Gradienten Gelelektrophorese DNA desoxy ribonucleic acid DOC dissolved organic carbon DOM dissolved organic matter DSM, DSMZ Deutsche Sammlung von Mikroorganismen und Zellkulturen et al. et alii FISH fluorescence in situ hybridization FL free living HT high tide LT low tide ml Milliliter MT mean tide NCBI National Center for Biotechnology Information n. a. not available n. d. not determined PA particle attached PCR polymerase chain reaction PIC particulate inorganic carbon POC particulate organic carbon psu practical salinity unit rRNA ribosomal ribonucleic acid SPM suspended particulate matter

I. Allgemeine Einleitung

1 Kapitel I Einleitung

I.1 Kleine Lebewesen, große Wirkung – Marine heterotrophe Bakterien im globalen Stoffkreislauf

Bei der Betrachtung der Gesamtgröße der Weltmeere erscheint es zunächst kaum vorstellbar, dass mikroskopisch kleine Lebewesen den Großteil des Umsatzes organischen Materials im Wasser bewirken sollen. Berücksichtigt man allerdings, dass in einem tausendstel Liter bereits durchschnittlich 1-3 Mio. Bakterien vorhanden sind, ist offenbar, warum sich die Forschung seit mehr als zwei Jahrzehnten bemüht, diese höchst bemerkenswerten Lebewesen besser kennen zu lernen. Bakterieller Abbau und Remineralisierung wirken sich auf den Stoffkreislauf aller Elemente aus (Schlegel, 1992; Madigan et al., 2003), wobei der Kohlenstoffkreislauf große Bedeutung nicht zuletzt für klimatische Veränderungen besitzt. So hat die Aktivität mariner phototropher und heterotropher Bakterien sowohl durch die

Fixierung als auch durch den Ausstoß von CO 2 Einfluss auf das Weltklima (Smith und Hollibaugh, 1993; Wollast, 1993; Falkowski et al., 1998), so dass der marinen mikrobiellen Ökologie im Zuge der globalen Erwärmung immer größere Bedeutung beigemessen wird. Auch das genetische Potential der marinen Bakterien, das durch moderne Methoden zwar detektiert, aber bei weitem noch nicht entschlüsselt wurde, gibt der Wissenschaft Rätsel auf (Venter et al., 2004). Die zukunftsträchtige und viel versprechende Vision einer möglichen medizinischen oder biotechnologischen Nutzung mariner Mikroorganismen bietet daher ein weiteres großes Interessensgebiet der ökologischen Forschung.

Die Nahrungsquelle heterotropher Bakterien, organischer Kohlenstoff, liegt in der Wassersäule in gelöster (dissolved organic carbon, DOC) oder in partikulär gebundener Form (particulate organic carbon, POC) vor. DOC umfasst bis zu 95% des Gesamtkohlenstoffs der Weltmeere (Hedges, 1992) und wird nach Zusammensetzung und bakterieller Verfügbarkeit in eine labile und eine refraktäre Fraktion unterteilt (Søndergaard und Middelboe, 1995). Die Entstehung von gelöstem organischen Material (dissolved organic matter, DOM) ist durch einen komplexen Kreislauf gekennzeichnet, in dem im Wesentlichen Phytoplankton, Bakterien und Viren eine Rolle spielen (Abb. 1). Anhand der Darstellung wird deutlich, dass Bakterien DOM direkt aufnehmen und verwerten und somit die größte Bedeutung für die Umsetzung des DOC besitzen. Durch das daraus resultierende Wachstum und den Fraß durch Protozoen gelangt der Kohlenstoff dann indirekt in das Nahrungsnetz höherer Organismen. Die Entstehung des DOM ist abhängig von verschiedenen Faktoren, da die am Kreislauf beteiligten Organismen

2 Kapitel I Einleitung räumlich-zeitlichen Gegebenheiten unterliegen. Phytoplankton bildet bei ausreichender Nährstoffversorgung und günstigen Licht- und Temperaturverhältnissen Blüten aus, die in direkter Form zu der Absonderung („Leakage“) von DOM führen kann (Bjørnsen, 1988). Nach Absterben der Blüte werden durch Lysis der Zellen ebenfalls gelöste Stoffe freigesetzt. Darüber hinaus entsteht bei diesem Vorgang Detritus (partikuläre Zellreste), der durch Bakterien hydrolysiert und somit dem DOM Pool zugeführt wird.

Abb. 1 : Kreislauf gelösten organischen Kohlenstoffs in der Wassersäule (dissolved organic matter, DOM; modifiziert nach Riemann, 2001)

Neben dem DOC stellt in Aggregaten angereicherter POC eine weitere wichtige organische Kohlenstoffquelle dar (Alldredge, 1979) und bildet die Grundlage für komplexe Lebens- gemeinschaften, die sich im Weiteren aus Phytoplankton, Protozoen, Bakterien und Pilzen zusammensetzen (Alldredge and Silver, 1988). An Aggregate angeheftet, weisen sie, verglichen mit frei suspendierten Mikroorganismen, wesentlich höhere Zelldichten auf (Simon et al., 2002). Je nach Ursprung des partikulären Materials variieren der organische Anteil sowie dessen Zusammensetzung aus Kohlenhydraten und Proteinen, die den Bakterien als Substrat dienen (Azam und Cho, 1987; Smith et al., 1995, Azam und Cho, 1987; Biddanda und Benner, 1997). Auch hier ist das Phytoplankton, abhängig von seiner Artenzusammensetzung, hauptsächlicher Nährstofflieferant (Smith et al., 1995); weitere Bestandteile von Aggregaten sind hochrefraktäres oder auch anorganisches Material, z. B. resuspendiertes Sediment (Eisma, 1993). Die Freisetzung der Nährstoffe durch Bakterien erfolgt durch die Ausscheidung hydrolytischer Ektoenzyme, die partikulär gebundene

3 Kapitel I Einleitung

Makromoleküle in Oligo- und Monomere spalten (Madigan, 2003). Die hydrolysierten Stoffe werden teils von den Aggregat-assoziierten Bakterien selbst verwertet, teils diffundieren sie jedoch auch in das Umgebungswasser und stehen somit den frei suspendierten Bakterien und anderen planktischen Organismen zur Verfügung (Smith et al., 1992). Durch die Substrataufnahme wachsen die Bakterien und bilden somit Biomasse, die wiederum Zooplankton als Nahrungsquelle dient. Durch diesen Stoffkreislauf, der als microbial loop bezeichnet wird, werden Nährstoffe aus abgestorbenen Tier- und Pflanzenresten (Detritus) für höhere Trophiestufen wieder verfügbar (Azam et al., 1983).

In flachen Küstenzonen, Wattsystemen und Ästuaren unterscheidet sich die Situation im Vergleich zu den offenen Ozeanen durch ein sehr hohes Schwebstoffaufkommen. Die Schwebstoffe werden hier durch Flüsse oder von den angrenzenden Landgebieten eingetragen und sind zum Teil hohen Scherkräften ausgesetzt, die durch die flache Wassersäule und Tidenhub entstehen. Dadurch sind die Aggregate in den Küstenzonen wesentlich kleiner und häufiger (Lunau et al., 2006) und besitzen, verglichen mit Schwebstoffen in küstenfernen Gebieten, einen geringeren organischen Anteil (Postma, 1981; Lunau et al., 2006). Durch ständige Turbulenz werden die Aggregate fortwährend resuspendiert und somit in Schwebe gehalten, was ebenfalls Auswirkungen auf die angehefteten Bakterien hat. So weisen Aggregat-assoziierte Bakterien in schwebstoffreichen Gewässern wesentlich höhere Enzymaktivitäten und Biomasse auf, und können sogar bis zu 95% der Gesamtaktivität der suspendierten Bakterien ausmachen (Crump et al., 1998; Crump & Baross, 2000). Auch die Bakteriengemeinschaften können in Küstenzonen anders zusammengesetzt sein als in zulaufenden Flüssen oder im offenen Meer. So bilden sich entweder Mischformen von Süßwasser- und marinen Bakteriengemeinschaften (Rappé et al., 2000), oder auch distinkte Bakteriengemeinschaften von Süßwasser, Brackwasser und marinem Milieu aus (Selje et al., 2003, Crump et al. 1999).

Diese Zusammenhänge verdeutlichen, in welchem Umfang Bakterien das gesamte Nahrungsnetz beeinflussen und dass freilebende und Aggregat-assoziierte Bakterien vollkommen unterschiedliche Lebensbedingungen vorfinden. Daher ist eine differenzierte Untersuchung beider Lebensgemeinschaften essentiell, um die ökologischen Zusammenhänge in der Wassersäule verstehen zu können.

4 Kapitel I Einleitung

I.2 Who´s who – Die Zusammensetzung der Bakteriengemeinschaften

Da Bakterien unter dem Mikroskop und in der Kultivierung nur sehr wenige Unterschiede anhand von Zellmorphologie und Wachstum aufweisen, wurde die Artenvielfalt von natürlichen Bakteriengemeinschaften lange Zeit unterschätzt. Darüber hinaus bot die Kultivierung nur bedingt Einblick in das Vorkommen und die Häufigkeit von Bakterienarten, da die Bedingungen, die Bakterien im Labor vorfinden, nicht den natürlichen Gegebenheiten entsprachen. Einige Bakterienstämme oder auch phylogenetische Gruppen konnten leicht unter künstlichen Bedingungen angereichert werden und wurden somit auch häufiger in verschiedenen Habitaten nachgewiesen, während sich andere Bakterien nur unter bestimmten Voraussetzungen kultivieren ließen oder bis heute unkultiviert bleiben. Daher ergaben sich große Unterschiede zwischen mikroskopisch und durch Kultivierungsansätze ermittelte Zellzahlen („great plate count anomaly“, Staley & Konopka, 1985). So brachte die Einführung molekularbiologischer Methoden, die auf dem Vergleich des Erbguts anhand der ribosomalen RNA beruhten, neue Einblicke in die mikrobielle Ökologie und die phylogenetischen Zusammenhänge (Woese et al., 1987). Bis heute stellt die hochkonservierte 16S rRNA bzw. der 16S rRNA Genabschnitt eine wesentliche Grundlage für die Untersuchung von Bakteriengemeinschaften dar. Die Vervielfältigung und Sequenzierung von Genen ermöglichte es, Bakteriengenome und Phylogenie unabhängig von Kultivierungserfolgen zu erforschen (Saiki et al, 1988; Sanger et al., 1977). Gängige Methoden zur Detektion sind z.B. die Denaturierende Gradienten Gelelektrophorese (DGGE; Muyzer et al., 1993), Restriktionsfragment Längen-Polymorphismus (RFLP; Marsh, 1999), oder die rDNA Intergenic Spacer Analysis (RISA). Die Quantifizierung von Bakteriengruppen oder auch –arten kann durch Fluoreszenz in situ Hybridisierung (FISH; Giovannoni et al, 1988; Amann et al., 1990) bzw. Catalyzed Reporter Deposition-FISH (CARD-FISH; Pernthaler et al., 2002) sowie mittels Realtime PCR (Heid et al., 1996) erfolgen. Heute werden Kultivierungsansätze und kultivierungsunabhängige Methoden sowie Aktivitätsmessungen kombiniert, um möglichst viele Informationen über die Mikrobiologie eines Habitats zu gewinnen.

Durch den Einsatz dieser Methoden konnte die Struktur der am Stoffumsatz beteiligten Bakterien, die vorher als „Black Box“ betrachtet wurden, weiter aufgeklärt werden (Giovannoni & Rappé 2000). So stellte sich heraus, dass insbesondere die gram-negativen Proteobakterien sowie Bacteroidetes bedeutende Gruppen innerhalb des marinen

5 Kapitel I Einleitung heterotrophen Bakterioplanktons bilden. Darüber hinaus wurden u.a. methylotrophe Bakterien, Planctomycetales und die gram-positiven Actinobakterien in marinen Habitaten nachgewiesen. Innerhalb der Proteobakterien wurden die gamma-Proteobakterien lange Zeit als die dominanteste Gruppe des marinen Bakterioplanktons angenommen, da sich Vertreter dieser Gruppe leicht unter Laborbedingungen isolieren ließen. Über kultivierungsunabhängige Methoden fand man jedoch heraus, dass die meisten weltweit nachgewiesenen Phylotypen distinkte Cluster bildeten, die wiederum keine Isolate beinhalteten (Giovannoni und Rappé, 2000). Mittlerweile konnten teilweise auch für diese Cluster mit gezielten Anreicherungsversuchen einzelne Isolate gewonnen werden (Cho und Giovannoni, 2004), so dass die Erforschung der ökologischen Funktion dieser Organismen weiter voranschreiten kann. Physiologisch betrachtet sind gamma-Proteobakterien fakultativ anaerobe und chemoheterotrophe Organismen, die häufig Oberflächen-assoziiert vorkommen und somit im Sediment sowie auf Aggregaten eine zentrale Rolle einnehmen. Die alpha-Proteobakterien sind ebenfalls weltweit verbreitet und zumeist durch die Subgruppen Sphingomonas und Roseobacter vertreten. Weitere große Bedeutung besitzen die hoch spezialisierten Cluster SAR 11 (Rappé et al. 2002) und SAR 116 innerhalb der alpha- Proteobakterien . Die chemoorganotrophen Roseobacter wurden bisher ausschließlich im marinen Milieu nachgewiesen und stellen dort habitatabhängig bis zu 50% der gesamten alpha-Proteobakterien . Einige Vertreter gehören zu den aeroben anoxygenen phototrophen Bakterien und sind somit auch in der Lage, Photosynthese zu betreiben. Aktuell werden große Forschungsprojekte zur Genomentschlüsselung dieser Organismen 1 durchgeführt, die das Potential dieser hoch interessanten und vielfältigen Gruppe weiter aufklären sollen. Die aeroben oder fakultativ anaeroben, chemoorganotrophen Bacteroidetes bilden die dritte große Gruppe innerhalb des marinen Bakterioplankton. Sie sind hoch divers und leben in der Wassersäule sowohl frei suspendiert als auch Aggregat-assoziiert. Ihre Stoffwechselphysiologie ist äußerst vielfältig, doch es hat sich gezeigt, dass besonders schwer abbaubare, hochmolekulare Substanzen bevorzugt von Bacteroidetes abgebaut werden können, z. B. Chitin oder Cellulose (Cottrell und Kirchman, 2000). Darüber hinaus sind sie häufig beweglich und können auf Oberflächen gleiten, so dass durch diese Eigenschaften

1 Auch andere Bakteriengruppen, die in marinen Habitaten von Bedeutung sind, werden derzeit durch große Genomprojekte erforscht (z. B. Moran et al., 2004). Die Untersuchung des genetischen Potentials von Organismen führt neben der Entschlüsselung bisher unbekannter Gene auch zur Entdeckung neuer Stoffwechselwege oder biotechnologisch nutzbarer Substanzen (Fusetani, 2000). Man kann daher annehmen, dass die Ozeane ein riesiges Potential bisher unentdeckter Ressourcen bietet, deren Erforschung im Zuge interdisziplinärer Projekte immer mehr in den Vordergrund tritt.

6 Kapitel I Einleitung angenommen wurde, dass sie besonders auf Aggregaten eine große Bedeutung für Stoffumsatzprozesse einnehmen.

Diese Zusammenhänge zeigen, dass das Verständnis über die Vorgänge der Remineralisierung durch die Strukturaufklärung der beteiligten Bakteriengemeinschaften wesentlich verbessert werden konnte. Darüber hinaus können lokale Gegebenheiten einzelner Habitate jedoch übergeordnet Einfluss auf die Zusammensetzung der Bakteriengemeinschaften und somit auch auf die Effektivität des Stoffumsatzes nehmen.

7 Kapitel I Einleitung

I. 3 Geographie und Ökologie der Untersuchungsgebiete

I.3.1 Die Nordsee und die Deutsche Bucht

Die Nordsee liegt auf dem europäischen Kontinentalschelf und wird begrenzt von den Britischen Inseln und dem Europäischen Kontinent (Abb. 2). Sie ist mit einer durchschnittlichen Tiefe von 93 m ein flaches Schelfmeer und durch verschiedene angrenzende Land- und Wasserregionen beeinflusst. Salines Atlantikwasser dringt im Norden zwischen der schottischen und der norwegischen Küste sowie südlich durch den Ärmelkanal in die Nordsee. Der größte Eintrag von Süßwasser erfolgt über den Skagerrak aus der Ostsee und durch verschiedene große Flüsse, die in die Nordsee münden (z. B. von deutscher Seite die Flüsse Rhein, Ems, Weser, Elbe und Eider). Der mittlere Salzgehalt ist demnach mit durchschnittlich 15 – 25 Promille an den Küstengebieten geringer als in der offenen Nordsee, wo durchschnittlich 32 – 35 Promille vorherrschen (Alongi, 1997). Aufgrund der Amphidromie in der südlichen Nordsee (Defant, 1923) fließt das Wasser in der deutschen Bucht entgegen dem Uhrzeigersinn, wodurch auch der Transport von partikulären und gelösten Stoffen sowie von planktischen Organismen beeinflusst wird.

Abb. 2 : Geographische Lage der Nordsee und der Deutschen Bucht sowie des Nationalparks Deutsches Wattenmeer von der ostfriesischen zur nordfriesischen Küste (modifiziert nach http://www.bsh.de)

Die deutsche Bucht, der südliche Teil der Nordsee, reicht von Jütland in Dänemark über die Friesischen Inseln (Nord-, Ost-, und Westfriesische Inseln) bis zur niederländischen Grenze im Westen. Im Nordwesten wird sie begrenzt von der Doggerbank, einer flachen Sandbank

8 Kapitel I Einleitung innerhalb der Nordsee, die durch große Fischvorkommen insbesondere für die Fischerei eine wesentliche Rolle spielt. Das Küstengebiet der Deutschen Bucht bildet das Deutsche Wattenmeer, eine einzigartige Flachwasserzone, die sich hinter den Friesischen Inseln erstreckt (vgl. I.3.2).

Grundlage des Nahrungsnetzes in der Nordsee bilden einzellige Algen, das Phytoplankton, welches im Jahresverlauf mehrere Blüten ausbildet (Alongi, 1997). Im Frühjahr (März bis April) führen steigende Temperatur- und Lichteinstrahlung sowie hohe Nährstoffkonzentrationen dazu, dass sich eine Blüte aus Kieselalgen (Diatomeen, Bacillariophyceae) ausbildet (Drebes, 1974). Durch Nährstofflimitierung und sukzessiven Fraßdruck durch Zooplankton endet die Blüte zumeist im Juni, bis im Spätsommer eine zweite, meist weniger intensive Phytoplanktonblüte entsteht (Alongi, 1997). Während der Blüte scheiden die Diatomeen gelösten organischen Kohlenstoff aus, der von Bakterien genutzt werden kann (Ittekott et al. 1981). Auch nach dem Zusammenbruch einer Phytoplanktonblüte profitieren heterotrophe Bakterien vom nährstoffhaltigen Lysat. Dies kann sich in vermehrter bakterieller Aktivität, Abundanz und Veränderungen in der Artenzusammensetzung ausdrücken (Reinthaler et al., 2005; Smith et al, 1995, Riemann et al., 2000, Fandino et al., 2001).

Seit Mitte der 1950er Jahre wurde in der Nordsee ein stetiger Anstieg von Nährstoffen bedingt durch anthropogene Einflüsse gemessen, der dazu geführt hat, dass die Nordsee stark eutrophiert ist (Alongi, 1997). Dies führte zu Verschiebungen sowohl in der Artenzusammensetzung als auch in der Biomasse des Phytoplanktons und resultierte in der Einschränkung der gesamten Artenvielfalt in der Nordsee. Da die Nährstoffe zumeist über die Zuflüsse in die Nordsee eingetragen werden, sind erhöhte Konzentrationen von Phosphat, Nitrat oder auch von Schwermetallen als Gradient von der Küste in die offenen Gewässer zu beobachten. Darüber hinaus ist durch die globale Erwärmung auch eine Erwärmung der Wassersäule in der Deutschen Bucht um durchschnittlich 1,1°C beobachtet worden (Wiltshire and Manly, 2004). Die damit verbundene Verschiebung der Algenblüten könnte durch eine Veränderung der temperaturabhängigen Rahmenbedingungen entstanden sein. Der Zustand der Nordsee wird daher seit Jahrzehnten durch verschiedene Institutionen in Monitoringserien untersucht, um die Nutzung und Belastung der Gewässer zu überwachen und den Lebensraum zu schützen (Bundesamt für Seeschiffahrt und Hydrographie, BSH; http://www.bsh.de ).

9 Kapitel I Einleitung

Untersuchungen zu den Bakteriengemeinschaften in der Nordsee liegen von verschiedenen Autoren vor. So untersuchten Eilers et al. (2001) die Kultivierbarkeit von Nordseebakterien bei Helgoland. Es konnten hauptsächlich alpha- und gamma-Proteobakterien sowie Bacteroidetes isoliert werden. FISH-Zählungen mit spezifischen Sonden ergaben, dass diese Gruppen ebenfalls einen großen Teil der Bakteriengemeinschaften in der Nordsee darstellen. Große saisonale Unterschiede in der Hybridisierbarkeit waren erkennbar, die zeigten, dass in den biologisch hoch produktiven Sommermonaten wesentlich höhere Effizienz erreicht wurde als in den Wintermonaten. Gerdts et al. (2004) gaben eine Übersicht der Aktivität, Abundanz und saisonale Veränderungen bakterieller Gemeinschaften, die mit verschiedenen Methoden als Langzeitmonitoring bei Helgoland durchgeführt wurden. Besonders in den produktiven Sommermonaten ergaben sich deutliche Änderungen in der Aktivität und Zusammensetzung. In der südlichen Nordsee wurden saisonal bakterielle Respiration, Artenreichtum („Richness“) und Biomasseproduktion entlang von Transekten (Reinthaler et al., 2005) sowie in Abhängigkeit von Phytoplanktonblüten untersucht (Reinthaler & Herndl., 2005). Es zeigte sich, dass die bakterielle Biomasseproduktion saisonal stark variiert und korreliert ist mit der Primärproduktion.

Diese Zusammenhänge verdeutlichen die Notwendigkeit, durch weitere Erforschung der relevanten Bakterienarten in der Nordsee Schlüsselorganismen zu erkennen und zu beschreiben. Anhand solcher Indikatororganismen könnten sowohl Änderungen der ökologischen Gegebenheiten sowie detaillierte Aussagen über Stoffumsatz und äußere Einflüsse möglich sein. Die Erforschung der Wechselwirkungen zwischen physiko- chemischen und biologischen Kräften stellt eine essentielle Brücke dar zum Verständnis der Umwelt und der Bedeutung für das gesamte Ökosystem.

10 Kapitel I Einleitung

I.3.2 Das Wattenmeer

Die südöstliche Nordseeküste besteht aus einem besonderen ökologischen Lebensraum, dem europäischen Wattenmeer (Abb. 2). Es bildet mit einer Gesamtfläche von ca. 7500 m 2 und einer Gesamtlänge von 500 km die größte zusammenhängende Wattfläche der Welt und reicht von Den Helder (Niederlande) bis Esbjerg (Dänemark). Es ist gekennzeichnet durch eine hohe Artenvielfalt und wurde daher 1985 zum Nationalpark erklärt. Der deutsche Teil des Wattenmeeres wird unterteilt in das Niedersächsische-, das Hamburgische- und das Schleswig-Holsteinische Wattenmeer. Die Friesischen Inseln sind dem Wattenmeer in Richtung Nordsee vorgelagert und bilden so eine natürliche Begrenzung.

Als Lebensraum ist das Wattenmeer stark durch die Gezeiten geprägt, wodurch große Teile des Watts in regelmäßigem Abstand trocken fallen und daher extreme Lebensbedingungen bieten. Nach Lozan et al. (1994) kann das Watt in vier Ablagerungsbereiche unterteilt werden:

a) Das Sublitoral: Ständig von Salzwasser bedeckte Flächen, z. B. Seegat, Wattrinnen b) Das Eulitoral: Bereiche, die bei Hochwasser überflutet sind und bei Niedrigwasser trockenfallen, z. B. Wattflächen zwischen Inseln und Festland c) Das Supralitoral: Nur bei hochauflaufender Flut von Salzwasser bedeckt, z. B. Salzmarschen der Inseln und des Festlands d) Die Dünen: Keine Überspülung mit Salzwasser

Die im Watt lebenden Organismen müssen daher eine hohe Anpassungsfähigkeit besitzen, da durch die zeitweise Exponierung der Wattfläche und eine insgesamt flache Wassersäule starke Temperaturschwankungen entstehen. Auch Salinitätsschwankungen sind sehr ausgeprägt, insbesondere bei starken Regenfällen und an Flußmündungen. Der Austausch der Wasserkörper zwischen der Nordsee und dem Wattenmeer geschieht über die Seegatten, Durchlässe zwischen den Inseln, in denen bei jeder Ebbe und Flut sehr hohe Strömungsgeschwindigkeiten von bis zu 2 m s -1 erreicht werden.

Durch die Eutrophierung der Nordsee insbesondere an den Küsten (vgl. Abschnitt I.3.1) sind seit den 70er Jahren diverse Projekte zur Beobachtung der Stoffflüsse im Wattenmeer durchgeführt worden (Baretta and Ruardij, 1988; Cadée, 1984; de Wilde und Beukema, 1984). Um die komplexen Zusammenhänge zwischen der Hydrographie, der Biologie und

11 Kapitel I Einleitung anthropogenen Einflüssen im Wattenmeer zu studieren, wurden darüber hinaus umfassende interdisziplinäre Forschungsprojekte ins Leben gerufen. Diese sind zusammenfassend für das nordfriesische Wattenmeer von Gätje und Reise (1998) sowie für das Spiekerooger Rückseitenwatt von Dittmann (1999) veröffentlicht worden und bilden die Grundlage für das Verständnis der Wattenmeerökologie. In beiden Werken wurden Entstehung, Geologie, Nährstoffkonzentrationen und Stoffflüsse sowie Flora und Fauna untersucht. Eine wesentliche Rolle nahmen Phyto- und Zooplankton ein; das Bakterioplankton wurde zwar als relevant erachtet, erschien jedoch als „Black Box“, da nur die Abundanz des gesamten Bakterioplanktons gemessen wurde. Die Artenzusammensetzung wurde in diesen Arbeiten nicht berücksichtigt.

Im Rahmen der DFG-geförderten interdisziplinären Forschergruppe „BioGeoChemie des Watts“, die 2001 ins Leben gerufen wurde, sind weitere Erkenntnisse über die Vorgänge im Wattenmeer gewonnen worden. Hierzu wurde ein Messpfahl im Spiekerooger Rückseitenwatt (Standort Otzumer Balje; http://www.icbm.de/watt) errichtet, über den ein fortwährendes Monitoring der physiko-chemischen Parameter durchgeführt wird. Zusätzlich wurden konzertierte Meßkampagnen sowie regelmäßige Beprobungen von Tagesgängen durchgeführt. Durch diese umfangreichen Datensätze manifestierte sich die Bedeutung biologischer Vorgänge auch im sediment-dominierten Wattenmeer. Besonders zu Niedrigwasser während des Tages waren biologische Einflüsse erkennbar, sowie saisonal zu produktiven Jahreszeiten, in denen Phytoplanktonblüten auftraten (Grossart et al., 2004; Lunau et al., 2006). So wurden im Gegensatz zu den Herbst- und Wintermonaten beispielsweise im Mai und Juni erhöhte bakterielle Biomasseproduktion und Abundanz gemessen. Partikelabundanz und –größe verhielt sich gegenläufig, indem in den Wintermonaten höhere Abundanzen kleinerer Partikel detektiert wurden, im Frühjahr und Sommer jedoch größere Aggregate in kleinerer Anzahl. Es ist daher anzunehmen, dass der Einfluß von Phyto- und Bakterioplankton auf Aggregation und Disaggregation durch Ausscheidung klebriger Substanzen (TEP, transparente Exopolymere; EPS, Exopolysaccharide; Passow, 2002; Bhaskar et al., 2005) auch in Wattsystemen von elementarer Bedeutung ist. Auf tidaler Ebene konnten regelmäßig wiederkehrende Signaturen des suspendierten partikulären Materials (SPM) nachgewiesen werden, die mit weiteren partikulären Parametern wie partikulärem organischem Kohlenstoff (POC) und Chlorophyll a korrelierten. Die Abundanz partikel-assoziierter Bakterien verhielt sich in der Gesamtprobe weitestgehend

12 Kapitel I Einleitung konstant, wie auch schon in anderen Arbeiten gezeigt (Stevens et al., 2005). Die Auftrennung in eine absinkende und eine frei schwebende Fraktion zeigte jedoch deutliche Unterschiede in der Besiedelung (Lunau et al., 2004). Diese Hinweise deuten darauf hin, dass sich die distinkten Bakteriengruppen unterschiedlich verhalten und Einfluss nehmen.

Erste Untersuchungen zur Zusammensetzung der bakteriellen Gemeinschaft im Watt wurden 1998 durch Llobet-Brossa und Kollegen im Wattenmeersediment durchgeführt. Mittels Fluoreszenz In Situ Hybridisierung (FISH) wurden sulfatreduzierende Bakterien sowie Bacteroidetes im Jadebusen nachgewiesen (Llobet-Brossa et al. 1998, 2002). Im Rahmen der Forschergruppe „BioGeoChemie des Watts“ wurden umfangreiche Untersuchungen auch in tieferen Sedimentschichten an mehreren Probenahmeorten im Spiekerooger Rückseitenwatt durchgeführt (Mußmann et al., 2005; Köpke et al., 2005; Willms et al., 2006). Kultivierungsansätze ergaben eine hohe Artenvielfalt, und die Isolate konnten den phylogenetischen Gruppen der Proteobakterien , Bacteroidetes , Fusobakterien , Actinobakterien und Firmicutes zugeordnet werden. Die molekularbiologischen Untersuchungen ergaben ein ähnliches Spektrum innerhalb der phylogenetischen Gruppen und die zusätzliche Detektion methanogener Archaeen. Beide Untersuchungen ergaben Zusammenhänge sowohl zu Aktivitätsmessungen als auch zu sedimentologischen Parametern und geben somit deutliche Hinweise auf eine ökologische Bedeutung der nachgewiesenen Stämme und Phylotypen.

Auch in der Wassersäule des Spiekerooger Rückseitenwatts wurden umfassende Untersuchungen über Bakteriengemeinschaften von Stevens et al. (2005a, b) durchgeführt. In den Jahren 1999 bis 2000 wurden monatlich Proben genommen, in denen mittels DGGE nachgewiesen werden konnte, dass hier distinkte Bakteriengruppen existieren: Auf dem Wattsediment, auf Schwebstoffen, und frei lebend in der Wassersäule. In jedem dieser Kompartimente waren Phylotypen nachgewiesen worden, die sich ausschließlich in diesem Lebensraum befanden, sowie Schnittmengen zwischen den einzelnen Gruppen. Vor allem auf Schwebstoffen bildeten die nachgewiesenen Phylotypen eine Mixtur aus Sediment- und freilebenden Bakterien. Im Allgemeinen wurden Phylotypen verschiedener Proteobakterien (alpha-, beta-, gamma- und delta-Proteobakterien ), Bacteroidetes und Gram-positive Bakterien gefunden. Saisonale Veränderungen wurden vorwiegend in den Sommermonaten während oder nach Phytoplanktonblüten beobachtet. Ein parallel durchgeführter umfassender

13 Kapitel I Einleitung

Kultivierungsansatz ergab nur wenige Übereinstimmungen zu den molekularbiologischen Ergebnissen. Die monatliche Probenahme in der Wassersäule war somit geeignet, einen ersten Einblick in den mikrobiologischen Lebensraum Wattenmeer zu geben, gab jedoch keinen Aufschluss über die Zeitskala, in der sich Veränderungen der bakteriellen Zusammensetzung ereigneten. Ebenfalls waren bestimmte Bakteriengruppen unterrepräsentiert, von denen sich in Kultivierungsansätzen (z. B. Gram-positive Bakterien) sowie anhand von FISH-Zellzahlen (Bacteroidetes ) gezeigt hat, dass sie einen großen Anteil der Bakteriengemeinschaft im Wattenmeer bilden.

Die weitere Aufklärung der Zusammensetzung der Bakteriengemeinschaften auf Ebene einzelner, als häufig erkannter phylogenetischer Gruppen ist daher dringend erforderlich. Ebenso stellt sich die Frage, wie die Bakteriengemeinschaften in kleineren Zeitskalen beeinflusst werden und inwiefern sich die oszillierenden SPM-Signaturen im Tidenzyklus auf einzelne Bakteriengruppen auswirken. Es ergaben sich daher folgende Zielsetzungen für die vorliegende Arbeit:

14 Kapitel I Einleitung

I.4 Zielsetzungen der Arbeit

Ziel der Arbeit war die detaillierte Analyse der frei lebenden und aggregat-assoziierten Bakteriengemeinschaften im Wattenmeer und der Deutschen Bucht mittels DGGE und FISH anhand von 16S rRNA und 16S rRNA Genabschnitten. Um auch unterrepräsentierte Bakteriengruppen erfassen zu können, wurden in allen Arbeiten nicht nur Bacteria-, sondern auch gruppenspezifische Oligonukleotide (PCR-Primer) verwendet.

Ein Teilaspekt dieser Zielsetzung war die Untersuchung der Zusammenhänge zwischen Phytoplanktonblüten und Veränderungen in den Bakteriengemeinschaften beider Kompartimente im ostfriesischen Wattenmeer mit einem engen Probenahmeraster ( Kapitel II ). Einen weiteren Aspekt stellte die Untersuchung tidaler Einflüsse auf die Bakteriengemeinschaften in Abhängigkeit von saisonalen Aspekten dar ( Kapitel III ). Aufgrund der extremen Verhältnisse, die im Wattenmeer herrschen (vgl. Abschnitt I.3.2), war die Anwendung hoch sensitiver Methoden erforderlich, die zunächst etabliert und getestet werden mussten. So wurde die DGGE nicht nur DNA, sondern auch RNA basiert durchgeführt; neben der FISH wurde für einen Vergleich ebenfalls die CARD-FISH Methode angewendet. Anhaltspunkte über außergewöhnlich hohe Abundanzen von Gram-positiven Bakterien im Wattenmeer (Stevens, 2004) erforderten molekularbiologische Untersuchungen, um die Relevanz dieser Aussage zu untermauern ( Kapitel IV ). Mit Hilfe einer spezifischen DGGE sowie CARD-FISH Untersuchungen konnten zusätzliche Hinweise über Zusammensetzung und Abundanz von Gram-positiven Bakterien an verschiedenen Standorten im Wattenmeer nachgewiesen werden. Eine weitere Zielsetzung stellte die Untersuchung von Bakteriengemeinschaften im Watten- meer und der Deutschen Bucht dar, um den Austausch von Wassermassen und die räumliche Verteilung der dominierenden Bakteriengruppen darzustellen ( Kapitel V ). Hierzu wurden zwei Messkampagnen im Sommer 2002 und 2003 mit umfassenden Probenahmen durchgeführt, um lokale Gegebenheiten und ihren Einfluss auf die Bakteriengemeinschaften zu erfassen.

15 Kapitel I Einleitung

I.5 Literatur

Alldredge, A. L. 1979. The chemical composition of macroscopic aggregates in two neritic seas. Limnol Oceanogr 24:855-866 Alldredge, A. L., and M. W. Silver. 1988. Characteristics, dynamics, and significance of marine snow. Prog Oceanogr 20:41-82 Amann, R. I., L. Krumholz, and D. A. Stahl. 1990. Fluorescent-oligonucleotide probing of whole cells for determinative, phylogenetic, and environmental studies in microbiology. J Bacteriol 172: 762-770 Azam, F., T. Fenchel, J. G. Field, J. S. Gray, L. A. Meyer-Reil, and T. F. Thingstad. 1983. The ecological role of water-column microbes in the sea. Mar Ecol Prog Ser 10:257- 263 Azam, F., and B. C. Cho. 1987. Bacterial utilization of organic matter in the sea. In Ecology of Microbial Communities, Cambridge University Press, 261-281 Baretta, J., and P. Ruardij. 1988. Tial flat estuaries. Berlin: Springer-Verlag. Bhaskar, P., H. P. Grossart, N. Bhosle, and M. Simon. 2005. Production of macroaggregates from dissolved exopolymeric substances (EPS) of bacterial and diatom origin. FEMS Microbiol Ecol 53: 255-264 Biddanda, B. A., and R. Benner. 1997. Carbon, nitrogen and carbohydrate fluxes during the production of particulate and dissolved organic matter by marine phytoplankton. Limnol Oceanogr 42:506-518 Bjørnsen, P. K. 1988. Phytoplankton exudation of organic matter: Why do healthy cells do it? Limnol Oceanogr 33:151-154 Cadée, G. C. 1984. Has input of organic matter into the western part of the Dutch Wadden Sea increased during the last decades? Neth Inst Sea Res Publ Ser 10:71-82 Cho, J. C., and S. J. Giovannoni. 2004. Cultivation and growth characteristics of a diverse group of oligotrophic marine Gammaproteobacteria . Appl Environ Microbiol 70: 432- 440 Cottrell, M., and D. L. Kirchman. 2000. Natural assemblages of marine Proteobacteria and members of Cytophaga-Flavobacter cluster consuming low- and high-molecular- weight dissolved organic matter. Appl Environ Microbiol 66:1692-1697 Crump, B. C., J. A. Baross, and C. A. Simenstad. 1998. Dominance of particle-attached bacteria in the Columbia River estuary, USA. Aquat Microb Ecol 14:7-18 Crump, B. C., E. V. Armbrust, and J. A. Baross (1999) Phylogenetic analysis of particle- attached and free-living bacterial communities in the Columbia River, its estuary and the adjacent coastal ocean. Appl Environ Microbiol 65:3192-3204 Crump, B. C., and J. A. Baross. 2000. Characterization of the bacterially-active particle fraction in the Columbia River estuary. Mar. Ecol. Prog. Ser. 206: 13-22 De Wilde, P. A. W. J., and J. J. Beukema. 1984. The role of zoobenthos in the consumption of organic matter in the Dutch Wadden Sea. Neth Inst Sea Res Publ Ser 10:145-158 Dittmann, S. 1999. The Wadden Sea ecosystem: stability, properties and mechanisms. New York: Springer-Verlag.

16 Kapitel I Einleitung

Eilers, H., J. Pernthaler, J. Peplies, F. O. Glöckner, G. Gerdts, and R. Amann. 2001. Isolation of novel pelagic bacteria from the German Bight and their seasonal contributions to surface picoplankton. Appl Environ Microbiol 67:5134-5142 Eisma, D. 1993. Suspended matter in the aquatic environment. Sprimger, Heidelberg Falkowski, P. G., R. T. Barber, and V. V. Smetacek. 1998. Biogeochemical controls and feedbacks on ocean primary production. Science 281: 200-207 Fusetani, N. 2000. Drugs from the Sea. Basel: Karger-Verlag, S. 1ff Gätje, C., and K. Reise. 1998. Ökosystem Wattenmeer. Austausch-, Transport- und Stoffumwandlungsprozesse. New York: Springer-Verlag. Gerdts, G., A. Wichels, H. Döpke, K. W. Klinge, W. Gunkel, und C. Schütt. 2004. 40-year long-term study of microbial parameters near Helgoland (German Bight, North Sea): historical view and future perspectives. Helgol Mar Res 58: 230-242 Giovannoni, S. J., E. F. DeLong, G. J. Olsen, and N. R. Pace. 1988. Phylogenetic group- specific oligodeoxynucleotide probes for identification of single microbial cells. J Bacteriol 170: 720-726 Giovannoni, S. J., and M. Rappé. 2000. Evolution, diversity, and molecular ecology of marine prokaryotes. In: D Kirchman (ed) Microbial Ecology of the Oceans, Wiley-Liss, Inc, 47-84 Hedges, J. I. 1992. Global biogeochemical cycles: progress and problems. Mar Chem 39:67- 93 Heid C. A., J. Stevens, K. J. Livak, and P. M. Williams. 1996. Real-time quantitative PCR. Genome Res 6:986-994. Ittekott, V., U. Brockmann, W. Michaelis, und E. T. Degenes. 1981. Dissolved free and combined carbohydrates during a phytoplankton bloom in the Northern North Sea. Mar Ecol Prog Ser 4: 299-305 Köpke, B., R. Willms, B. Engelen, H. Cypionka, and H. Sass. 2005. Microbial diversity in coastal subsurface sediments: A cultivation approach using various electron acceptors and substrate gradients. Appl Environ Microbiol 71: 7819-7830 Llobet-Brossa, E., R. Rossello-Mora, and R. Amann. 1998. Microbial community composition of Wadden Sea sediments as revealed by fluorescence in situ hybridization. Appl Environ Microbiol 64:2691-2696 Llobet-Brossa, E., R. Rabus, M. E. Böttcher, M. Könneke, N. Finke, A. Schramm, R. L. Meyer, S. Grotzschel, R. Rossello-Mora, and R. Amann. 2002. Community structure and activity of sulfate-reducing bacteria in an intertidal surface sediment: a multi- method approach. Aquat Microb Ecol 29:211-226 Lozan, J. L., E. Rachor, K. Reise, H. v. Westernhagen, und W. Lenz. 1994. Warnsignale aus dem Wattenmeer. Blackwell Wissenschafts-Verlag, Berlin. Lunau, M., A. Sommer, A. Lemke, H. P. Grossart, und M. Simon. 2004. A new sampling device for microaggregates in turbid aquatic systems. Limnol Oceanogr: Methods 2:387-397 Lunau, M., A. Lemke, O. Dellwig, und M. Simon. 2006. Physical and biogeochemical controls of microaggregate dynamics in a tidally affected coastal ecosystem. Limnol Oceanogr 51: 847-859 Marsh, T. L. (1999) Terminal restriction length polymorphism (T-RFLP): an emerging

17 Kapitel I Einleitung

method for characterization diversity among homologous populations of amplification products. Curr. Opin. Microbiol. 2: 323-327. Moran, M. A., und Kollegen. 2004. Genome sequence of Silicibacter pomeroyi reveals adaptations to the marine environment. Nature 432: 910-913 Mussmann, M., K. Ishii, R. Rabus, und R. Amann. 2005. Diversity and vertical distribution of cultured and uncultured Deltaproteobacteria in an intertidal mud flat of the Wadden Sea. Environ Microbiol 7: 405-418 Muyzer, G., E. C. de Waal, and A. G. Uitterlinden. 1993. Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl Environ Microbiol 59:695-700 Passow, U. 2002. Production of transparent exopolymer particles (TEP) by phyto- and bacterioplankton. Mar Ecol Prog Ser 238: 1-12 Pernthaler, A., J. Pernthaler, and R. Amann. 2002. Fluorescence in situ hybridization and catalyzed reporter deposition for the identification of marine bacteria. Appl. Environ. Microbiol. 68: 3094-3101 Postma, H. 1981. Exchange of materials between the North Sea and the Wadden Sea. Mar Geol 40: 199-213 Rappé, M. S., K. Vergin, und S. J. Giovannoni. 2000. Phylogenetic comparisons of a coastal bacterioplankton community with its counterparts in open ocean and freshwater systems. FEMS Microbiol Ecol 33: 219-232 Rappé, M. S., S. A. Connon, K. L. Vergin, and S. J. Giovannoni. 2002. Cultivation of the ubiquitous SAR11 marine bacterioplankton clade. Nature 418:630-633 Reinthaler, T., and G. Herndl. 2005. Seasonal dynamics of bacterial growth efficiencies inr elation to phytoplankton in the southern North Sea. Aquat Microb Ecol 39:7-16 Reinthaler, T., C. Winter, and G. Herndl. 2005. Relationships between bacterioplankton richness, respiration, and production in the southern North Sea. Appl Environ Microbiol 71:2260-2266 Saiki R. K, D. H. Gelfand, S. Stoffel, S. J. Scharf, R. Higuchi, G. T. Horn, K. B. Mullis, and H. A. Ehrlich. 1988. Primer-directed Enzymatic amplification of DNA with a thermostable DNA Polymerase. Science 239:487-491 Sanger, F., G. M. Air, B. G. Barrell, N. L. Brown, A. R. Coulson, C. A. Fiddes, C. A. Hutchinson, P. M. Slocombe, and M. Smith. 1977. Nucleotide sequence of bacteriophage phi X174 DNA. Nature 265:687-695 Simon, M., H. P. Grossart, B. Schweitzer, and H. Plough. 2002. Microbial ecology of organic aggregates in aquatic ecosystems. Aquat Microb Ecol 28:175-211 Smith, D. C., M. Simon, A. L. Alldredge, and F. Azam. 1992. Intense hydrolytic enzyme activity on marine aggregates and implications for rapid particle dissolution. Nature 359:139-142 Smith, S. V., and J. T. Hollibaugh. 1993. Coastal metabolism and the oceanic organic carbon balance. Rev. Geophy. 31:75. Smith, D. C., G. F. Steward, R. A. Long, and F. Azam. 1995. Bacterial mediation of carbon fluxes during a diatom bloom in a mesocosm. Deep Sea Res 42:75-97 Staley, J. T., und A. Konopka. 1985. Measurements of in situ activities of nonphotosynthetic microorganisms in aquatic and terrestrial habitats. Annu Rev Microbiol 39: 321-346

18 Kapitel I Einleitung

Stevens, H. 2004. Heterotrophe Bakteriengemeinschaften des Deutschen Wattenmeeres – Diversität, Dynamik und Abundanz Stevens, H., T. Brinkhoff, and M. Simon. 2005a. Composition of free-living, aggregate- associated and sediment surface-associated bacterial communities in the German Wadden Sea. Aquat Microb Ecol 38:15-30 Stevens, H., M. Stübner, M. Simon, and T. Brinkhoff. 2005b. Phylogeny of Proteobacteria and Bacteroidetes from oxic habitats of a tidal flat system. FEMS Microb Ecol 54:351- 365 Søndergaard, M., and M. Middelboe. 1995. A cross-system analysis of labile dissolved organic carbon. Mar Ecol Prog Ser 118:283-294 Willms, R., H. Sass, B. Köpke, J. Köster, H. Cypionka, and B. Engelen. 2006. Specific bacterial, archaeal, and eukaryotic communities in tidal-flat sediments along a vertical profile of several meters. Appl Environ Microbiol 72: 2756-2764 Woese, C., R. 1987. Bacterial evolution. Microbiol Rev 51:221-271 Wollast, R. 1993. Interactions of carbon and nitrogen cycles in the coastal zone. In Interactions of C, N, P and S biogeochemical cycles and global change. Wollast, R., Mackenzie, F. T., and Chou, L. (eds.) Springer-Verlag, Berlin

19

II. Effects of a phytoplankton bloom in a coastal ecosystem on the composition of bacterial communities

20 Kapitel II Effects of a phytoplankton bloom on bacterial communities

Effects of a phytoplankton bloom in a coastal ecosystem on the composition

of bacterial communities

Beate Rink 1 , Susanne Seeberger 1 , Torben Martens 1 , Claus-Dieter Duerselen 2,

Meinhard Simon 1 , Thorsten Brinkhoff 1*

1 Institute for Chemistry and Biology of the Marine Environment (ICBM), University of

Oldenburg, P.O. Box 2503, D-26111 Oldenburg, Germany

2 AquaEcology, Marie-Curie-Str. 1, D-26129 Oldenburg, Germany

*Corresponding author. E-mail: [email protected]

Running head: Effects of a phytoplankton bloom on bacterial communities

KEY WORDS: Free-living and particle-attached bacteria, Bacteroidetes , Roseobacter ,

phytoplankton, DGGE

21 Kapitel II Effects of a phytoplankton bloom on bacterial communities

ABSTRACT: We studied the composition of free-living and aggregate-associated bacterial communities during the course of the phytoplankton succession in spring and early summer in the German Wadden Sea, a tidal flat ecosystem in the southern North Sea. We applied the DGGE approach based on PCR amplified 16S rRNA gene fragments, and, in addition to Bacteria -specific primers, used primers specific for alpha -Proteobacteria , the Roseobacter clade, and the Bacteroidetes phylum. Even though the application of Bacteria- and alpha - Proteobacteria -specific primers detected some changes, changes were most pronounced with the Roseobacter - and Bacteroidetes -specific primer sets. They were supported by a correspondence analysis, which showed a highly significant correlation of the DGGE banding patterns of the Roseobacter specific PCR with the composition of the phytoplankton. This indicates that changes of the phytoplankton composition in this habitat are not reflected by the patterns of the most abundant or most readily amplifiable phylotypes. The findings rather suggest that few, specialized heterotrophic bacteria are most responsive to the organic matter supplied by senescent phytoplankton and that the main part of organic matter in the German Wadden Sea is utilized by generalists. Sequence analyses of excised bands revealed a high diversity for the Bacteria- and Bacteroidetes-specific approaches. The bacterial community detected by the alpha -Proteobacteria -specific primer set, however, was mainly composed of bacteria affiliated to the Roseobacter clade.

22 Kapitel II Effects of a phytoplankton bloom on bacterial communities

INTRODUCTION

Today it is well established that heterotrophic bacteria are an important component of and key players in the biogeochemical cycling of elements and the flux of energy in aquatic ecosystems. Depending on the ecosystem and on various environmental and biotic factors the composition of the bacterial communities involved may exhibit distinct differences and variations in time and space. Temperature preferences certainly select for certain bacterial taxa but little direct information is available within this context. The most important factor for selecting specific bacterial groups is supply by specific monomeric and polymeric components of the dissolved organic matter (DOM) pool and of inorganic nutrients such as phosphate, ammonium or nitrate. It has been shown that alpha -Proteobacteria prefer monomers such as amino acids and N-acetyl-glucosamine, whereas Cytophaga/Flavobacteria (now Sphingobacteria/Flavobacteria ) of the Bacteroidetes phylum prefer polymers such as chitin and protein, and gamma -Proteobacteria amino acids and proteins (Cottrell & Kirchman 2000). Various mesocosm studies have shown that distinct DOM components via direct supply or the experimental induction of phytoplankton blooms select for specific bacterial subcommunities or populations (LeBaron et al. 1999, Pinhassi et al. 2004, Riemann et al. 2000, Schäfer et al. 2001). The specific organic matter profile of various algae appears also to be an important selection factor for distinct bacterial communities and populations evolving in the phycosphere of algae (Grossart 1999, Grossart et al. 2005, Schäfer et al. 2002). In fact, alpha -Proteobacteria , in particular the Roseobacter clade, and the Bacteroidetes appear to be most responsive to inputs of phytoplankton-born DOM (Fandino et al. 2001, Grossart et al. 2005, Pinhassi et al. 2004, Riemann et al. 2000, Schäfer et al. 2001). It is also well established that the community composition of particle-associated (PA) bacteria differs from that of free-living (FL) bacteria. Several studies have shown that Sphingobacteria and Flavobacteria preferentially colonize particles whereas alpha- and gamma -Proteobacteria mainly dwell in free-living marine bacterial communities (Fandino et al. 2001, Grossart et al. 2005, Simon et al. 2002). Our knowledge on the development and succession of specific subcommunities and populations within PA bacterial communities during phytoplankton blooms, however, is still fragmentary. Experimental studies are important to elucidate single factors affecting the composition of bacterial communities. As the aim of such studies is to better understand how the composition of bacterial communities is controlled at ambient, but much more complex conditions it is important to complement these studies by appropriate field observations. Such

23 Kapitel II Effects of a phytoplankton bloom on bacterial communities studies have been carried out in various ecosystems and shown that the composition of bacterial communities undergoes temporal changes during phytoplankton blooms (Fandino et al. 2001, Larsen et al. 2004, Yager et al. 2001). These changes often reflect the changing environmental conditions and DOM supply and also indicate which bacteria are mainly involved in the biogeochemical cycling of elements and flux of energy. Denaturing gradient gel electrophoresis (DGGE) of PCR-amplified 16S rRNA gene fragments using Bacteria - specific primers (Muyzer et al. 1993) has been proven to be a powerful tool to assess the composition and temporal changes of bacterial communities. Using Bacteria -specific primers for this approach appears to be selective against the Bacteroidetes group (Cottrell & Kirchman 2000, Selje et al. 2005, but see Castle & Kirchman 2004). Therefore, and to obtain a more detailed insight into the composition of bacterial communities and their major players, it is desirable to apply primers targeting specifically important groups such as Bacteroidetes and alpha -Proteobacteria . The aim of our study was to investigate the composition of free-living and aggregate- associated bacterial communities during the course of the phytoplankton succession in spring and early summer in the Wadden Sea, a tidal flat ecosystem of the southern North Sea. Based on previous studies, we hypothesized that the expected bacterial response to the phytoplankton succession would be reflected most pronounced by alpha -Proteobacteria and the Sphingobacteria/Flavobacteria group. Therefore, we applied the DGGE approach and, in addition to Bacteria -specific primers, primers specific for alpha -Proteobacteria , the Roseobacter clade, and the Bacteroidetes group.

MATERIALS AND METHODS

Sample collection and processing. Surface water samples were collected weekly by bucket from shipboard at high tide from 12 April to 29 June 2000 in the Backbarrier tidal flat ecosystem of the German Wadden Sea near Spiekeroog Island (53° 44.4 N, 7° 41 E). This is a mesotidal ecosystem characterized by high loads of suspended particulate matter (SPM). For further details see Stevens et al. (2005a) and Lunau et al. (2006). For analysis of SPM and the particulate carbon fractions 0.5-1 L of seawater was filtered onto pre-combusted (2 h at 550°C) and pre-weighed glass fiber filters (GF/F, Whatman, USA) and stored at –20°C in the dark until further processing. For DGGE analysis, 250 ml of seawater were pre-filtered onto 5.0 µm polycarbonate-filters (Nuclepore) to obtain the fraction of aggregate-associated and

24 Kapitel II Effects of a phytoplankton bloom on bacterial communities subsequently onto 0.2 µm polycarbonate-filters to obtain that of free-living bacteria. Filters were stored at –20°C in the dark until further processing. For enumeration of bacterial and phytoplankton cells 100 ml of water sample were fixed with formaldehyde (final concentration 2% vol/vol) or Lugol and stored at 4°C. Hydrographic data (temperature, salinity, pH, and oxygen) were measured by probes (LF 196, pH192, OXI 196, WTW, Weilheim, Germany). SPM dry weight, particulate carbon fractions. Filters were dried for 1 hour at 110°C and weighed on a micro-balance (Sartorius, Germany). Total particulate carbon (TC) and particulate inorganic carbon (PIC) were determined after high temperature combustion and titration of the CO 2 produced against Ba(ClO 4)2. Particulate organic carbon (POC) was calculated as the difference of TC and PIC. For further details see Stevens et al. (2005a). Bacterial and algal cell counts. Abundance of free-living and aggregate-associated bacteria was enumerated after DAPI (4´-6-diamidino-2-phenylindole) staining by epifluorescence microscopy at 1000x magnification according to Crump et al. (1998). To distinguish particle-attached and free-living bacteria, seawater was fractionated by filtration onto 5.0 µm and subsequently onto 0.2 µm polycarbonate-filters. To reduce the background fluorescence by inorganic matter filters were counter-stained with an acridine orange solution (0.1%). Lugol-fixed phytoplankton samples were enumerated by inverted microscopy. Phytoplankton was identified on the species level when possible. For estimating phytoplankton biomass cell numbers were multiplied by cell carbon. The latter was estimated from measured cell sizes of individual cells converted to carbon according to empirical carbon/cell volume conversion factors from the Biologische Anstalt Helgoland (J. Berg, unpubl. results). Nucleic acid extraction. The isolation of genomic DNA was performed by phenol- chloroform extraction after bead beating as described earlier with slight modifications (Selje & Simon, 2003). The precipitation was done overnight at –20°C using isopropanol. The DNA was resuspended in molecular grade water (Eppendorf, Germany) and stored at –20°C until further processing. Primer sets. PCR amplification of 16S rRNA gene fragments was performed with primer pairs specific for Bacteria (GC 341F and 907RM), the Bacteroidetes phylum (GC-CF319f and 907RM), alpha -Proteobacteria (GC 341F and ALF968r), and the Roseobacter clade within alpha -Proteobacteria (GC ROSEO536Rf and GRb735r). Primer sequences and references are given in Table 1. ’GC’ indicates that a GC clamp was added to the primer (Muyzer et al. 1993). For the primer GC ROSEO536Rf the following GC clamp was used: 5’-

25 Kapitel II Effects of a phytoplankton bloom on bacterial communities

CGCCCGCCGCGCCCCGCGCCCGTCCCGCCGCCCCCGCCCG-3’. For the sequences of the other GC clamps used in this study see the references cited in Table 1. Specificity of the primers used for Bacteroidetes was described earlier (Jaspers et al. 2001, Kirchman 2002). The oligonucleotide probe ALF968r (Neef, 1997), used as reverse primer for alpha - Proteobacteria , was tested theoretically using the BLAST function of the NCBI server (http://www.ncbi.nlm.nih.gov). Search results for this primer sequence revealed up to 10% matches to other phylogenetic groups with 100% sequence similarity for the first one hundred matches. The primer set used for the Roseobacter -group was tested theoretically with the whole database of the ARB software package (Ludwig et al. 2004) and recently published sequences present in GenBank (www.ncbi.nlm.nih.gov) of cultivated and uncultivated organisms affiliated with the Roseobacter clade. In total 183 sequences affiliated with this group were considered. Specificity was also tested in PCR assays using several described species as positive and negative controls (Table 2), and 25 isolates affiliated with the Roseobacter group, taken from our culture collection. PCR amplification of 16S rRNA gene fragments. PCR amplifications were performed with an Eppendorf Mastercycler (Eppendorf, Hamburg, Germany) as follows: One µl of template was added to 49 µl of PCR mixture containing 1 U of Sigma RedTaq TM polymerase and 5 µl 10 x RedTaq TM PCR buffer (Sigma, Deisenhofen, Germany), bovine serum albumin -1 (10 mg ml ), 250 µM of each deoxynucleotide triphosphate, 2.1 µM MgCl 2, and 20 pmol of each primer. The PCR protocol for the Bacteria -specific primer set was performed as described previously (Brinkhoff & Muyzer, 1997). Amplification of the 16S rRNA gene fragments of alpha -Proteobacteria was performed under the same conditions with an annealing temperature of 65°C for 10 cycles and subsequently 55°C for 20 cycles. Roseobacter -specific PCR conditions were 5 cycles at 65°C and 25 cycles with an annealing temperature of 63°C. For highest specificity, a maximum of 30 cycles is recommendable at this step. PCR with the primer set specific for Bacteroidetes was performed as described previously (Jaspers et al. 2001). Four µl of the amplification products were analyzed by electrophoresis in 2% (w/v) agarose gels and stained with ethidium bromide (1 µg ml -1) (Sambrook et al. 1989). For subsequent sequencing analysis PCR products were purified by using the Qiaquick PCR purification kit (Qiagen Inc., Chatsworth, Calif.). DGGE analysis of PCR products. DGGE was performed with the D-Code system (Bio- Rad Laboratories, Inc.). For gene fragments of Bacteria and alpha -Proteobacteria , the protocol described by Brinkhoff & Muyzer (1997) was used. For 16S rRNA gene fragments obtained with the primer pair GC-CF319f and 907 RM the gradient was modified to 15 to

26 Kapitel II Effects of a phytoplankton bloom on bacterial communities

85% denaturant. DGGE analysis of Roseobacter 16S rRNA gene fragments was performed with 20 to 70% denaturant and 9% (wt/vol) polyacrylamide content. After electrophoresis, the gels were stained with SYBR Gold (Molecular Probes, Inc.) and photographed using a BioDoc Analyze Transilluminator (Biometra, Göttingen, Germany). Bands were excised with a scalpel sterilized with ethanol and transferred to sterile Eppendorf caps. Fifty µl of water (molecular grade, Eppendorf, Germany) were added and the samples were stored at –20°C. Cloning. Twenty four DGGE bands (GWS-e1-FL to GWS-e13-PA, GWS-c3-FL, GWS- c16-PA, GWS-c9-PA, GWS-c10-PA, GWS-c18-PA and GWS-a10-PA to GWS-a13-PA, GWS-a4-FL, GWS-a8-FL) were cloned using the pGEM ®-T Vector System II (Promega, Madison, USA) following the instruction manual. Clones with inserts were picked, resuspended in molecular grade water (Eppendorf, Germany) and screened by DGGE to check if the insert position matches the position of the corresponding DGGE band. Adequate clones were amplified and subsequently sequenced using the primers pUC/M13f and pUC/M13r (Messing, 1983) with an annealing temperature of 48°C. Sequencing and phylogenetic analysis. PCR products were sequenced using the DYEnamic Direct cycle sequencing kit (Amersham Life Science, Inc.) and a Model 4200 Automated DNA Sequencer (LI-COR, Inc.). Sequencing primers were 341F and 907RM for direct sequencing of DGGE bands, or M13 primers as described above for cloned bands labeled with IRDye TM 800. For all sequences, at least 400 bp were determined. Phylogenetic affiliation of the sequences was compared to those in GenBank using the BLAST function of the NCBI server (http://www.ncbi.nlm.nih.gov/BLAST/). Phylogenetic trees were constructed using the ARB software package (Ludwig et al. 2004, http://www.arb-home.de). The backbone tree was calculated with the maximum likelihood method using sequences with a minimum length of 1300 bp including type strains of the selected phylogenetic groups. For tree calculation, positions were excluded at which less than 50% of all sequences showed the same residues to avoid uncertain alignments. Sequences with less than 1300 bp were added to the backbone tree with the maximum parsimony method using the same filter. As an outgroup, 16S rRNA gene sequences of seven type strains belonging to Cyanobacteria were used. The sequences obtained in this study are available from GenBank under accession no. DQ080919 to DQ080962. Statistics. Cluster analyses of DGGE banding patterns were performed using Gel Compar II, version 2.5 (Applied maths, Kortrijk, Belgium). Calculations were done curve based using Pearson correlation and UPGMA. A correspondence analysis of the DGGE

27 Kapitel II Effects of a phytoplankton bloom on bacterial communities banding patterns and the phytoplankton composition was performed using ADE-4 (Thioulouse et al. 1997). To analyze the bacterial community structure, we exported the raw data of the cluster analysis and generated a matrix based on the specific band heights. For phytoplankton, we used relative species abundance. A modified correspondence analysis was performed row weighted on a biplot scale. After calculation of the COA for each community a Coinertia analysis was performed to connect the data. A permutation test based on the Monte Carlo method was calculated using the Coinertia test (– Fixed D; number of random matching: 1000).

RESULTS

Environmental conditions and SPM properties

From the start of the study period in mid-April until 10 May 2000 the water temperature continuously increased from 8-17 °C (Fig. 1A). Thereafter it fluctuated between 17 and 13 °C. Salinity ranged between 29 and 32‰ (Fig. 1A) and SPM dry weight from 80-120 mg l -1 in April and May, but increased to 160 mg L -1 on 14 June (data not shown). POC concentrations varied between 0.8 mg L -1 on 3 May and 4.7 mg L -1 on 26 April (Fig. 1B). They steadily increased from 3 to 17 May and from 24 May to 14 June.

Phytoplankton and bacterial dynamics

The phytoplankton consisted exclusively of diatoms and few dinoflagellates. From 12 April to 3 May diatom cell numbers strongly decreased from 6.5x10 3-1.2x10 3 L -1 but thereafter continuously increased until 24 May (Fig. 1C). After the decline of this bloom in late May only low numbers remained. Whereas the initial bloom on 12 April exhibited a high diversity and evenness, the bloom in May became more and more dominated by Guinardia delicatula , constituting 70% of algal cell numbers and biomass on 24 May (Fig. 1C). One week later, when diatom cell numbers declined to ~30% of the previous week, abundance of Guinardia delicatula had strongly decreased while Pseudonitzschia pungens constituted 50% of the cell numbers. At the onset of the study phytoplankton constituted 50% of POC, but on 26 April only 4%. Thereafter during the Guinardia bloom, phytoplankton carbon continuously increased to 13% on 24 May (Fig. 1B).

28 Kapitel II Effects of a phytoplankton bloom on bacterial communities

Cell numbers of FL bacteria increased from 12 April until 3 May from 2.4x10 6-4.0x10 6 ml -1 and fluctuated thereafter around this value (Fig. 1D). Cell numbers of PA bacteria were lower and ranged from 0.5-1.1x10 6 ml -1, accounting for 11 to 22% of total bacterial numbers.

Specificity of the Roseobacter primer set

Comparison of 16S rRNA gene fragments present in our ARB database revealed that the forward primer GC ROSEO536Rf matched 131 of a total of 183 target sequences affiliated to the Roseobacter clade. Forty three Roseobacter sequences had no or incomplete information at the target site of the primer, and 8 sequences of uncultured Roseobacter -affiliated organisms showed up to three mismatches to the primer sequence. Sulfitobacter pontiacus (Acc. no. Y13155) had one mismatch at position 17 of the primer sequence. Reverse primer GRb735r targeted 133 sequences after insertion of a wobble (G/A) at Escherichia coli position 752. Forty four Roseobacter sequences had no or incomplete information at the target site of the 16S rRNA gene. Six sequences of target bacteria had up to three mismatches: Roseobacter sp. J8W (AF026462; two mismatches), Roseobacter sp. J2W (AF026462, three mismatches), Roseobacter sp. KT11117 (AF173971, one mismatch), Adriatic 72 (AF030780, one mismatch, two non-defined bases), Sulfitobacter pontiacus (Y13155, one mismatch) and GWS-BW-H66M (AY515422, one mismatch). The non-target sequences of Rhodovulum iodosum and clone SAR102 (Acc. no. L35460) had no mismatches to the primer sequence. Considering all respective sequences in the ARB data base, the use of both primers paired resulted in at least one mismatch to all other phylogenetic groups. PCR results showed that the specificity and sensitivity of the Roseobacter primer set was very high. With one step down from 65°C -63°C and 1 U of Taq polymerase, 0.2 ng genomic DNA µl -1 of Roseobacter gallaeciensis was detectable. DNA of the non-target organism Paracoccus aminophilus (1 mismatch to the target sequence) was detected down to 2 ng µl -1. Specificity was higher but less sensitive under the same conditions with 0.5 U of polymerase, detecting 0.1-1 ng DNA µl -1 of R. gallaeciensis and 20 ng DNA µl -1 of P. aminophilus . To determine a possible sequence preference of the primer set, a DNA mixture of both organisms with equal DNA amounts was amplified and the PCR products were analyzed using DGGE. By this approach only amplicons of R. gallaeciensis could be detected (data not shown). This notion suggests that the amplification of non-target organisms is suppressed under the chosen PCR and DGGE conditions.

29 Kapitel II Effects of a phytoplankton bloom on bacterial communities

DGGE banding patterns

The DGGE analyses with the various primer sets showed distinctly different banding patterns of the FL as well as PA bacterial communities (Fig. 2A-D). The Bacteria -specific primer set yielded 12 to 15 bands per lane in the PA bacterial fraction and 12 to 18 bands in the FL bacterial fraction (Fig. 2A). Changes in the banding patterns occurred mainly during the Guinardia bloom in May, showing a slight increase of band numbers in the FL fraction on 17 May and the appearance of a strong band in the PA fraction (GWS-e11-PA). The cluster analysis yielded distinct clusters for the FL and PA bacterial communities (Fig. 3A). Only during the Guinardia bloom between 10 and 24 May the PA bacteria clustered separately and closer with the FL bacterial community. Correspondence analysis did not yield a significant correlation with the phytoplankton composition. The aggregate sample of 31 May was reamplified from a former PCR product and showed reduced band numbers compared to the other samples. Hence, the fingerprint of this sample appeared as an outgroup in the cluster analysis and was not regarded for the further discussion. The Bacteroidetes –specific banding patterns revealed 7-12 and 9-18 amplicons per lane in the FL and PA bacterial fractions, respectively (Fig. 2B). Low numbers of 7-9 bands occurred in the FL bacterial fraction before and after the Guinardia bloom and higher numbers of 10-12 bands during the bloom. In contrast, the number of bands in the PA bacterial community was high before and after and decreased during the bloom. The cluster analysis showed a distinct cluster of the FL bacterial community, excluding the dates towards the end of the Guinardia bloom, when banding patterns clustered together with those of the PA bacterial community during the bloom (Fig. 3B). Furthermore, the latter fraction exhibited clearly different patterns before and after the bloom. A correspondence analysis revealed a moderate correlation of the banding patterns with the composition of the phytoplankton (P=0.067). DGGE banding patterns specific for alpha -Proteobacteria showed 7-8 and 8-13 bands per lane in the FL and PA bacterial communities, respectively (Fig. 2C). Most of the bands were permanently present but a few bands in both fractions occurred in the course of the bloom (GWS-a11-PA, GWS-a6-FL, and GWS-a5-FL). The cluster analysis showed generally rather complex patterns and that PA bacteria during the Guinardia bloom clustered together with FL bacteria (Fig. 3C). The correspondence analysis did not yield a significant correlation of the banding patterns with the phytoplankton composition.

30 Kapitel II Effects of a phytoplankton bloom on bacterial communities

The Roseobacter -specific DGGE banding patterns showed 5-8 bands per lane in the FL bacterial fraction and 8-18 bands in the PA bacterial fraction (Fig. 2D). Quite a few bands were permanently present in both fractions, but additional bands occurred during the decline of the bloom in April and the Guinardia bloom in May, mainly in the PA bacterial fraction. The cluster analysis yielded complex patterns with several subclusters both of FL and PA associated bacterial fractions. A distinct subcluster comprised the banding patterns of both fractions during the Guinardia bloom (Fig. 3D). The correspondence analysis showed a highly significant correlation of the banding patterns with the composition of the phytoplankton (P=0.03). Phylogenetic affiliation

The sequence analysis of excised bands revealed a high diversity of the obtained phylotypes for the 16S rRNA gene fragments of the Bacteria- and Bacteroidetes-specific approaches (sequences obtained with Bacteria -specific primers were designated GWS-e and sequences obtained with Bacteroidetes -specific primers as GWS-c; Figs. 4A, B). The bacterial community detected by the alpha -Proteobacteria -specific primer set (sequences GWS-a) was mainly composed of bacteria belonging to the Roseobacter clade. Most phylotypes of this group, detected by the Bacteria - and alpha-Proteobacteria -specific primer set, clustered within the recently described WAC I cluster (Stevens et al. 2005b) or RCA cluster (Selje et al. 2004). The primer set used for alpha -Proteobacteria turned out to be not specific, as sequencing results revealed that two sequences affiliated to delta -Proteobacteria (GWS-a12-PA, GWS-a13-PA) and one to Bacteroidetes (GWS-a8-FL). In contrast, although the primer GC-CF319f used for specific amplification of 16S rRNA gene sequences of bacteria belonging to Bacteroidetes is known to be unspecific (Kirchman et al. 2003), all our phylotypes of the sequenced bands fell into this phylum. During the Guinardia bloom, DGGE derived phylotypes belonging to the WAC I cluster dominated the FL bacterial fraction. DGGE band GWS-e7-FL was present during the bloom. This phylotype was closely related to GWS-a6-FL and GWS-a5-FL (sequence differences <0.8%, Fig. 4A) which were also present only during the Guinardia bloom. While these organisms seem to be highly responsive to the phytoplankton composition, other members of the WAC I cluster were present during the whole investigation period, e.g. GWS-e6-FL (Fig. 4A). This phylotype is closely related to DGGE band GWS-FL-3, which was persistently detected throughout the year in the Wadden Sea, indicating that this organism is well adapted to highly variable biotic and environmental conditions in this habitat (Stevens et al. 2005a). In

31 Kapitel II Effects of a phytoplankton bloom on bacterial communities the PA bacterial fraction chloroplast DNA (GWS-e11-PA) represented the most significant change within the community detected by the Bacteria -specific primer set. Sequencing of other conspicuous bands was not possible, as the diffuse bands in the upper part of the gel could not be reamplified. One of the Bacteroidetes –specific phylotypes appeared during 24 May to 14 June at the end of the Guinardia bloom both in the FL and PA bacterial fractions (GWS-c6-FL, GWS-c5- FL and GWS-c15-PA). This phylotype is closely related to GWS-e9-FL in the FL bacterial fraction (Fig. 4B). BLAST results revealed that the closest related sequence of these bands is DGGE band GWS-AG-8, which was detected on aggregates in June 2000 in the same area (Stevens et al. 2005a). Other phylotypes affiliated to the Bacteroidetes were present during the whole investigation period. DGGE band GWS-c8-PA was detected in the PA bacterial fraction from April to June and is closely related to strain T15 (AY177723, 99% similarity, 502/505 bp), isolated from the same habitat in October 1999 (Brinkhoff et al. 2004). The phylotype was also detected in a seasonal sampling campaign in this habitat from 2000 to 2002 using GC-CF319f and 907RM for DGGE analysis (S. Seeberger, unpublished results).

DISCUSSION

Our results indicate that the composition of the bacterial communities in the Wadden Sea underwent changes during the phytoplankton succession in spring and early summer. These changes were detected as the disappearance of DGGE bands and the appearance of new ones, were most pronounced during the Guinardia bloom and its decline in May, and occurred in the FL as well as in the PA bacterial communities. Even though the application of Bacteria- and alpha -Proteobacteria -specific primers in the DGGE approach detected some of these changes, they were detected most clearly with the Roseobacter - and Bacteroidetes -specific primer sets and supported by a correspondence analysis. Whereas the number of bands of the FL bacterial fraction within the Bacteroidetes increased during the Guinardia bloom it decreased in the PA bacterial fraction. Within the Roseobacter clade, the number of bands of the PA bacterial fraction increased during the decline of both blooms, in late April and late May. Hence our results show that the bacterial communities respond to the changing phytoplankton community and organic matter field on a rather specific phylogenetic level and call for applying class- and subclass-specific primer sets in the DGGE approach for such investigations.

32 Kapitel II Effects of a phytoplankton bloom on bacterial communities

Our investigation complements mesocosm experiments which obtained similar findings (LeBaron et al. 1999, Pinhassi et al. 2004, Riemann et al. 2000, Schäfer et al. 2001) and demonstrate that changes in FL as well as PA bacterial communities during the development of phytoplankton blooms also occur and can be detected at ambient conditions in a natural ecosystem. Our results, however, go beyond the mentioned experimental studies by showing in a much more detailed way the different responses of bacteria affiliated to the Roseobacter clade and the Bacteroidetes . To achieve these results we applied sets of published primers for all bacterial target groups and modified specific probes for the Roseobacter clade from earlier studies to optimize its specificity (Table 1). In addition, we developed a PCR to achieve highest specifity for this primer set. As expected from the BLAST search, the results revealed that the alpha -Proteobacteria -specific primer set was not specific. We detected two sequences affiliated to delta -Proteobacteria and one to the Bacteroidetes phylum, indicating that sequencing of bands is essential when applying this primer set. However, the great majority of the bands sequenced affiliated to alpha -Proteobacteria and exclusively to the Roseobacter clade, underscoring the significance to apply a primer set specific for this clade. In this study, the Bacteroidetes-specific primer set, in fact, was specific, as all sequenced bands affiliated to the respective target group. The fact that most changes within the bacterial communities were only detected by applying primer sets for a lower phylogenetic level indicates that these changes do not affect the most abundant or most readily amplifiable phylotypes. Although various prominent bands were visible using the group-specific primer set, only two Bacteroidetes affiliated phylotypes could be detected with the Bacteria -specific primer set. Only one of these two phylotypes was also detected with the group-specific primer set (GWS-e9-FL, Fig. 4B) suggesting that the Bacteria -specific primer set discriminates the Bacteroidetes affiliated bacteria, as has been reported previously (Cottrell & Kirchman, 2000). In contrast, we had no indication of a biased amplification of phylotypes affiliated to alpha -Proteobacteria by the Bacteria -specific primer set. Seven of the 17 sequences of this subclass were amplified by the Bacteria -specific primer set and all except one sequence (GWS-e10-FL) were very closely related or similar to those amplified by the alpha - Proteobacteria -specific primer set (Fig. 4A). However, the latter and the Roseobacter - specific primer set yielded a much better resolution and detected substantially more phylotypes with a presumably lower abundance. The decreasing number of bands of Bacteroidetes in the PA bacterial fraction detected during the Guinardia bloom and its decline indicates the formation of a more specialized

33 Kapitel II Effects of a phytoplankton bloom on bacterial communities environment on suspended particles and aggregates to which fewer bacteria of this phylogenetic group were able to adapt. The number of bands of the Roseobacter clade, however, remained rather unchanged and even increased at the decline of the bloom on 31 May, indicating that the niche diversity for this bacterial group did not decrease. In contrast to aggregates, the number of bands of FL bacterial phylotypes belonging to the Bacteroidetes increased during the bloom, indicating that the DOM supply became more diverse, presumably including a variety of polymers released from growing and decaying diatoms and solubilizing phytodetrital aggregates. Two of the newly occurring phylotypes cluster together (GWS-c6-FL, GWS-c5-FL) and also together with other phylotypes which occurred on 14 June (GWS-e9-FL) in the FL bacterial fraction and in the PA bacterial fraction (GWS-c15-PA, Fig. 4B). These phylotypes are closely related to phylotypes which were retrieved from the associated bacterial communities of two diatoms [SB-42-DB (Schäfer et al. 2001), Flo-21 (Grossart et al. 2005)], suggesting that they are particularly adapted to the organic matter profile of diatoms. Our results show that organisms of the Roseobacter clade and the Bacteroidetes are most responsive to the changing organic matter field during the phytoplankton blooms. This is in line with other studies (Dang & Lovell, 2002, Lebaron et al. 1999, Riemann et al. 2000, Fandino et al. 2001, Pinhassi et al. 2004, Grossart et al. 2005) and thus indicates that members of these two bacterial groups appear to be particularly adapted to such conditions, at least in temperate waters. The significance of the Bacteroidetes affiliated bacteria in consuming complex and polymeric DOM in marine systems is well known and fairly well understood, mainly because of their specific properties to hydrolyze polymers (Cottrell & Kirchman, 2000; Kirchman, 2002). The significance of the Roseobacter clade is much less understood. Some members of this clade exhibit aerobic anoxygenic photosynthesis but the significance of this metabolic pathway at ambient conditions and varying trophic state is still unclear (Schwalbach & Fuhrman, 2005). Other members of this clade appear to be involved in the decomposition of DMS (Moran et al. 2003). Roseobacter strains have been isolated from FL as well as PA bacterial communities and quite a few of them exhibit antibiotic and quorum sensing properties (Long & Azam 2001, Gram et al. 2002; Grossart et al. 2004). Hence, it appears conceivable that varying adaptive properties make this clade well suitable to dwell successfully in marine systems. More work, however, is needed to better understand the success of this clade at a physiological and genetic level. The combined application of Bacteria - and group-specific primer sets revealed that a hierarchical structure exists in the bacterial communities, both in the FL as well as the PA

34 Kapitel II Effects of a phytoplankton bloom on bacterial communities fractions. The Bacteria -specific primer set detected mainly those phylotypes which constitute the main and often dominant components of the bacterial communities, persisting throughout most of the time and thus comprising bacteria able to adapt to rather variable environmental conditions and exhibiting a rather generalistic life style. These phylotypes include members of the RCA-cluster (Selje et al. 2004), and the WAC I cluster (Stevens et al. 2005b) of the Roseobacter clade. In contrast, the group- and clade-specific primer sets detect, besides some of these generalistic phylotypes, others which are probably less abundant but appear at distinct environmental and biotic conditions, such as during certain periods of phytoplankton blooms. The phylotypes detected by these primer sets reflect in a much more sensitive way these changing conditions and thus allow a more detailed analysis of bacterial communities at varying environmental conditions. The application of Bacteria -specific primer sets appears to be appropriate to study the main components of bacterial communities and their variability at greatly varying environmental conditions such as in salinity gradients (Selje & Simon, 2003; Troussellier et al. 2002), PA vs. FL bacterial communities (Stevens et al. 2005a), or in manipulated mesocosms (Lebaron et al. 1999, Pinhassi et al. 2004, Riemann et al. 2000). In other cases when more subtle variations or discrimination against specific target groups may occur, this approach appears not sensitive enough to comprehensively detect these changes. Then the application of more specific primer sets is a valuable tool to detect these changes which are an important indication of distinct responses of the bacterial communities to their changing environment.

Acknowledgements . We appreciate the hospitality and assistance of the RV Senckenberg crew. We thank B. Kuerzel and R. Weinert for dry weight analyses and H.-P. Grossart for valuable discussions. This work was supported by the Deutsche Forschungsgemeinschaft (DFG) within the research group “BioGeoChemistry of Tidal Flats” (FG 432 TP5).

35 Kapitel II Effects of a phytoplankton bloom on bacterial communities

LITERATURE CITED

Brinkhoff T, Muyzer G (1997) Increased species diversity and extended habitat range of sulfur-oxidizing Thiomicrospira spp. Appl Environ Microbiol 63: 3789-3796 Brinkhoff T, Bach G, Heidorn T, Liang L, Schlingloff A, Simon M (2004) Antibiotic production by a Roseobacter clade-affiliated species from the German Wadden Sea and its antagonistic effects on indigenous isolates. Appl Environ Microbiol 70: 2560-2565 Brinkmeyer R, Rappé M, Gallacher S, Medlin L (2000) Development of clade- ( Roseobacter and Alteromonas ) and taxon- specific oligonucleotide probes to study interactions between toxic dinoflagellates and their associated bacteria. Eur J Phycol 35: 315-329 Castle D, Kirchman DL (2004) Composition of estuarine bacterial communities assessed by denaturing gradient gel electrophoresis and fluorescence in situ hybridization. Limnol Oceanogr Methods 2: 303-314 Cottrell MT, Kirchman DL (2000) Natural assemblages of marine proteobacteria and members of the Cytophaga-Flavobacter cluster consuming low- and high-molecular- weight dissolved organic matter. Appl Environ Microbiol 66: 1692-1697 Crump BC, Baross JA, Simenstad CA (1998) Dominance of particle-attached bacteria in the Columbia River estuary, USA. Aquat Microb Ecol 14: 7-18 Dang H, Lovell CR (2002) Seasonal dynamics of particle-associated and free-living marine Proteobacteria in a salt marsh tidal creek as determined using fluorescence in situ hybridization. Environ Microbiol 4:287-295 Fandino LB, Riemann L, Steward GF, Long RA, Azam F (2001) Variations in bacterial community structure during a dinoflagellate bloom analyzed by DGGE and 16S rDNA sequencing. Aquat Microb Ecol 23:119-130 Giuliano L, De Domenico M, De Domenico E, Höfle MG, Yakimov MM (1999) Identification of culturable oligotrophic bacteria within naturally occurring bacterioplankton communities of the Ligurian Sea by 16S rRNA sequencing and probing. Microb Ecol 37:77-85 Gram L, Grossart HP, Schlingloff A, Kiorboe T (2002) Possible quorum sensing in marine snow bacteria: production of acylated homoserine lactones by Roseobacter strains isolated from marine snow. Appl Environ Microbiol 68:4111-4116 Grossart HP (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 Grossart HP, Schlingloff A, Bernhard M, Simon M, Brinkhoff T (2004a) Antagonistic activity of bacteria isolated from organic aggregates of the German Wadden Sea. FEMS Microbiol Ecol 47:387-396 Grossart HP, Levold F, Allgaier M, Simon M, Brinkhoff T (2005) Marine diatom species harbour distinct bacterial communities. Environ Microbiol 7:860-873 Jaspers E, Nauhaus K, Cypionka H, Overmann J (2001) Multitude and temporal variability of ecological niches as indicated by the diversity of cultivated bacterioplankton. FEMS Microbiol Ecol 36:153-164 Kirchman DL (2002) The ecology of Cytophaga-Flavobacteria in aquatic environments. FEMS Microbiol Ecol 39:91-100

36 Kapitel II Effects of a phytoplankton bloom on bacterial communities

Kirchman DL, Yu L, Cottrell MT (2003) Diversity and abundance of uncultured Cytophaga - like bacteria in the Delaware estuary. Appl Environ Microbiol 69:6587-6596 Larsen A, Flaten GAF, Sandaa RA, Castberg T, Thyrhaug R, Erga SR, Jacquet S, Bratbak G (2004) Spring phytoplankton bloom dynamics in Norwegian coastal waters: Microbial community succession and diversity. Limnol Oceanogr 49:180-190 Lebaron P, Servais P, Troussellier M, Courties C, Vives-Rego J, Muyzer G, Bernard L, Guindulain T, Schäfer H, Stackebrandt E (1999) Changes in bacterial community structure in seawater mesocosms differing in their nutrient status. Aquat Microb Ecol 19:255-267 Long RA, Azam F (2001) Antagonistic interactions among marine pelagic bacteria. Appl Environ Microbiol 67:4975-4983 Ludwig W, Strunk O, Westram R et al. (2004) ARB: a software environment for sequence data. Nucl Acids Res 32:1363-1371 Lunau M, Lemke A, Dellwig O, Simon M (2006) Physical and biogeochemical controls of microaggregate dynamics in a tidally affected coastal ecosystem. Limnol Oceanogr 51:847-859 Manz W, Amann R, Ludwig W, Vancanneyt M, Schleifer KH (1996) Application of a suite of 16S rRNA-specific oligonucleotide probes designed to investigate bacteria of the phylum cytophaga-flavobacter-bacteroides in the natural environment. Microbiology 142:1097-1106 Messing J (1983) New M13 vectors for cloning. Meth Enzymol 101:20-78 Moran MA, Gonzalez JM, Kiene RP (2003) Linking a bacterial taxon to sulfur cycling in the sea: Studies of the marine Roseobacter group. Geomicrobiol J 20:375-388 Muyzer G, de Waal EC, Uitterlinden AG (1993) Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction- amplified genes coding for 16S rRNA. Appl Environ Microbiol 59:695-700 Neef A (1997) Anwendung der in-situ-Einzell-Identifizierung von Bakterien zur Populationsanalyse in komplexen mikrobiellen Biozönosen. PhD Thesis, Technische Universität München. Pinhassi J, Sala MM, Havskum H, Peters F, Guadayol O, Malits A, Marrase CL (2004) Changes in bacterioplankton composition under different phytoplankton regimens. Appl Environ Microbiol 70:6753-6766 Riemann L, Steward GF, Azam F (2000) Dynamics of bacterial community composition and activity during a mesocosm diatom bloom. Appl Environ Microbiol 66:578-587 Sambrook J, Frisch EF, Maniatis T (1989) Northern Hybridisation. In Molecular Cloning: a laboratory manual, 2nd edn., Cold Spring Harbour Laboratory Press, New York Schäfer H, Bernard L, Courties C, et al. (2001) Microbial community dynamics in Mediterranean nutrient-enriched seawater mesocosms: changes in the genetic diversity of bacterial populations. FEMS Microbiol Ecol 34:243-253 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 Schwalbach MS & Fuhrman JA (2005) Wide-ranging abundances of aerobic anoxygenic phototrophic bacteria in the world ocean revealed by epifluorescence microscopy and quantitative PCR . Limnol Oceanogr 50:620-628

37 Kapitel II Effects of a phytoplankton bloom on bacterial communities

Selje N, Simon M (2003) Composition and dynamics of particle-associated and free-living bacterial communities in the Weser estuary, Germany. Aquat Microb Ecol 30:221-237 Selje N, Simon M, Brinkhoff T (2004) A newly discovered Roseobacter cluster in temperate and polar oceans. Nature 427:445-448 Selje N, Brinkhoff T, Simon M (2005) Detection of abundant bacteria in the Weser estuary using culture-dependent and culture-independent approaches. Aquat Microb Ecol 39:17- 34 Simon M, Grossart HP, Schweitzer B, Ploug H (2002) Microbial ecology of organic aggregates in aquatic ecosystems. Aquat Microb Ecol 28:175-211 Stevens H, Brinkhoff T, Simon M (2005a) Composition of free-living, aggregate-associated and sediment surface-associated bacterial communities in the German Wadden Sea. Aquat Microb Ecol 38:15-30 Stevens H, Stübner M, Simon M, Brinkhoff T (2005b) Phylogeny of Proteobacteria and Bacteroidetes from oxic habitats of a tidal flat ecosystem. FEMS Microbiol Ecol 54:351-365 Thioulouse J, Chessel D, Dolédec S, Olivier JM (1997) ADE-4: a multivariate analysis and graphical display software. Statistics and Computing 7:75-83 Troussellier M, Schäfer H, Batailler N, Bernard L, Courties C, Lebaron P, Muyzer G, Servais P, Vives-Rego J (2002) Bacterial activity and genetic richness along an estuarine gradient (Rhone River plume, France). Aquat Microb Ecol 28:13-24 Yager PL, Connelly TL, Mortazavi B, Wommack KE, Bano N, Bauer JE, Opsahl S, Hollibaugh JT (2001) Dynamic bacterial and viral response to an algal bloom at subzero temperatures. Limnol Oceanogr 46:790-801

38 Kapitel II Effects of a phytoplankton bloom on bacterial communities

Table 1. Primers used in this study. ______Primer Sequence (5´- 3´) E. coli 16S Target group Reference rRNA position ______GC-341F CCTACGGGAGGCAGCAG 341 – 358 Bacteria (Muyzer et al. 1993) 907RM CCGTCAATTCMTTTGAGTTT 907 – 924 Universal (Muyzer 1998) GC-CF319f GTACTGAGACACGGACCA 319 – 336 Bacteroidetes (Manz et al. 1996) ALF968r GGTAAGGTTCTGCGCGTT 968 – 985 alpha-Proteobacteria (Neef 1997) GC-ROSEO536Rf CGGAGGGGGTTAGCGTTG 536 – 553 Roseobacter clade (Brinkmeyer et al. 2000) GRb735r a GTCAGTATCGAGCCAGT(G/A)AG 735 – 754 Rhodobacter group (Giuliano et al. 1999) ______a modified

Table 2. Phylogenetic affiliation of strains and species used for the specificity test of the Roseobacter primer set. Strains are from the culture collection of our lab or the DSMZ: German collection for cell cultures and microorganisms. DSMZ strain numbers are given in parenthesis.

Class Strain or species (DSM No.) Acc. No. alpha -Proteobacteria a TL AY177716 (Roseobacter clade) T5 AY177712 T11 AY177714 TY AY841788 D1 AY841770 D4 AY841771 HP12 AY841769 HP14w AY841773 HP29w AY239008 HP30 AY239009 HP32 AY841774 HP37 AY239010 HP44 AY841765 HP47 AY841776 HP50 AY841778 ROS2 AY841779 ROS4 AY841780 ROS7 AY841781 ROS8 AY841782 AP-27 AY145564 H43-35 AY841784 H55 AY841765 GWS-BW-H22M AY515407 GWS-BW-H66M AY515422 GWS-BW-H71M AY515423 Roseobacter gallaeciensis (12440) Y13244 Roseobacter denitrificans (7001) M59063 Ruegeria algicola (10251) X78315 Ruegeria gelatinovorans (5887) D88523 Roseovarius tolerans (11457) Y11551 Leisingera methylohalidivorans (14336) AY005463 Sulfitobacter pontiacus (10014) Y13155

39 Kapitel II Effects of a phytoplankton bloom on bacterial communities

Table 2 cont.

Class Strain or species (DSM No.) Acc. No. alpha -Proteobacteria b Paracoccus aminophilus (8538) Y16929 beta -Proteobacteria b Aquaspirillum delicatum (11558) AF078756 Burkholderia pyrrocinia (10685) AB021369 Sphaerotilus natans (6575) L33980 gamma -Proteobacteria b Pseudomonas putida (548) AF094741 Pseudeo alteromonas atlantica (6839) X82134 Fundibacter jadensis (12178) AJ001150 delta -Proteobacteria b Desulfococcus multivorans (2059) AF418173 Desulfobulbus mediterraneus (13871) AF354663 Pelobacter venetianus (2394) U41562 Flavobacteria b Muricauda ruestringensis (13258) AF218782 Bacilli b Bacillus marinus (1297) AJ237708 Bacillus subtilis (7, 10) AJ276351 Lactobacillus plantarum (20205) n. a. Actinobacteria b Streptomyces violaceoruber (40701) n. a. Streptomyces glaucescens (40155) D44092 Streptomyces antibioticus (40715) n. a. Arthrobacter nicotinovorans (420) X80743 a positive control b negative control n. a. : not available

40 Kapitel II Effects of a phytoplankton bloom on bacterial communities

Figure legends

Fig. 1. Temperature and salinity (A), total particulate carbon, particulate organic carbon (POC) and phytoplankton carbon as % of POC (B), phytoplankton cell counts and species composition (C), and abundance of particle-attached and free-living bacteria (D) in the German Wadden Sea from 12 April to 29 June 2000.

Fig. 2. DGGE fingerprints of the free-living (FL) and particle-attached (PA) bacterial communities of the German Wadden Sea from 12 April to 29 June 2000 using primer sets for 16S rRNA genes of Bacteria (EUB; A), Bacteroidetes (CFB; B), alpha-Proteobacteria (ALF; C) and the Roseobacter clade (ROS; D). The numbered arrows mark excised and sequenced bands. Because of the small fragment size of the Roseobacter amplicons the DGGE bands were not excised for sequencing.

Fig. 3. Cluster analyses of the DGGE banding patterns of particle-attached (PA) and free- living (FL) Bacteria (EUB; A), Bacteroidetes (CFB; B), alpha -Proteobacteria (ALF; C) and the Roseobacter clade (ROS; D) using UPGMA. The similarity matrix was calculated using Pearson correlation.

Fig. 4. Phylogenetic trees of Proteobacteria (A) and the Bacteroidetes phylum (B) calculated with Maximum-Likelihood based on 16S rRNA gene fragments. Sequences obtained in this study are highlighted in bold.

41 Kapitel II Effects of a phytoplankton bloom on bacterial communities

20 35 A 15 °C 10 30 ‰

5 Temperature Salinity 25 8 12 April 26 April 10 May 24 May 07 June 21 June 8 C total 60 POC B 6 -1 6 % C phyt./C org. 40 -1

mg l 4 %% mg l 4 20 2

0 0 8 12 April 26 April 10 May 24 May 07 JuneCryptophyceae 21 June sp. Dinophyceae

C 3 Pennate diatoms 6 Thalassionema nitzschioides x10

-1 Raphoneis amphiceros Pseudonitzschia pungens 4 Plagiogrammopsis vanheurckii Cylindrotheca closterium Centric diatoms

Cell Counts l 2 Thalassiosira punctigera Guinardia delicatula

0

6 5 attached free-living D x10 -1 4

2 Cell Counts ml-1*106

Cell Counts ml 0 12 April 26 April 10 May 24 May 07 June 21 June DateDate

Fig. 1. Rink et al.

42 Kapitel II Effects of a phytoplankton bloom on bacterial communities

April May June April May June

Std. 12 26 03 10 17 24 31 14 29 12 26 03 10 17 24 31 14 29 Std. EUB

A CFB

B ALF

C ROS

D FL PA

Fig. 2. Rink et al.

43 Kapitel II Effects of a phytoplankton bloom on bacterial communities

A

EUB

B

CFB

C

ALF

D

ROS

Fig. 3. Rink et al.

44 Kapitel II Effects of a phytoplankton bloom on bacterial communities

A

WAC I

RCA

alpha

gamma

beta

Fig. 4A. Rink et al.

45 Kapitel II Effects of a phytoplankton bloom on bacterial communities

B

Fig. 4B. Rink et al.

46

III.

Tidal effects on coastal bacterioplankton

47 Kapitel III Tidal effects on coastal bacterioplankton

Tidal effects on coastal bacterioplankton

Beate Rink, Torben Martens, Doreen Fischer, Andreas Lemke, Hans-Peter Grossart, 1

Meinhard Simon, and Thorsten Brinkhoff *

Institute for Chemistry and Biology of the Marine Environment (ICBM),

University of Oldenburg, P.O. Box 2503, D-26111 Oldenburg, Germany

1 Present address: Institute of Freshwater Ecology and Inland Fisheries, Department of

Limnology of Stratified Lakes, Alte Fischerhuette 2, D-16775 Stechlin, Germany

Running head: Tidal effects on coastal bacterioplankton

Key words: Aggregates, bacteria, CARD-FISH, DGGE, tidal flat, Wadden Sea

______

* To whom correspondence should be addressed ([email protected]).

48 Kapitel III Tidal effects on coastal bacterioplankton

Acknowledgements We appreciate the hospitality and assistance of the RV Senckenberg crew. We gratefully acknowledge K. Ishii and R. Amann for the introduction of the CARD-FISH method and M. Mußmann for valuable methodical discussions. We thank B. Kuerzel and R. Weinert for Chl a and dry weight analyses and R. Reuter and T. Badewien for the supply of temperature and salinity data for July 2005. This work was supported by the Deutsche Forschungsgemeinschaft (DFG) within the research group “BioGeoChemistry of Tidal Flats” (FG 432-5).

Abstract Tidal flats are highly dynamic and productive ecosystems, strongly influenced by hydrodynamic forces and tidal events. We examined the impact of the tide on the composition of free-living (FL) and particle-attached (PA) bacterioplankton in the Wadden Sea, southern North Sea, in November 1999, May 2000 and July 2005 complementing the study of Grossart et al. (2004) which focused mainly on the dynamics of suspended particulate matter (SPM) and bacterial abundance. FISH, CARD-FISH and DGGE fingerprints with primer sets for Bacteria , Bacteroidetes , alpha-Proteobacteria and the Roseobacter Clade based on 16S rRNA genes and 16S rRNA were applied. In addition, suspended particulate matter (SPM), particulate organic carbon (POC), chlorophyll a (Chl a), bacterial cell counts and bacterial protein production rates (BPP) were measured. The introduction of North Sea water in the Wadden Sea system was shown by changes of the water temperature, salinity values and oxygen saturation during high tides (HT). In spring and summer, particulate organic carbon (POC) ratio and cell specific bacterial production was increased and altered with the tide. Surprisingly, these strong variations had only slight effect on the composition of the bacterial communities. FISH counts revealed abundance of Bacteroidetes , alpha- and gamma- Proteobacteria . On particles, also beta-Proteobacteria were detected with 15.29% DAPI in May 2000 and 8.19% DAPI in July 2005. The Roseobacter Clade constituted almost all FL but only one third of the PA alpha-Proteobacteria . In spring and summer, FISH counts showed higher variations within the bacterial groups compared to autumn and increased abundances of PA alpha-Proteobacteria . A clear relationship between these variations and the tide was not found. DGGE banding patterns of 16S rRNA gene fragments were highly stable even on the group-specific level. At higher resolution based on cDNA amplicons, higher richness was detected but only few variations related to the tide appeared. Phylogenetic analysis revealed that prominent bands were affiliated to the RCA, the WAC I and the

49 Kapitel III Tidal effects on coastal bacterioplankton

SAMMIC cluster. These clusters were found in previous studies of the Wadden Sea and are distributed worldwide in temperate and polar marine regions. Our results suggest that the stability of the abundant bacterial groups is a consequence of the strong changes within the Wadden Sea ecosystem selecting highly adapted species which persist even on a long-term scale.

Introduction Tidal flats are coastal ecosystems appearing in temperate and tropical regions and are composed of sandy or muddy sediment. They are strongly exposed to the tide and thus form a very stressful environment for the indwelling organisms. Furthermore, tidal flats provide complex substrates due to high loads of suspended particulate matter (SPM) deriving from the land and the sea, and belong to the most productive ecosystems in the world (Dittmann, 1999). The German Wadden Sea, a tidal flat system in the southern part of the North Sea, is the largest tidal flat system worldwide comprising an area of 7500 km 2. Several monitoring studies examined the status of this ecosystem on spatial, diurnal and seasonal scales. It is a well studied area in terms of suspended particulate matter flux and particle size distribution (Eisma and Li 1993; van Leussen 1996; Behrends und Liebezeit 1999; Mikkelsen 1998; Fugate and Friedrichs 2003; Grossart et al. 2004; Lunau et al. 2006), primary production (van Duyl et al. 1999; Tillmann et al. 2000; Niesel and Günther 1999) and several planktonic and benthic organisms (Günther, 1999; Dittmann, 1999), but the main mediators for organic matter decomposition, the bacterial communities, remained disregarded. Correlation of SPM composition and -variation and microbial activities in the Wadden Sea were investigated by Grossart et al. (2004) during two tidal cycles in November 1999 and May 2000. Strong changes in several parameters were detected due to differences in tidal currents and subsequent introduction of North Sea water masses into the Wadden Sea ecosystem at high tides (HT). In November 1999, water temperatures and salinity were higher during HT. SPM and PA bacteria were correlated and fluctuated during the tidal cycle dependent from current velocities. In May 2000, water temperatures were higher but oxygen saturation was increased during HT due to the North Sea water influence. Cell specific bacterial production increased with rising tide indicating response of the bacterial community to tidal changes. Also seasonal influences were detected particularly for the POC ratio and FL cell counts which were increased in May 2000. These observations substantiated the assumption that the tide may strongly affect the composition of pelagic bacterial communities. So far, only few studies focused on short-term influences on bacterioplankton.

50 Kapitel III Tidal effects on coastal bacterioplankton

Pernthaler and Pernthaler (2005) described variations in the proliferation of North Sea bacteria during a tidal cycle. Diurnal changes in pigment concentrations derived from bacteria in Baltic Sea water were examined by Koblizek et al. (2005). In dilution cultures of North Sea bacteria, the impact of UV on bacterial communities during a 24 hours period was investigated using DGGE profiles (Winter et al., 2001). In previous studies, the main interest of Wadden Sea investigations focused on the composition and abundance of sediment bacteria. Llobet-Brossa et al. (1998) showed dominance of Bacteroidetes and sulfate-reducing bacteria in the sediment surface layer. At two different sampling sites of Wadden Sea sediment, Mussmann et al. (2005) isolated several delta-Proteobacteria . In the water column, the Wadden Sea bacterioplankton was recently examined on a qualitative level by Stevens et al. (2005a, b) and Rink et al. (2006) using DGGE. It was demonstrated that many bands persisted over long-term periods, but also variations in the banding patterns were detected in spring and summer months. Rink et al. (2006) showed correlation of phytoplankton composition changes and alteration in the bacterial community composition using group-specific primer sets. As these studies were performed on a monthly or weekly scale, the influences of tidal currents and the fast exchange of different water bodies in the Wadden Sea on the bacterioplankton remained unknown. Thus, the aim of this study was to examine the impact of the tide on the composition and abundance of the bacterial communities in the Wadden Sea with respect to seasonal aspects. With this study, we complete the results by Grossart et al. (2004) with detailed bacterial community composition analysis based on the same sampling campaigns and an additional sampling in July 2005.

51 Kapitel III Tidal effects on coastal bacterioplankton

Materials and methods

Sample collection and processing– Samples were taken in November 1999, May 2000 and July 2005 from shipboard in the major channel of the backbarrier tidal flat system of the German Wadden Sea (53° 44.9´N, 07° 40.0´E). The sampling period was 19 hours in November 1999, 22 hours in May 2000 and 12 hours in July 2005. In November 1999 and May 2000, surface water was collected every hour and filtrated for suspended particulate matter (SPM), chlorophyll a (Chl a), particulate organic carbon (POC) and bacterial cell counts according to Grossart et al. (2004). In July 2005, samples for SPM, Chl a, POC and bacterial cell counts were taken and processed according to Lunau et al. (2006). Hydrographic data of July 2005 were measured by a CTD probe (SeaCat 19plus, Seabird, Washington, USA). For DGGE analyses, one hundred ml of seawater were filtered at high tides, mean tides and low tides, fractionated onto polycarbonate-filters (diameter 47 mm, Nuclepore) and stored at -20°C in the dark. A pore size of 5.0 µm was used to obtain particle-attached (PA) and 0.2 µm to obtain free-living (FL) bacteria. For Fluorescent In Situ Hybridization (FISH) and Catalyzed Reporter Deposition (CARD)- FISH analysis, 2 - 4 ml were filtered onto 5.0 µm polycarbonate-filters (diameter 25 mm, Nuclepore) and 2 ml of the filtrate were subsequently filtered onto 0.2 µm polycarbonate-filters. After fixation with paraformaldehyde (4% w/v) for one hour filters were stored at -20°C in the dark until further processing. Sediment cores in November 1999 and May 2000 were taken with Plexiglas tubes (36 mm diameter) at low tide on an intertidal mud flat about 200 m away from the ship. Two mm of the upper surface layer were sliced, transferred into sterile caps and kept frozen at -20°C.

SPM dry weight, particulate organic carbon and chlorophyll a – Filters for SPM dry weight were dried for 12 h at 60°C, adapted to room temperature and weighed on a microbalance. Dry weight was calculated as the difference of filter weight before and after filtration (500 ml filtration volume, glass fiber filters, GF/F, Whatman). For particulate organic carbon, filters were exposed to hydrochloric acid fume for 12 h and subsequently analyzed by a FlashEA 1112 CHN-analyzer (Thermo Finnigan). Chlorophyll a was determined photometrically after extraction in hot ethanol according to Von Tuempling and Friedrich (1999). For further details concerning SPM, POC and Chl a analyses see Lunau et al. (2006).

52 Kapitel III Tidal effects on coastal bacterioplankton

Bacterial cell counts – Bacterial abundance was determined with SybrGreen I as described by Lunau et al. (2006) with slight modifications. The detachment of particulate bacteria was performed using methanol (30% v/v) and ultrasonic treatment for 15 min at 35°C. For the enumeration of bacterial cells, subsamples of 1 ml were filtered onto black polycarbonate filters (0.2 µm pore size, Poretics). Abundance of particulate bacteria was calculated as the difference between total and free-living cell counts.

Bacterial Production –The bacterial production rate was estimated by the incorporation of 14 C-Leucin (Simon and Azam 1989). Samples were incubated in triplicates with 14 C-Leucin (306 mCi/mmol, Hartmann Analytic, Germany) at a final concentration of 70 nmol l -1 to ensure saturation of uptake systems. Formalin fixed water samples (2% v/v) were used as controls. The samples were incubated in 5 ml plastic tubes in the dark at in situ temperature for up to 1 hour. To avoid sedimentation, incubation was performed on a plankton wheel. The incubation was linear for at least 1 hour and terminated by the addition of formaldehyde (2% v/v). After fixation, the samples were filtered onto 0.45 µm nitrocellulose filters (Sartorius, Germany) and extracted with ice-cold 5% trichloracetic acid (TCA) for 5 min. Subsequently, filters were rinsed twice with 3 ml ice-cold TCA (5% v/v). To dissolve the filters, 4.5 ml of scintillations cocktail was added. Vials were shaken vigorously and radioactivity was determined afterwards in a scintillation counter. Standard deviation of triplicate measurement was usually <15%.

FISH and CARD-FISH – FISH filters in November 1999 and May 2000 were rinsed with 1 ml phosphate buffered saline (PBS 1x) and dehydrated with 50%, 80% and 100% ethanol. Subsequently, bacterial cells were hybridized for 5 h and washed 20 min at 46°C following the protocol of Glöckner et al. (1996). Oligonucleotide probes for various phylogenetic groups were used in November 1999: EUB 338 (Amann et al. 1990), ALF968, GAM42a (Manz et al. 1992), CF319a (Cytophaga/Flavobacteria, Manz et al. 1996), ARCH915 (Stahl & Amann 1991) and SRB385 (Amann et al. 1990). In May 2000, the additional probes BET42a (Manz et al. 1992) and SRB385db (Rabus et al. 1996) were used. All filters were counterstained with DAPI (4´,6´-Diamidino-2-phenylindol, 1µg/ml). In July 2005, hybridizations were performed using the CARD-FISH method following the protocol of Sekar et al. (2003). After fixation, samples were embedded in agarose (0.2%) and treated with lysozyme (10mg/ml). Hybridization conditions were 3 h of hybridization at 35°C, 10 min washing at 37°C and 30 min. amplification at 37°C. The following oligonucleotide probes

53 Kapitel III Tidal effects on coastal bacterioplankton labeled with horseradish-peroxidase (HRP) were used for CARD-FISH: EUB338, ALF968, BET42a, GAM42a, ROS536 (Brinkmeyer et al. 2000) and NON338 (Wallner et al. 1993). Tyramine-HCl was labeled with Fluorescein-5-isothiocyanate (FITC) as described by Pernthaler et al. (2002). To avoid unspecific accumulation of dye in the cells, the last washing step in PBS (1x) amended with TritonX-100 (0.05%) was extended to 30 min. Counterstaining was performed with Vectashield-mounting medium with DAPI (1.5µg/ml; Vector Laboratories, Peterborough, England).

Nucleic acid extraction – For samples taken in November 1999 and May 2000, genomic DNA was isolated by phenol-chloroform extraction after bead beating as described by Selje and Simon (2003). In July 2005, DNA and RNA were co-extracted from the same filter using phenol-chloroform calibrated with sodium acetate RNA-buffer (50 mM, pH 4.2) containing EDTA (10 mM) and Polyvinylpolypyrrolidone (PVPP, 2 g l -1). All steps were done under sterile conditions using Diethylpyrocarbonate (DEPC, 0,1%)-treated ingredients. After precipitation in isopropanol at -20°C overnight nucleic acids were resuspended in DNase/RNase-free molecular grade water (Eppendorf, Germany). The RNA was incubated with DNase I (5 U ml -1) for 1 h at 37°C and precipitated as described above. DNase digestion was repeated until no DNA contamination could be detected by PCR with primers specific for bacterial 16S rRNA genes. Samples were stored at -20°C until further processing.

PCR amplification – 16S rRNA gene fragments for subsequent DGGE analyses were amplified using the primer pair GC 341F and 907RM (Muyzer et al. 1998) in an Eppendorf Mastercycler (Eppendorf, Hamburg, Germany). In July 2005, the additional primer pairs GC CF319aF and 907RM, GC 341F and ALF968R, and GC ROS 536F and GRB 735R were used to investigate the bacterial community composition on a group-specific level. Primer sequences and PCR conditions were described earlier by Rink et al. (2006). Four µl of the amplification products were analyzed by electrophoresis in 2% (w/v) agarose gels and stained with ethidium bromide (1 µg ml -1). For sequencing analysis, PCR products were purified by using the E.Z.N.A. Microspin Cycle-Pure Kit (Peqlab Biotechnologie GmbH, Erlangen, Germany) following the instruction manual. To amplify bacterial RNA, the Qiagen RT-PCR kit (Qiagen, Hilden, Germany) was used following the instruction manual for reverse transcription of the RNA. Reverse transcription was done directly with bacteria- or group- specific primer sets as described above. Subsequent PCR was performed in the same reagent mix directly after transcription using specific conditions for the applied primer set.

54 Kapitel III Tidal effects on coastal bacterioplankton

DGGE analysis of PCR products – DGGE was performed with the INGENY phorU System (INGENY International BV, Goes, Netherlands) using specific conditions for the applied primer sets according to Rink et al. (2006). Gels were stained with SYBR Gold (Molecular Probes, Inc.) after electrophoresis and documented digitally using a BioDoc Analyze Transilluminator (Biometra, Göttingen, Germany). Bands were excised with a sterile scalpel, suspended in 50 µl of molecular grade water (Eppendorf, Germany) and stored at –20°C until reamplification.

Cloning– DGGE bands GWS-TC-a2-PA, GWS-TC-a6-FL, GWS-TC-c4-PA, GWS-TC-e9- FL, GWS-TC-e11-FL, GWS-TC-e3-SE, GWS-TC-e4-PA, GWS-TC-e1-SE and GWS-TC-e2- SE were cloned using the pGEM®-T Vector System II (Promega, Madison, USA) following the instruction manual. Clones with inserts were picked, resuspended in molecular grade water (Eppendorf, Germany) and screened by DGGE to check if the insert position matches the position of the corresponding DGGE band. Adequate clones were amplified and subsequently sequenced using the primers pUC/M13f and pUC/M13r (Messing 1983).

Sequencing and phylogenetic analysis – PCR products were sequenced using the DYEnamic Direct cycle sequencing kit (Amersham Life Science, Inc.) and a Model 4200 Automated DNA Sequencer (LI-COR, Inc.) using GM8R (5´-TGGGTATCTAATCCT-3´) as sequencing primer labeled with IRDye TM 800. In addition, DGGE bands were sequenced by GeneArt (Regensburg, Germany) using the primer 907RM to enhance the sequence quality by repeat determination. For all sequences, at least 400 bp were determined. Construction of the phylogenetic trees was performed using the ARB software package (http:/www.arb-home.de). Calculation of the backbone trees was done with the maximum likelihood method using sequences of type strains of the selected phylogenetic groups with a minimum sequence length of 1300 bp. To avoid uncertain alignments, positions at which less than 50% of all sequences showed the same residues were excluded. Sequences with less than 1300 bp were added to the backbone tree using the maximum parsimony method and the same filter. Five type strains belonging to Cyanobacteria were used as outgroup.

Nucleotide sequence accession number –The sequences obtained in this study are available from GenBank under accession no. DQ911822 to DQ911842.

55 Kapitel III Tidal effects on coastal bacterioplankton

Statistics – Cluster analyses of DGGE banding patterns were performed using Gel Compar II, version 2.5 (Applied maths, Kortrijk, Belgium). Calculations were done curve based using Pearson correlation and UPGMA.

Results

Hydrographic data–In July 2005 water temperature ranged between 19.1 and 20.1°C showing slight decrease at HT with incoming North Sea water. Salinity was almost constant around 32 psu (Fig. 1a). For hydrographic data as well as results concerning SPM dry weight, POC, Chl a, bacterial abundance and bacterial production from November 1999 and May 2000 see Grossart et al. (2004).

SPM dry weight, POC and Chl a–Dry weight of suspended particulate matter in July 2005 ranged from 7.5 to 8.2 mg l -1 at HT and reached the highest value of 36.8 mg l -1 at low tide (Fig. 1b). Despite highest current values at mean tide, SPM dry weight was average (21 to 29.2 mg l -1). POC values (see supplementary data) were low around LT (0.39 mg l -1) and increased towards HT (up to 2.15 mg l -1) and constituted 1.1% (LT) to 28.7% (HT) of the SPM dry weight. The average value was 7.6% dry weight. Chlorophyll a increased from 3.13 µg l -1 at HT to 6.7 µg l -1 around LT and decreased again to 3.76 µg l -1 at high tide (Fig. 1c). Phaeopigments showed the same tidal dynamic and ranged from 0.30 to 5.12 µg l -1 with the highest ratio phaeopigments/chlorophyll a of 1.18 at mean tide.

Bacterial abundance and bacterial production – In July 2005, total bacterial cell counts varied between 1.28 to 3.56 x 10 6 cells ml -1 (Fig 1d). Free-living (FL) bacteria showed a minimum of 0.97 x 10 6 cells ml -1 one hour after the first high tide and increased slightly to a maximum of 2.30 x 10 6 cells ml -1 at one hour after mean tide 1 (MT 1). Afterwards, FL cell numbers decreased again until next mean tide (MT2) and then ranged between 1.46 and 1.75 x 10 6 cells ml -1 until second high tide. Particle-attached (PA) bacteria were lowest around high tide with 0.09 to 0.45 x 10 6 cells ml -1 and highest between MT1 to MT2 (0.96 to 1.33 x 10 6 cells ml -1) with a slight decrease around LT. In average, PA bacteria accounted for 28% of total bacterial cell counts. To compare cell densities of PA bacteria, cell counts per mg dry weight were calculated (see supplementary data). Density of PA cells was highest one hour after MT and two hours after LT due to lower dry weight values and equal cell counts.

56 Kapitel III Tidal effects on coastal bacterioplankton

Bacterial production (BPP, Fig. 1e) in July 2005 ranged between 0.55 µg l -1 h -1 at both high tides to 3.33 to 3.64 µg l -1 h -1 during LT and MT2.

FISH counts – FISH counts detected with probe EUB338 were lowest in May 2000 with a mean value of 56.77% DAPI on particles and 48.42% DAPI in the FL fraction (Table 1, Fig. 2). Highest FISH counts were measured in November 1999 with 73.8% DAPI on particles and 61.24% DAPI in the FL fraction. In July 2005, FISH counts were similar to November with 70.42% DAPI on particles and 59.44% DAPI for free-living bacteria. Highest variation within the cell counts were detected in May 2000 during a spring phytoplankton bloom, while lowest variation was observed in November 1999 when hydrodynamic forcing controlled the processes in the water column (Fig. 2). Generally, a higher percentage (% DAPI) of cells could be detected on particles compared to the free-living fraction. Specific probes revealed high numbers of Bacteroidetes , alpha-, beta- and gamma- Proteobacteria in the Wadden Sea (Table 1, Fig. 2). In November 1999 and May 2000, also low cell counts of Archaea and sulphate-reducing bacteria were detected (below 5% each). In May 2000, cell counts of Bacteroidetes were 27.59% DAPI on particles and 13.55% DAPI in the FL fraction. In November 1999, Bacteroidetes showed lower FISH counts on particles 20.93% DAPI, but higher numbers (19.16% DAPI) in the FL fraction (mean values). In July 2005, lowest cell counts of Bacteroidetes were detected with 15.02% DAPI in the PA fraction and 11.52% DAPI in the FL fraction. Abundances of alpha-Proteobacteria on particles were lowest in November 1999 with 9.53% DAPI and highest in May 2000 with 30.59% DAPI. In the free-living fraction, alpha- Proteobacteria contributed 13.76% of the total cell counts in November 1999 but only 7.21 to 7.87% DAPI in spring and summer. In July 2005, the Roseobacter Clade, as part of the alpha- Proteobacteria , represented 4.34% DAPI on particles and 5.09% DAPI in the FL fraction contributing nearly 25% of the PA and 71% of the FL alpha-Proteobacteria . Abundances of gamma-Proteobacteria ranged between 16.49 and 28.83% DAPI on particles and between 9.83 and 17.12% DAPI in the FL fraction. Hybridization with probe BET42a revealed 8.19 to 15.29% DAPI counts on particles in spring and summer showing a certain abundance of PA beta-Proteobacteria in this coastal environment.

DGGE banding patterns and Cluster analysis –DGGE analysis of the tidal cycles in November 1999, May 2000, and July 2005 revealed distinct bacterial populations for the sediment surface, the PA and the FL bacterial communities (Fig. 3, 4, 5). In November 1999

57 Kapitel III Tidal effects on coastal bacterioplankton and May 2000, fingerprints within the different fractions showed almost identical banding patterns during the tidal cycles (data not shown). Highest band numbers in all seasons were determined in the PA fraction (24 bands in Nov 1999, 13 bands in May 2000, 25 bands in July 2005) and a direct comparison of November 1999 and May 2000 showed only few variations in the banding patterns at all (Fig. 3). The banding patterns of the PA bacterial communities indicate an overlap with those resulted from sediment and the FL bacteria, as both fractions show bands on the same height as in the PA fraction. In July 2005, the FL and PA bacterial communities were also distinct as shown by the DGGE fingerprints and cluster analysis. In addition, high stability of the communities was observed on the DNA level, however, some bands appeared only in one or two samples, i. e. at certain times of the tidal cycle (Fig. 4). Analysis with the bacteria specific primers revealed for the PA fraction, e. g., that samples obtained during the first mean tide (MT1) and LT showed one additional band (GWS-TC-e6-PA), which probably led to higher similarity of these two samples in the cluster analysis (Fig. 5). In the FL fraction, one band appeared exclusively at the first HT (GWS-TC-e8-FL). The cDNA fingerprints of Bacteria revealed a slightly higher richness for the FL fraction with up to 23 bands at HT compared to a maximum of 18 bands in the DNA banding patterns. At HT, two additional bands appeared in the FL fraction which were neither detected in the DNA fingerprints nor in the other cDNA samples (GWS-TC-e10-FL and GWS-TC-e11-FL). On the group-specific level, DGGE analysis of Bacteroidetes showed stable DNA based banding patterns in the PA fraction (Fig. 4, CFB). In the FL fraction, one prominent band appeared exclusively at MT1 and LT (GWS-TC-c1-FL). The cDNA banding patterns showed few differences compared to the DNA fingerprints and during the tidal cycle as well. In the PA fraction, bands no. 2, 3, and 4 (GWS-TC-c2-PA, GWS-TC-c3-PA and GWS-TC-c4-PA) occurred individually at different times and did not appear in the DNA based banding patterns. Cluster analysis of the Bacteroidetes fingerprints revealed distinct fractions of PA und FL fractions except for the second mean tide (MT2) of the cDNA FL fraction, which showed higher similarity to the PA cluster (Fig. 5). The FL cluster was divided in two subclusters of DNA and cDNA samples showing high similarity within DNA samples (>90% Pearson correlation). The PA cluster was also subdivided into DNA and cDNA clusters except the DNA sample at LT which fell into the cDNA group. DGGE analysis of the alpha-Proteobacteria showed low richness within the DNA fingerprints with 14 bands in the PA fraction and 8 bands in the FL fraction (Fig. 4). Slight differences during the tidal cycle were detected on particles at LT with decreased band

58 Kapitel III Tidal effects on coastal bacterioplankton intensity of bands at the standard height, also reflected by cluster analysis. The cDNA banding patterns showed higher richness in both fractions compared to the DNA level. A maximum of 25 bands was counted at LT on particles and 11 bands were visible at HT in the FL fraction. These differences between DNA and cDNA resulted in distinct clusters as shown by cluster analysis for both, PA and FL fraction (Fig. 5). Tidal differences traced back to alpha-Proteobacteria were detected in the cDNA fingerprints during MT1 and LT (Fig. 4, GWS-TC-a2-PA, and GWS-TC-a5-PA) or generated by chloroplast phylotypes (GWS-TC- a3-PA, GWS-TC-a4-PA). For the Roseobacter Clade, DNA based fingerprints revealed a maximum of 14 bands in the PA fraction and 11 bands in the FL fraction. The cDNA banding patterns showed significantly higher richness in the PA fraction (20 bands at LT) compared to DNA fingerprints, forming distinct clusters (Fig. 5). Slight differences during the tidal cycle were also found for this group (Fig. 4). While the cluster analysis demonstrated high similarity of the DNA and cDNA patterns of the FL bacterial communites, the PA samples of the Roseobacter Clade clustered distinct in DNA and cDNA (Fig. 5).

Phylogenetic affiliation – Sequence analysis of DGGE bands revealed that most of the prominent bands were affiliated to recently described clusters of Wadden Sea bacteria (WAC I, Stevens et al. 2005b) or of worldwide distributed bacteria (RCA, Selje et al. 2004; SAMMIC, Stevens et al. 2005b) as shown in Fig 6A. From the DGGE banding patterns of Bacteria , phylotypes belonging to the alpha-, gamma- and beta-Proteobacteria were obtained reflecting the FISH and CARD-FISH results. One phylotype of delta-Proteobacteria and two sequences of Actinobacteria were detected in the sediment fractions of November and May (Fig. 3 and 6, GWS-TC-e1-SE, GWS-TC-e2-SE, GWS-TC-e3-SE). In total, four bands were identified as chloroplasts (Fig. 3: GWS-TC-e5-PA; Fig. 4: GWS-TC-e7-FL, GWS-TC-a3-PA, GWS-TC-a4-PA). GWS-TC-e11-FL was distantly related to Acidocella aminolytica (Fig. 6A) and was not related to any sequence found in the Wadden Sea or North Sea before. Two sequences fell in the WAC I cluster (GWS-e6-FL, GWS-e7-FL), GWS-TC-e9-FL clustered with the Wadden Sea clone GWS-FL-5 (Stevens et al. 2005a), three sequences belonged to the RCA cluster (GWS-TC-e6-PA, GWS-e12-PA, GWS-e5-FL) and GWS-e3-FL was related to clone NAC11-7 (Fig. 6 A). Two phylotypes of DNA and cDNA fingerprints clustered together within the gamma- Proteobacteria (GWS-TC-e8-FL, GWS-TC-e10-FL). They were affiliated to Stenotrophomonas maltophilia and appeared solely during HT in the FL fraction. Two of the

59 Kapitel III Tidal effects on coastal bacterioplankton sediment derived sequences (GWS-TC-e3-SE, delta-Proteobacteria ; GWS-TC-e2-SE, Actinobacteria ) were affiliated to clones which were found in the North Frisian Wadden Sea sediments (Sylt 19, clones Sylt 21, Fig. 6 A). Bacteroidetes affiliated phylotypes were exclusively obtained with the specific primer set (GWS-TC-c1-FL to GWS-TC-c4-PA) confirming the bias against this phylum by the use of the Bacteria -specific primer set (Kirchman 2002). Within the Bacteroidetes phylum, GWS- TC-c1-FL represented the most pronounced diurnal change in the FL fraction of the DNA fingerprints. This phylotype was related to GWS-c14-PA which was detected in May 2000 with specific primer sets for Bacteroidetes (Rink et al. 2006). Both were affiliated to clone CF60 (AY274866) derived from the Delaware estuary. GWS-TC-c2-PA was detected during HT in the PA fraction of the cDNA fingerprints and the phylogenetic affiliation showed no relationship to known Wadden Sea or North Sea organisms. Another cDNA phylotype, GWS- TC-c3-PA, was related to Lutibacter litoralis , isolated from a tidal flat system in Korea (Choi and Cho 2006). GWS-TC-c4-PA was related to two novel species, Krokinobacter genikus (Khan et al. 2006), isolated from marine sediment in Japan, and Dokdonia donghaensis (Yoon et al. 2005), isolated from seawater of the Korean East Sea. The application of the ALF968r primer revealed two chloroplast sequences (GWS-TC-a3- PA, GWS-TC-a4-PA), and three phylotypes of alpha-Proteobacteria . Within the alpha- Proteobacteria , the cDNA derived phylotypes clustered with Silicibacter lacuscaerulensis (GWS-TC-a2-PA) and Sphingomonas paucimobilis (GWS-TC-a5-PA).

Discussion

Grossart et al. (2004) demonstrated that bacterial dynamics in the Wadden Sea are controlled by two major factors: resuspension of sediment and phytoplankton growth. Another strong influence is the introduction of North Sea water in the tidal basin at high tides (HT). In November 1999, sediment resuspension was the dominating process in the water column (Grossart et al. 2004). SPM flux and changes in particulate carbon directly reflected the tidal dynamics. In May 2000, a strong influence of a phytoplankton bloom was observed resulting in higher Chl a concentrations and an increase of the POC ratio and free-living bacterial cells. In July 2005 the Chl a concentrations were lower than during the phytoplankton bloom in May but typical for summer months in the Wadden Sea (Lunau et al., 2006). SPM values in July were much lower compared to November and May, probably due to a lower resuspension rate and lower abundance of phytoplankton, respectively (see

60 Kapitel III Tidal effects on coastal bacterioplankton

Grossart et al. 2004). Bacterial abundance on particles was similar in all seasons following SPM dynamics. Cell counts of free-living bacteria were lower in July than in May but similar to November suggesting lower dissolved organic carbon concentrations in the water column (Grossart et al. 2004). The bacterial production was lowest around HT contrary to May and other investigations (Lunau et al. 2006). Bacterial production values of May showed not a tidal, but a diurnal, pattern, with high rates during the day and reduced values at night (Grossart et al. 2004). Grossart et al. (2004) showed that free-living bacteria were influenced by SPM concentrations, total carbon (TC) and particulate organic carbon in November 1999 and May 2000. The abundance of particulate bacteria correlated with SPM, Chl a, total carbon and particulate organic carbon. Thus, we investigated the composition of the bacterial communities in the Wadden Sea by FISH and DGGE expecting strong changes of the populations following the tidal dynamics of these parameters. By contrast, the fluctuation of particulate carbon, SPM, Chl a and bacterial cell counts was not reflected by changes within the phylogenetic groups during tidal cycles in all investigated seasons as shown by the FISH results. Even the application of the highly sensitive CARD-FISH method in July 2005 did not show any systematic impact of the tide on the bacterial community composition. In all seasons, alpha- and gamma-Proteobacteria as well as Bacteroidetes were most abundant in both fractions. This is in line with other studies from the North Sea (Eilers et al. 2000) and the Weser estuary (Selje et al. 2003). The seasonal comparison of the FISH results revealed very low variation within the bacterial groups during the tidal cycle in November in contrast to May and July (Fig. 2). Largest variations and highest group-specific FISH counts were obtained in May 2000 on particles. The exceeding of the group-specific over the EUB338 counts indicated high activity of the PA bacteria, as the signal strength of directly fluorochrome-labeled oligonucleotide probes depends on the ribosomal RNA content of the cells (Schut 1994). The SPM in May was enriched with phytoplankton derived POC which intensely stimulated bacterial degradation processes of the organic matter as suggested by the decoupling of activity parameters and tidal dynamics (Grossart et al. 2004). This may be one reason for the fact that despite large variations within the bacterial groups, no clear relationship was found between the FISH counts and the tide. In spring and summer, beta-Proteobacteria were also investigated and very high abundances were detected on particles. In May 2000, they constituted even 15.29% of the total PA bacterial community (mean value, Table 1). This is in line with FISH counts of a study showing that this group constituted 6% of DAPI cell counts in the marine section of the

61 Kapitel III Tidal effects on coastal bacterioplankton nearby Weser estuary (Selje et al. 2003) and findings of several other authors (Rappé et al. 2000; Beja et al. 2002). High abundance of beta- Proteobacteria in this habitat was supported by detection of phylotype GWS-e8-FL in May 2000, which clustered with other phylotypes derived from the Wadden Sea (GWS-e4-FL, Rink et al. 2006; GWS-FL-6, Stevens et al. 2005) and the Weser estuary (DC11-51-11, Selje et al. 2005). The latter phylotype was obtained from a 10 -5 dilution step on marine medium even suggesting abundance of beta- Proteobacteria in this saline environment. DGGE results demonstrated that the composition of the bacterial communities in November, May and July showed almost no changes during the tidal cycles. Sequencing of prominent bands revealed that most of these phylotypes were affiliated to previously described phylogenetic clusters, e. g. the RCA and the WAC I cluster within the alpha- Proteobacteria (Selje et al. 2004; Stevens et al. 2005b) and the SAMMIC cluster within the gamma-Proteobacteria (Stevens et al. 2005b). Organisms affiliated with the RCA cluster are globally distributed in temperate and polar regions. They live exclusively in marine environments freely suspended in the water column. The SAMMIC cluster comprises organisms living on surfaces in marine environments. Bacteria of the SAMMIC cluster were permanently detected on particles and on the sediment surface in the Wadden Sea during a seasonal study (Stevens et al. 2005a). In parallel, Bowman et al. (2005) described the same cluster with phylotypes from polar and temperate marine sediments detecting abundances of up to 4% of the total bacterial community by quantitative RealTime-PCR. The distribution and abundance of WAC I affiliated organisms is not yet clarified, but phylotypes belonging to this cluster were also present in the Wadden Sea during all seasons (Stevens et al. 2005a). Thus, bacteria belonging to these phylogenetic groups are not affected by short-term changes generated by the tide as confirmed by our results. In contrast to these permanently present organisms, some phylotypes were not detected during the whole tidal cycles but in single samples (Fig. 4). The application of group-specific primer sets revealed additional bands present at specific points of time. Although the use of specific primer sets enhances the resolution of microbial studies (Abell & Bowman 2005; Gich et al. 2005; Rink et al. 2006) the bacterial communities were remarkably stable even on the specific level indicating low influence of the tide on the investigated groups. Most tidal changes were detected by the RNA approach showing significantly higher richness compared to the DNA fingerprints. The rRNA content of metabolically active bacteria is higher than in dormant cells (Poulsen et al. 1993) and correlates with bacterial growth rates (Delong et al. 1989). Thus, rRNA is suggested as activity indicator and reflects

62 Kapitel III Tidal effects on coastal bacterioplankton responses of bacterial communities to environmental changes more pronounced than the rRNA gene. Higher sensitivity concerning species diversity and appearance of community shifts have been reported before, using rRNA based fingerprinting methods for marine bacterioplankton (Schäfer et al. 2001; Moeseneder et al. 1999), drinking water supply systems (Eichler et al. 2006) and bacterial assemblages of mariculture biofilter systems (Cytryn et al. 2005). Our study provides the first RNA based insights into bacterial communities of the Wadden Sea tidal flat system, showing that most of the cDNA derived phylotypes, which occured exclusively at specific points of the sampling, were not closely related to previously described Wadden Sea organisms. Within the alpha-Proteobacteria , two of these phylotypes were affiliated to described species (GWS-TC-a5-PA and GWS-TC-a2-PA, Fig. 6A). Phylotype GWS-TC-a6-FL, which appeared at rising tide (MT2) was related to clone ZD0117 derived from the North Sea during a phytoplankton bloom (Zubkov et al. 2002). Largest differences between DNA and cDNA fingerprints were reflected by the Roseobacter specific fingerprints in the PA fraction. The richness within the cDNA banding patterns showed almost two-fold increase compared to DNA. These findings were reflected by distinct clusters of DNA and cDNA as revealed by cluster analysis. Hence, our results suggest that significantly more particle-attached organisms are active than previously indicated by DNA based methods and that the additional application of RNA based fingerprinting methods on a group-specific level is essential to detect small-scale processes within microbial communities. Relationships between additionally appearing bands and tidally generated processes were hardly detected. In July 2005, bands GWS-TC-e8-FL and GWS-TC-e10-FL were pronounced at HT in the FL fraction of DNA and cDNA samples. At HT, increased salinity, %POC values and lower water temperature indicated influence of water from the open North Sea. The two above mentioned phylotypes were closely related to Stenotrophomonas maltophilia (Fig. 6A, 96-97% sequence similarity). The genus Stenotrophomonas is widespread in terrestrial and limnic habitats, but was also found in different marine samples, several times associated with algae, sponges or dinoflagellates (e. g. Hagstrom et al. 2000; Seibold et al. 2001; Sfanos et al. 2005). Seibold et al. (2001) found a phylotype of Stenotrophomonas associated with the dinoflagellate Noctiluca scintillans obtained from plankton hauls taken at Helgoland Roads (German Bight, North Sea). Thus it is possible that the additional DGGE bands we observed in our study derived from a Stenotrophomonas sp. introduced to the Wadden Sea with the incoming tide. The same probably holds true for band GWS-TC-e11-FL (Fig. 4), which belongs to the SAR116 cluster (Giovannoni et al. 1990) and band GWS-TC-a6-FL, which appeared at rising tide (MT2) and was related to a clone obtained from the open North Sea

63 Kapitel III Tidal effects on coastal bacterioplankton during a phytoplankton bloom (Zubkov et al. 2002). Appearance of these phylotypes only at HT or rising tide, however, indicates low significance of the organisms within the Wadden Sea ecosystem. Cunha et al. (2001) showed that the activity of marine bacteria exposed to brackish water increased significantly while bacteria of the estuary were less active when exposed to marine water. Thus, the detection of differences at HT or rising tide may also be a single response of marine bacteria to high concentrations of suspended substrates in the Wadden Sea. Band GWS-TC-e6-PA (RCA cluster) disappeared in July 2005 at HT and MT2 (rising tide) in parallel to the incoming North Sea water, but only in the particle fraction. This effect was also detected in May 2000, but not in November 1999 (data not shown). In spring and summer, SPM concentrations were lower compared to November with inversed tidal dynamics (Grossart et al., 2004; this study). In May and July the inflow of North Sea water resulted in lower amounts of SPM at HT suggesting that the disappearance of this band may be due to a dilution effect. This assumption is supported by data showing much lower numbers of RCA related phylotypes in the open North Sea compared to the Wadden Sea (H. Giebel, University of Oldenburg, unpublished results). As no corresponding bands were visible in the cDNA pattern of the particle fraction (Fig. 4), the significance of this organism for SPM degradation in the Wadden Sea is also questionable. On the group-specific level, GWS-TC-c1-FL appeared at MT1 and LT in the FL fraction of the Bacteroidetes in the DNA and cDNA banding patterns. Closest relative was clone GWS-c14-PA found in the Wadden Sea during a phytoplankton bloom in May 2000 (Rink et al. 2006). This indicates that the organism is resident and active in the Wadden Sea. The ecological function, however, remains unknown as this organism was found on particles (Rink et al. 2006) and free-living (this study) as well. Several other bands appeared on the group specific level at MT1 and LT (Fig. 4: GWS-TC- c3-PA, GWS-TC-c4-PA, GWS-TC-a2-PA, GWS-TC-a5-PA) what might be explained by resuspension of material from the sediment surface but this remains speculation. The fingerprints of the Roseobacter Clade, which constituted a high fraction within the alpha-Proteobacteria (approx. 25% on particles and 71% in the FL fraction as shown by CARD-FISH results), were most stable even on the RNA level. Despite significantly higher richness on particles (see above), almost no changes in the community composition were observed for this group during the tidal cycle. Thus, our results indicate that the combined application of group-specific primer sets and DNA/RNA based fingerprinting is sufficient for the reliable examination of bacterial communities, i. e. to detect small changes within the

64 Kapitel III Tidal effects on coastal bacterioplankton bacterial communities. In the tidal flat ecosystem of the Wadden Sea, however, only very few changes could be observed and thus connection of these data with processes generated by the tide were only sparely visible. Overall our results demonstrate that even application of very sensitive investigation methods like CARD-FISH and rRNA based DGGE analyses with specific primer sets revealed highly stable FL and PA bacterial communities, almost not influenced by strong tidal effects of this dynamic ecosystem. Seasonal changes and phytoplankton blooms may result in larger tidal variations of group-specific abundances and the appearance of additional organisms on a short time-scale but do not seem to influence the composition of the bacterial communities significantly (Stevens et al. 2005a; Rink et al. 2006). Thus, we conclude that the recurrent short- and long-term changes in the Wadden Sea resulted in the selection of highly adapted organisms and lead to an exceptional stability of the bacterial communities in this ecosystem. Further investigations are now required to clarify how the abundant bacteria are involved in the microbial degradation processes and what makes them superior to other organisms.

65 Kapitel III Tidal effects on coastal bacterioplankton

References

ABELL , G. C., AND J. P. BOWMAN . 2005. Ecological and biogeographic relationships of class Flavobacteria in the Southern ocean. FEMS Microbiol. Ecol. 51 : 265-77. AMANN , R. I., B. J. BINDER , R. J. OLSON , S. W. CHISHOLM , R. DEVEREUX , AND D. A. STAHL . 1990. Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 56 : 1919-1925. BEHRENDS , B. AND G. LIEBEZEIT . 1999. Particulate amino acids in Wadden Sea waters – seasonal and tidal variations. J. Sea. Res. 41 : 141-148. BEJA , O., AND OTHERS . 2002. Unsuspected diversity among marine aerobic anoxygenic phototrophs. Nature 415 : 630-633 BOWMAN , J. P., S. A. MCCAMMON , AND A. L. DANN . 2005. Biogeographic and quantitative analyses of abundant uncultivated γ-proteobacterial clades from marine sediment. Microb. Ecol. 49 :451-460. BRINKMEYER , R., M. RAPPE , S. GALLACHER , AND L. MEDLIN . 2000. Development of clade- (Roseobacter and Alteromonas ) and taxon- specific oligonucleotide probes to study interactions between toxic dinoflagellates and their associated bacteria. Eur. J. Phycol. 35: 315-329. CHOI , D.H, AND B.C. CHO . 2006. Lutibacter litoralis gen. nov., sp. nov., a marine bacterium of the family Flavobacteriaceae isolated from tidal flat sediment. Int. J. Syst. Evol. Microbiol. 56 : 771-776. CORZO -DELGADO , J.E., AND J.M.G OMEZ -MATEOS . 2006. Stenotrophomonas maltophilia , an increasingly important nosocomial pathogen. Enferm. Infecc. Microbiol. Clin. 24 : 1-3. CUNHA , M. A., M. A. ALMEIDA , AND F. ALCANTARA . 2001. Short-term responses of natural planktonic bacterial community to the changing water properties in an estuarine environment: Ectoenzymatic activity, glucose incorporation, and biomass production. Microb Ecol 42 :69-79. CYTRYN , E., J. VAN RIJN , A. SCHRAMM , A. GIESEKE , AND D. MINZ . 2005. Identification of bacteria potentially responsible for oxic and anoxic sulfide oxidation in biofilters of a recirculating mariculture system. Appl. Environ. Microbiol. 71 :6134-41. DELONG , E.F., G.S. WICKHAM , AND N.R. PACE . 1989. Phylogenetic strains: ribosomal RNA- based probes for the identification of single cells. Science 243 :1360-1363. DITTMAN , S. 1999. The Wadden Sea ecosystem: Stability properties and mechanisms. New York: Springer-Verlag. EICHLER , S., R. CHRISTEN , C. HOLTJE , P. WESTPHAL , J. BOTEL , I. BRETTAR , A. MEHLING , AND M.G. HÖFLE . 2006. Composition and dynamics of a drinking water supply system as assessed by RNA- and DNA-based 16S rRNA gene fingerpinting. Appl. Environ. Microbiol. 72 :1858-1872. EILERS , H., J. PERNTHALER , F. O. GLÖCKNER , AND R. AMANN . 2000. Culturability and in situ abundance of pelagic bacteria from the North Sea. Appl. Environ. Microbiol. 66 : 3044- 3051. EISMA , D., AND A. LI. 1993. Changes in suspended-matter floc size during the tidal cycle in the Dollard Estuary. Neth. J. Sea Res. 31 : 107-117. FUGATE , D.C., AND C.T. FRIEDRICHS . 2003. Controls on suspended aggregate size in partially mixed estuaries. Estuar. Coast. Shelf Sci. 58 : 389-404.

66 Kapitel III Tidal effects on coastal bacterioplankton

GICH , F., K. SCHUBERT , A. BRUNS , H. HOFFELNER , AND J. OVERMANN . 2005. Specific detection, isolation, and characterization of selected, previously uncultured members of the freshwater bacterioplankton community. Appl. Environ. Microbiol. 71 : 5908-19. GLÖCKNER , F.O., R. AMANN , A. ALFREIDER , J. PERNTHALER , R. PSENNER , K. TREBESIUS , AND K. H. SCHLEIFER . 1996. An in situ hybridization protocol for detection and identification of planktonic bacteria. Syst. Appl. Microbiol. 19 : 403-406. GROSSART , H. P., T. BRINKHOFF , T. MARTENS , C. DUERSELEN , G. LIEBEZEIT , AND M. SIMON . 2004. Tidal dynamics of dissolved and particulate matter and bacteria in a tidal flat ecosystem in spring and fall. Limnol. Oceanogr. 49 : 2212-2222. GÜNTHER , C.-P. 1999. Settlement, secondary dispersal and turnover rate of benthic macrofauna, p. 133-146. In S. Dittmann [ed.] The Wadden Sea Ecosystem. New York: Springer Verlag. HAGSTROM , A., J. PINHASSI , AND U. L. ZWEIFEL . 2000. Biogeographical diversity among marine bacterioplankton. Aquat. Microb. Ecol. 21 : 231-244. KARAKURT , A., O. ABDIK , A. SENGUN , R. KARADAG , S. SARICAOGLU , H. Y. SARIKATIPOGLU , AND H. HASIRIPI . 2006. Stenotrophomonas maltophilia Endophthalmitis after cataract extraction. Ocul. Immunol. Inflamm. 14 : 41-6. KHAN , S. T., Y. NAKAGAWA , AND S. HARAYAMA . 2006. Krokinobacter gen. nov. , with three novel species, in the family Flavobacteriaceae . Int. J. Syst. Evol. Microbiol. 56 : 323- 328. KIRCHMAN , D. L. 2002. The ecology of Cytophaga-Flavobacteria in aquatic environments. FEMS Microbiol. Ecol. 39 : 91-100. KOBLIZEK , M., J. STON -EGIERT , S. SAGAN , AND Z.S. KOLBER . 2005. Diel changes in bacteriochlorophyll a concentration suggest rapid bacterioplankton cycling in the Baltic Sea. FEMS Microbiol. Ecol. 51 : 353-361. LLOBET -BROSSA , E., R. ROSSELLO -MORA , AND R. AMANN . 1998. Microbial community composition of Wadden Sea sediments as revealed by fluorescence-in-situ- hybridization. Appl. Environ. Microbiol. 64 : 2691-2696. LUDWIG , W., AND OTHERS . 2004. ARB: a software environment for sequence data. Nucl. Acids Res. 32: 1363-1371. LUNAU , M., A. LEMKE , O. DELLWIG , AND M. SIMON . 2006. Physical and biogeochemical controls of microaggregate dynamics in a tidally affected coastal ecosystem. Limnol. Oceanogr. 51 : 847-859. MANZ , W., R. AMANN , W. LUDWIG , M. WAGNER , AND K. H. SCHLEIFER . 1992. Phylogenetic oligodeoxynucleotide probes for the major subclasses of Proteobacteria – Problems and solutions. Syst. Appl. Microbiol. 15 : 593-600. MANZ , W., R. AMANN , W. LUDWIG , M. VANCANNEYT , AND K.H. SCHLEIFER . 1996. Application of a suite of 16S rRNA-specific oligonucleotide probes designed to investigate bacteria of the phylum Cytophaga-Flavobacter-Bacteroides in the natural environment. Microbiology 142: 1097-1106. MESSING , J. 1983. New M13 vectors for cloning. Meth. Enzymol. 101: 20-78. MIKKELSEN , O., AND A. PEJRUP . 1998. Comparison of flocculated and dispersed suspended sediment in the Dollard estuary, p. 199-209. In K. Black, D. Paterson and A. Cramp. [eds.] Sedimentary Processes in the Intertidal Zone. London, Geological Society.

67 Kapitel III Tidal effects on coastal bacterioplankton

MOESENEDER , M. M., J. M. ARRIETA , G. MUYZER , C. WINTER , AND G. J. HERNDL . 1999. Optimization of Terminal-Restriction Fragment Length Polymorphism Analysis for complex marine bacterioplankton communities and comparison with Denaturing Gradient Gel Electrophoresis. Appl. Environ. Microbiol. 65 : 3518-3525. MUSSMANN , M., K. ISHII , R. RABUS , AND R. AMANN . 2005. Diversity and vertical distribution of cultured and uncultured Deltaproteobacteria in an intertidal mudflat of the Wadden Sea. Environ. Microbiol. 7: 405-18. MUYZER , G., T. BRINKHOFF , U. NÜBEL , C. SANTEGOEDS , H. SCHÄFER AND C. WAWER . 1998. Denaturing gradient gel electrophoresis (DGGE) in microbial ecology. In : Akkermans ADL. J.D. van Elsas, F. J. Bruijn (eds) Molecular microbial ecology manual. Kluwer Academic Publishers, Dordrecht, 1-27. NEEF , A. 1997. Anwendung der in-situ-Einzell-Identifizierung von Bakterien zur Populationsanalyse in komplexen mikrobiellen Biozönosen. Ph.D. thesis. Technische Universität München. NIESEL , V., AND C. P. GÜNTHER . 1999. Distribution of nutrients, algae and zooplankton in the Spiekeroog backbarrier system, p. 77-94. In S. Dittmann [ed.] The Wadden Sea Ecosystem. Stability properties and mechanisms. Springer Verlag. PERNTHALER , A., J. PERNTHALER , AND R. AMANN . 2002. Fluorescence in situ hybridization and catalyzed reporter deposition for the identification of marine bacteria. Appl. Environ. Microbiol. 68 : 3094-3101. ------, AND J. PERNTHALER . 2005. Diurnal variation of cell proliferation in three bacterial taxa from coastal North Sea waters. Appl. Environ. Microbiol. 71 :4638-4644. POULSEN , L. K., G. BALLARD , AND D. A. STAHL . 1993. Use of rRNA fluorescence in situ hybridization for measuring the activity of single cells in young and established biofilms. Appl. Environ. Microbiol. 59 :1354-1360. RABUS R., M. FUKUI , H. WILKES , AND WIDDEL F. (1996). Degradative capacities and 16S rRNA-targeted whole-cell hybridization of sulfate-reducing bacteria in an anaerobic enrichment culture utilizing alkylbenzenes from crude oil. Appl. Environ. Microbiol. 62 : 3605-3613 RAPPÉ , M. S., K. VERGIN , AND S. J. GIOVANNONI . 2000. Phylogenetic comparison of a coastal bacterioplankton community with its counterparts in open ocean and freshwater systems. Fems Microbiol Ecol 33 : 219-232 RINK , B., S. SEEBERGER , T. MARTENS , C. DUERSELEN , M. SIMON , AND T. BRINKHOFF . 2006. Effects of a phytoplankton bloom in a coastal ecosystem on the composition of bacterial communities. Aquat. Microb. Ecol., submitted. SAMBROOK , J., E. F. FRISCH , AND T. MANIATIS . 1989. Northern Hybridisation. In Molecular Cloning: a laboratory manual, 2nd edn., Cold Spring Harbour Laboratory Press, New York. SCHÄFER , H., AND OTHERS . 2001. Microbial community dynamics in Mediterranean nutrient- enriched seawater mesocosms: changes in the genetic diversity of bacterial populations. FEMS Microbiol. Ecol. 34: 243-253. SCHUT , F. (1994) Ph.D. thesis. University of Groningen, Groningen, The Netherlands. SEIBOLD , A., A. WICHELS , AND C. SCHÜTT . 2001. Diversity of endocytic bacteria in the dinoflagellate Noctiluca scintillans . Aquat. Microb. Ecol. 25 : 229-235.

68 Kapitel III Tidal effects on coastal bacterioplankton

SEKAR , R., A. PERNTHALER , J. PERNTHALER , F. WARNECKE , T. POSCH , AND R. AMANN . 2003. An improved protocol for quantification of freshwater Actinobacteria by fluorescence in situ hybridization. Appl. Environ. Microbiol. 69 : 2928-2935. SELJE N., AND M. SIMON . 2003. Composition and dynamics of particle-associated and free- living bacterial communities in the Weser estuary, Germany. Aquat. Microb. Ecol. 30 :221-237. SELJE , N., M. SIMON , AND T. BRINKHOFF . 2004. A newly discovered Roseobacter cluster in temperate and polar oceans. Nature 427: 445-448. SFANOS , K., D. HARMODY , P. DANG , A. LEDGER , S. POMPONI , P. MCCARTHY , AND J. LOPEZ . 2005. A molecular systematic survey of cultured microbial associates of deep-water marine invertebrates. Syst. Appl. Microbiol. 28 : 242-264. SIMON , M., AND F. AZAM . 1989. Protein content and protein synthesis rates of planktonic marine bacteria. Mar. Ecol. Prog. Ser. 51 : 201-213. STAHL , D. A., AND R. AMANN . 1991. Development and application of nucleic acid probes. 205-248. In E. Stackebrandt and M. Goodfellow [eds.], Nucleic acid techniques in bacterial systematics. John Wiley & Sons Ltd., Chichester, England. STEVENS , H., T. BRINKHOFF , AND M. SIMON . 2005a. Composition of free-living, aggregate- associated and sediment surface-associated bacterial communities in the German Wadden Sea. Aquat. Microb. Ecol. 38: 15-30. ------, M. STÜBNER , M. SIMON , AND T. BRINKHOFF . 2005b. Phylogeny of Proteobacteria and Bacteroidetes from oxic habitats of a tidal flat ecosystem. FEMS Microbiol. Ecol. 54 : 351-365. TILLMANN , U., K.-J. HESSE AND F. COLIJN . 2000. Planktonic primary production in the German Wadden Sea. J. Plankton Res. 22 : 1253-1276. VAN DUYL , F. C., B. DE WINDER , A. J. KOP , AND U. WOLLENZIEN . 1999. Tidal coupling between carbohydrate concentrations and bacterial activities in diatom-inhabited intertidal mudflats. Mar. Ecol. Prog. Ser. 191 : 19-32. VAN LEUSSEN , W. 1996. Erosion/sedimentation cycles in the Ems estuary. Arch. Hydrobiol. Spec. Issues Advanc. Limnol. 47 : 179-193. WALLNER G., R. AMANN , AND W. BEISKER . 1993. Optimizing fluorescent in situ hybridization with rRNA-targeted oligonucleotide probes for flow cytometric identification of microorganisms. Cytometry. 14 : 136-143. WINTER , C., M. M. MOESENEDER , AND G. J. HERNDL . 2001. Impact of UV radiation on bacterioplankton community composition. Appl. Environ. Microbiol. 67 :665-672. YOON , J. H., S. J. KANG , C. H. LEE , AND T. K. OH. 2005. Dokdonia donghaensis gen. nov., sp. nov., isolated from sea water. Int. J. Syst. Evol. Microbiol. 55 : 2323-2328. ZUBKOV , M. V., B. M. FUCHS , S. D. ARCHER , R. P. KIENE , R. AMANN , AND P. H. BURKILL . 2002. A population of the alpha-Proteobacteria dominates the bacterioplankton and dimethylsulphoniopropionate uptake after an algal bloom in the North Sea. Deep-sea Res. II, Top. Stud. Oceanogr. 49 : 3017-3038.

69 Kapitel III Tidal effects on coastal bacterioplankton

Figure legends

Fig. 1. Temperature and salinity (a), dry weight (DW) and percent particulate organic carbon (%POC) (b), chlorophyll a (chl a), phaeopigments and phaeopigment/chl a ratio (c), abundance of total, particle-attached (PA) and free-living (FL) bacteria (d), and bacterial protein production (BPP) (e) in the Wadden Sea during a tidal cycle in July 2005. X-axis: Time of high tide (HT), mean tide (MT1/2) and low tide (LT).

Fig. 2. FISH counts for samples taken in November 1999, May 2000 and July 2005 during tidal cycles at high tide, mean tide, and low tide. Results of July 2005 were obtained using CARD-FISH. Calculation includes all values of single counting grids to display outliners (dots). Box-Whisker-Plots show the 25 th /75 th percentile (box), the mean (dashed line), the median (solid line) and the 5 th /95 th percentile (error bars). Abbreviations: EUB (EUB338), CFB (CFB319a), ALF (ALF968), GAM (GAM42a), BET (BET42a), ROS (ROS536), ARCH (ARCH915), SRB (SRB385), SRBdb (SRB385db), NON (NON338).

Fig. 3. DGGE fingerprints of sediment surface (SE), particle-attached (PA), and free-living (FL) bacterial communities of the Wadden Sea during tidal cycles in November 1999 (Nov) and May 2000 using primer sets for 16S rRNA genes of Bacteria . The arrows mark excised and sequenced bands. Bands sequenced in this study include “TC” for tidal cycle, bands named GWS-eX- PA/FL refer to sequences of Rink et al. (2006). Std. = standard.

Fig. 4. DGGE fingerprints of free-living (FL) and particle-attached (PA) bacterial communities during a tidal cycle in July 2005, using primer sets for 16S rRNA genes of Bacteria (EUB), Bacteroidetes (CFB), alpha-Proteobacteria (ALF) and the Roseobacter clade (ROS). “DNA” marks banding patterns obtained from 16S rRNA gene fragments and “cDNA” refers to banding patterns derived from 16S rRNA after reverse transcription. HT = high tide, MT1/2 = mean tide, LT = low tide. The numbered arrows mark excised and sequenced bands. Because of the small fragment size of the Roseobacter amplicons (approx. 200 bp) the DGGE bands were not excised for sequencing. Std. = standard.

Fig. 5. Cluster analyses of DGGE banding patterns (July 2005, Fig. 4) of particle-attached (PA) and free-living (FL) Bacteria (EUB), Bacteroidetes (CFB), alpha -Proteobacteria (ALF) and the Roseobacter clade (ROS). “DNA” marks banding patterns obtained from 16S rRNA

70 Kapitel III Tidal effects on coastal bacterioplankton gene fragments and “cDNA” refers to banding patterns derived from 16S rRNA after reverse transcription. Samples were taken during a tidal cycle at high tide (HT), mean tide (MT1/2) and low tide (LT). The similarity matrix was calculated using UPGMA and Pearson correlation. Std. = standard.

Fig. 6. Phylogenetic trees of Proteobacteria and Actinobacteria (A) and Bacteroidetes (B) calculated with Maximum-Likelihood based on 16S rRNA gene fragments. Sequences obtained in this and one earlier study partially including the same samples [Rink et al. (2006), compare Fig. 3] are highlighted in bold. Alpha, delta, gamma and beta = different phylogenetic groups of Proteobacteria , WAC I = Wadden Sea Alpha Cluster (Stevens et al. 2005), RCA = Roseobacter Clade Affiliated (Selje et al. 2004), SAMMIC = Surface Attached Marine MICrobes (Stevens et al. 2005). “GWS” (German Wadden Sea) indicates clones obtained from the same habitat in this and earlier studies. The percentage of sequence divergence is indicated by the scale bars.

Supplementary Data

Fig. I. Particulate organic carbon (POC), ratio of POC per dry weight (DW) and density of particle-attached bacteria at high tide (HT), mean tide (MT) and low tide (LT) in July 2005.

Fig. II. DGGE fingerprints of sediment surface, particle-attached, and free-living bacterial communities of the Wadden Sea during tidal cycles in November 1999 and May 2000 using primer sets for 16S rRNA genes of Bacteria .

71 Kapitel III Tidal effects on coastal bacterioplankton

Table 1. FISH counts of November 1999, May 2000 and July 2005 during tidal cycles taken at high tide (HT), mean tide (MT) and low tide (LT) in percent. Results of July 2005 were obtained using CARD-FISH. n = number of analysed subsamples, std dev = standard deviation

Table. 1. Rink et al.

72 Kapitel III Tidal effects on coastal bacterioplankton

Fig. 1. Rink et al.

73 Kapitel III Tidal effects on coastal bacterioplankton

Fig. 2. Rink et al.

SE PA FL St No Ma No Ma No Ma St

GWS -TC -e5 -

GWS -e3 - GWS -e12 - GWS -e5 - GWS -e13 -

GWS -e6 - GWS -TC -e3 - GWS -TC -e1 - GWS -e7 - GWS -e8 - GWS -TC -e4 - GWS -TC -e2 -

Fig. 3. Rink et al.

74 Kapitel III Tidal effects on coastal bacterioplankton

DNA cDNA

Std. PA FL PA FL Std. H M L M H M L M H M L M H M L M T T T T T T T T

GWS -TC -e6 - GWS -TC -e7 - GWS -TC -e10 -

GWS -TC -e9 - GWS -TC -e8 - EUB GWS -TC -e11 - GWS -TC -c4 - GWS -TC -c3 - GWS -TC -c2 -

GWS -TC -c1 - CFB

GWS -TC -a1 - GWS -TC -a3 - GWS -TC -a6 - GWS -TC -a2 - GWS -TC -a4 -

GWS -TC -a5 - ALF

ROS

Fig. 4. Rink et al.

75 Kapitel III Tidal effects on coastal bacterioplankton

Pearson correlation [0.0% -100.0%]

EUB

CFB

ALF

ROS

Fig. 5. Rink et al.

76 Kapitel III Tidal effects on coastal bacterioplankton

A WAC I

Roseobacter alpha

RCA

delta

gamma SAMMIC

beta

Actinobacteria

Fig. 6A. Rink et al.

77 Kapitel III Tidal effects on coastal bacterioplankton

B

Fig. 6B. Rink et al.

78 Kapitel III Tidal effects on coastal bacterioplankton

Supplementary Data

Fig. I. Rink et al.

79 Kapitel III Tidal effects on coastal bacterioplankton

Fig. II. Rink et al.

80

IV. Diversity and abundance of Gram-positive bacteria in a tidal flat ecosystem

81 Kapitel IV Diversity and abundance of Gram-positive bacteria

Diversity and abundance of Gram-positive bacteria in a tidal flat ecosystem

Heike Stevens, Thorsten Brinkhoff, Beate Rink, John Vollmers, and Meinhard Simon*

Institute for Chemistry and Biology of the Marine Environment (ICBM), University of Oldenburg, Germany

Running title : Gram-positive bacteria in tidal flats

Key words: Gram-positive bacteria, Actinobacteria, Firmicutes, tidal flats, dilution cultures, fluorescence in situ hybridization, DGGE

______*Corresponding author Mailing address: ICBM, University of Oldenburg, PO Box 2503, D-26111 Oldenburg, Germany. Phone: +49 (0) 441 / 798-5361. Fax: +49 (0) 441 / 798-3438. E-mail: [email protected]

82 Kapitel IV Diversity and abundance of Gram-positive bacteria

Abstract. Gram-positive (Gram+) bacteria recently have been identified as important components of freshwater ecosystems and are also present in marine environments. However, their quantitative significance and possible role in the latter systems is still little studied, in particular in coastal regions. Therefore, we investigated the abundance and composition of Gram+ bacteria in the Wadden Sea, a tidal flat ecosystem in the German Bight of the North Sea. Applying fluorescence in situ hybridization we found that Actinobacteria constitute 4-7% of total bacteria in the Wadden Sea and slightly higher proportions in a freshwater drainage channel connected to the sea by a sluice. The application of DGGE of 16S rRNA gene fragments after amplification by an Actinobacteria -specific primer set and subsequent sequencing showed that the composition of the actinobacterial community in the Wadden Sea was distinctly different from that in the freshwater system. A clone library of 103 clones yielded 8 Gram+ phylotypes which are related closely to other marine phylotypes including the Marine Actinobacteria Clade but also to freshwater phylotypes. We applied dilution cultures, enriched with various biopolymers for isolating bacteria from the bulk water, suspended aggregates, the oxic surface and oxic/anoxic transition zone of the sediment, Marine Broth and Fucus vesiculosus extracts. Fifty three isolates affiliated to seven families of the order Actinomycetales and 9 isolates to the family Bacillaceae . The salinity range (1 to 45‰ NaCl) and growth optimum of fourteen strains from various families showed that all except one strain exhibited a rather broad range of sustained growth from 1 to >20‰ NaCl and several strains exhibited an optimum of >10‰ NaCl. The results indicate that the Gram+ bacterial community in the Wadden Sea is surprisingly diverse and consists mainly of indigenous species which appear to be well adapted to the environmental conditions of this coastal ecosystem.

83 Kapitel IV Diversity and abundance of Gram-positive bacteria

Introduction In the recent past, more and more evidence accumulated which indicated that Gram-positive (Gram+) bacteria, in particular Actinobacteria , are of hitherto unknown significance in aquatic ecosystems (Bull et al. , 2005). Studies in freshwater systems showed a surprising diversity and abundance of Actinobacteria even though the specific ecological role remains to be unveiled (Glöckner et al. , 2000; Hahn et al. , 2003; Stepanauskas et al. , 2003; Warnecke et al. , 2004; Haukka et al. , 2005; Allgaier and Grossart, 2006). It has been known for more than 60 years that Gram+ bacteria also exist in marine environments, but they were thought not to be indigenous but introduced from terrestrial habitats (Zobell and Upham., 1944; Goodfellow and Haynes, 1984). Since the mid-1990s, studies using culture-independent but also refined culture-dependent methods indicated that Gram+ bacteria reveal an unexpected diversity in marine bacterioplankton communities (Fuhrman et al. , 1993; Rappé et al. , 1997; Suzuki et al. , 1997; Rappé et al. , 1999; Fuchs et al. , 2005) as well as in marine sediments and a surprisingly high abundance of up to 13% (Jensen and Fenical, 1995; Moran et al. , 1995; Urakawa et al. , 1999; Mincer et al. , 2002; Maldonado et al. , 2005; Pathom-aree et al. , 2006). Phylogenetic analyses of the phylotypes and isolates obtained in these studies implied that, in fact, indigenous marine Gram+ bacteria exist. Some of the Gram+ bacteria obtained from marine habitats fall into distinct “marine” clusters, only distantly related to clusters comprising also Gram+ bacteria from freshwater and soil. The Marine Actinobacteria Clade, (Rappé et al. , 1999), which includes the “BD1-5 cluster” (Fuhrman et al. , 1993), a deeply branching cluster within the Actinobacteria , comprises exclusively marine bacterioplankton phylotypes. The MAR 1 cluster (Mincer et al. , 2002) consists of Actinobacteria isolates from tropical and subtropical marine sediments, but is branching not as deeply as the Marine Actinobacteria Clade. Quantitative studies on marine Gram+ bacteria are still scarce. A biomarker study on the basis of the composition of phospholipid ester-linked fatty acids (PLFA) indicated that Gram+ bacteria are major components of bacterial communities in sediments of a eutrophic bay (Rajendran et al. , 1994). Actinobacteria constituted up to 5% of total bacteria in shallow marine sediments and <1.4% in an Arctic deep-sea sediment as determined by dot blot and fluorescence in situ hybridization (Moran et al. , 1995; Llobet-Brossa et al., 1998; Ravenschlag et al. , 1999). In the Sargasso Sea and the Arabian Sea, Actinobacteria have been identified as substantial components of the bacterioplankton (Fuchs et al. , 2005; Morris et al. , 2005). In the Delaware estuary, USA, Actinobacteria constitute a decreasing fraction of total bacterioplankton numbers and of the proportions assimilating glucose and polysaccharide with

84 Kapitel IV Diversity and abundance of Gram-positive bacteria increasing salinity (Elifantz et al. , 2005; Kirchman et al. , 2005). It is of particular interest to reveal the significance, composition and abundance of Gram+ bacterial communities in coastal tidally affected ecosystems and whether in these systems indigenous Gram+ bacteria do exist These systems are severely understudied with respect to Gram+ bacteria as compared to other marine systems (Bull et al. , 2005). The Wadden Sea is a shallow and nutrient-rich tidally affected coastal ecosystem of the southern North Sea stretching from the Netherlands (Den Helder) to Denmark (Esbjerg). Due to the pronounced tidal dynamics and inputs of organic and inorganic nutrients from land, freshwaters, and the North Sea it can be considered as a melting pot in which microbial processes are of major significance (Dittmann, 1999; Poremba et al. , 1999). In the recent past extensive work on microbial processes and on the composition of bacterial communities in the bulk water, on suspended aggregates and in the sediment of the Wadden Sea has been carried out (e.g. Llobet-Brossa et al. , 1998; Köpke et al. , 2005; Stevens et al. , 2005a, 2005b; Lunau et al. , 2006; Wilms et al. , 2006). This study investigated the composition and abundance of the Gram+ bacterial community in the Wadden Sea. We applied cultivation-based approaches, using enrichment and dilution cultures amended with a variety of biopolymers, as well as cultivation-independent approaches using Gram+-specific CARD-FISH (CAtalyzed Reporter Deposition Fluorescence In Situ Hybridization, Sekar et al. , 2003), DGGE (Denaturing Gradient Gel Electrophoresis, Muyzer et al. , 1993) and clone library construction. The salt requirements for growth of representative isolates were examined by determining their growth adaptation to the ambient salinity range.

RESULTS

Isolation Seventeen of 63 bulk water isolates (27%) obtained from dilution cultures of the May sample were Gram+ strains. Isolates affiliating to α- and γ-Proteobacteria constituted proportions of 44.4% and 20.6%, respectively, and those affiliating to Bacteroidetes 7.9%. In October, the majority of strains affiliated to α -Proteobacteria (31.0%), Actinobacteria (30.2%) and γ - Proteobacteria (24.0%). Including the Firmicutes isolates (n=6) 45 of the 129 October strains (34.9%) were Gram+ bacteria, of which 35.6% originated from the bulk water, 20% from aggregates and the transition zone each and 24.4% from the oxic layer of the sediment. In other studies of which Gram+ isolates were included in the phylogenetic analysis, only the

85 Kapitel IV Diversity and abundance of Gram-positive bacteria bulk water (Bruns et al. , 2003; Selje et al. , 2005) or aggregates (Grossart et al. , 2004) were investigated. Gram+ isolates were obtained with all substrates used. The only isolate obtained with laminarin also affiliated to Gram+ bacteria. The highest MPN dilution steps which yielded isolates, was 10 -8. These isolates were derived from bulk water in May with casein and from the sediment transition zone in October with MB as substrate and affiliated to Actinobacteria and Firmicutes , respectively. Overall most Gram+ isolates obtained in May and October were isolated with alginate (n=11), followed by agar (n=9), MB and starch (n=8). Only few isolates were retrieved from dilution cultures amended with casein (n=2) and palmitate (n=4). Dilution steps 10 –1, 10 –2 and 10 –4 yielded >10 and the others not more than 7 Gram+ isolates (Fig. 2).

Salinity-dependent growth All strains except one exhibited a rather broad salinity range of sustained growth and 9 strains grew equally well from 1 to >20‰ NaCl (Table 1) and thus appeared well adapted to the salinities occurring in the Wadden Sea. Two strains grew significantly better at salinity ranges >10‰ (GWS-BW-H260, GWS-BW-H252). One strain, affiliating to Bacilli, grew significantly better at salinities <5‰.

CARD-FISH Hybridization efficiencies varied from 48.9 to 67.7% as indicated by the ratio of numbers of EUB338 positive over DAPI cell counts and subtracting the non-EUB338 numbers which remained below 1.4% (Table 2). Gram+ bacteria were detected at all five locations and proportions ranged from 4.2 to 8.0% of DAPI cell counts (Table 2). The highest proportion occurred in the freshwater location behind the sluice.

DGGE analysis The banding patterns of the Actinobacteria –specific DGGE revealed distinct communities of the freshwater and the marine locations (Fig. 3). The marine samples were rather similar with one prominent band occurring in all samples. However, the particle-associated and free-living bacterial fractions exhibited also different bands. Band numbers ranged between 4 in the particle-associated bacterial fraction of station 4 and 9 in the free-living bacterial fraction of

86 Kapitel IV Diversity and abundance of Gram-positive bacteria station 2. Sequencing of the excised bands yielded only actinobacterial 16S rRNA phylotypes, indicating the specificity of the primers.

Clone library Altogether 103 different clones were sequenced . Two of them were identified as chimera. Eight of the remaining clones (7.8%) affiliated to Gram+ bacteria and the majority to γ- and δ-Proteobacteria .

Phylogenetic affiliation A total of 94 Gram+ 16S rRNA gene sequences was phylogenetically analyzed (Fig. 4A and B, Tab. 3), 82 from isolates, 8 from the clone library and 4 from DGGE bands. To avoid an overestimation of diversity, sequences with similarities of ≥99% were merged into 15 “sequence-groups” (I – XV) resulting in a total of 58 different sequence types (Tab. 3, Fig. 4). Sixty-three sequences from isolates, 7 from clones and 4 from DGGE bands affiliated to Actinobacteria . The sequences mainly clustered into seven families of the order Actinomycetales , the Microbacteriaceae , Micrococcaceae , Mycobacteriaceae , Nocardiaceae , Nocardioidaceae , Promicromonosporaceae , Pseudonocardiaceae , and the Sanguibacteraceae . Most sequences group with the Microbacteriaceae (18 isolates, 1 clone), with the Micrococcaceae (16 isolates) and with the Nocardioidaceae (15 isolates). Because of their low sequence similarity to the next relative, Streptomyces cinnabarus and candidatus “Microthrix parvicella ”, clone K39 and DGGE bands GWS-DG2, GWS-DG1, GWS-FL-8, GWS-DG3 could only be classified on the class-level (Table 3). The latter three phylotypes clustered together with various phylotypes of uncultured marine Actinobacteria . Clone GWS- K46 and isolates GP-5 and GP-6 belong to the order Actinobacteria with next related species of the subclass Frankinae . Clones GWS-K72, GWS-K105 and GWS-K112 clustered within the Marine Actinobacteria Clade. Sixteen of the 58 Gram+ sequence types (28%) obtained from the Wadden Sea exhibit a closest relative of marine origin, 37 of other environments, such as freshwater, soil, plants, or endophytic habitats. For 5 sequence types no information is available on the source of isolation of the next related sequence (Table 3). Our strains affiliating with the Micrococcaceae were mainly isolated from bulk water, isolates affiliating with the Nocardioidaceae from bulk water and aggregates. For all other sequences and isolates no

87 Kapitel IV Diversity and abundance of Gram-positive bacteria relationship exists between phylogeny and habitat or substrate from which they were obtained. Eight of our strains, 10 from other studies of the same habitat and one clone (K48, not shown in Fig. 4B) affiliate to the phylum Firmicutes . Fifteen of the isolates were merged into five sequence groups. All isolates affiliate to the order Bacillales and, except sequence group XIV (GP-13, GP-14) and isolate GWS-TZ-H232, fall within the family Bacillaceae . Sequence group XIV affiliates to the Planococcaceae . The clone K48 affiliates within the class Clostridia to the “ Peptostreptococcaceae ” (family name not validly published), with the next relative Fusibacter paucivorans (Table 3). Isolates affiliating to the Firmicutes were never obtained from aggregates, but from the sediment layers and the bulk water. They were obtained from assays performed with MB, agar, alginate, and stearine.

Discussion

Our results show that Gram+ bacteria constitute a substantial fraction of the bacterial community in the Wadden Sea and that they are indigenous to this environment. They comprise around 5% of total bacterial numbers and are a major component of its so far cultivated fraction. However, the isolated strains were only distantly related to most phylotypes retrieved from the clone library and from the Actinobacteria -specific DGGE bands. Most of the Gram+ isolates tested grew at a rather wide salinity range including 20 and 30‰ NaCl, thus indicating that they are well adapted to growing in the Wadden Sea , in which salinities of 26 to 33 psu occur (http://las.physik.uni-oldenburg.de/wattstation). Even though it seems possible that Gram+ bacteria are also washed in from terrestrial run off and by releasing drained fresh water through the dike sluices these bacteria do not appear to constitute prominent members of the marine Gram+ bacterial community. The DGGE bands of the freshwater sample were distinctly different from the marine ones and only one isolated strain exhibited a distinct growth optimum at salinities <5‰. Further, several clones of the clone library affiliated to the Marine Actinobacteria Clade (Rappé et al. , 1999), and the sequence of one prominent DGGE band detected at all marine stations (GWS-DG3) affiliated to marine Actinobacteria phylotypes and most closely to a phylotype detected in the northwestern coastal Pacific (Fig. 4; Table 3; Morris et al. , 2006). Gram+ bacteria comprised 27 and 35% of all isolates retrieved in May and October 1999, respectively, and the majority affiliated to Actinobacteria . Thus, Gram+ bacteria constitute one of the two major cultivable phylogenetic classes of isolates in the Wadden Sea.

88 Kapitel IV Diversity and abundance of Gram-positive bacteria

Because the Gram+ bacteria were obtained with various substrates we assume that the frequent isolation of these organisms was not due to a cultivation bias but rather a result of the various biopolymers used as enrichment substrates. Even though most of the isolates were retrieved from low dilution steps, several isolates from various habitats were obtained from 10 -6 to 10 -8 dilutions, suggesting that some of them are significant constituents of the bacterial community in this ecosystem. This is particularly true for sequence groups I (Microbacteriaceae ) and X ( Nocardioidaceae ). Three of the strains within group I were obtained from 10 -6 or higher, and 6 of the 7 strains in group X from 10 -4 or higher (Fig. 4). This suggestion is supported by the finding that the recently described Aeromicrobium marinum, isolated from a 10 -4 dilution culture retrieved from the Wadden Sea and the first described marine species within the family Nocardioidaceae , constitutes up to 1% of total bacteria in the water column (Bruns et al. , 2003). Several other recent studies also isolated Actinobacteria from various marine environments, but in most cases they constituted lower fractions of all isolates as compared with our results. In studies from various regions including the German Bight of the North Sea, the Baltic Sea and the Oregon coast of the Pacific, which examined the diversity of isolates from bacterioplankton samples and enrichments with various substrates, between <1 and 15% of all isolates affiliated to Actinobacteria (Suzuki et al. , 1997; Eilers et al. , 2000; Hagström et al. , 2000; Uphoff et al., 2001). Du et al. (2006) reported that Gram+ bacteria were a prominent fraction of the pigmented cultivable bacterial community in Chinese estuaries and were also present in the coastal sea, even though to much lower proportions. In various marine sediments, ranging from shallow coastal regions to the deepest ocean, Gram+ bacteria of a rather wide diversity have been isolated in recent studies (Mincer et al. 2002; Köpke et al. , 2005 ; Maldonado et al. , 2005; Pathom-aree et al. , 2006). These notions indicate that Gram+ bacteria constitute a greater fraction of the so far cultivable proportion of the bacterial community in coastal environments than previously assumed. The fraction of Actinobacteria we assessed by CARD-FISH is in the same range as that reported from the Delaware Bay, USA (Kirchman et al. , 2005). These authors found that Actinobacteria constitute decreasing proportions of 20-30% of total bacterial numbers to 5% from the freshwater end of the estuary to the marine Bay. This study did not examine the phylogenetic composition of the Gram+ bacterial community in the Bay. According to our DGGE and sequencing results, however, it appears most likely that the composition of the Gram+ bacterial community in the Bay was different from the freshwater section and that the decreasing proportion of this phylogenetic class was not only a dilution effect. Our results

89 Kapitel IV Diversity and abundance of Gram-positive bacteria further imply that an indigenous marine actinobacterial community exists, also in such near shore environments with intense water exchange with estuarine and freshwater habitats. The proportions of Actinobacteria we and Kirchman et al. (2005) found in the water column of the Wadden Sea and the Delaware Bay are somewhat higher than proportions reported from the upper 4 cm of Wadden Sea sediments, <1 to 3.6% (Llobet-Brossa et al. , 1998) and also as those of the genus Streptomyces detected in a shallow marine sediment by dot blot hybridization with a genus-specific probe (2.0 to 5.1% of total extracted rRNA; Moran et al. , 1995). Studies based on PLFA indicate that Gram+ bacteria are significant components of the bacterial communities in eutrophic bays in Japan (Rajendran et al. , 1994; Rajendran and Nagatomo, 1999). In open ocean environments the significance of Actinobacteria appears to be variable. Whereas in the Sargasso Sea, Actinobacteria constitute <2% of total bacterial numbers (Morris et al. , 2005), in the Arabian Sea variable proportions of 2 and 13% of total bacteria were found in oligotrophic waters and in the oxygen minimum zone, respectively (Fuchs et al. , 2005). Hence, these data indicate that Actinobacteria are a prominent component of the bacterial community in shallow coastal ecosystems, in the water column as well as in the sediment, even though its quantitative proportion is lower than that of other important classes such as - and -Proteobacteria and Bacteroidetes (Kirchman et al. , 2005; Llobet-Brossa et al. , 1998) and lower than that of Actinobacteria in freshwater ecosystems (Glöckner et al. , 2000; Allgaier and Grossart, 2006). Our phylogenetic analysis of sequences from the isolates, the DGGE bands and clone library shows a surprisingly high diversity of the Gram+ bacteria. The sequences from the DGGE bands and the clone library affiliated to distinctly different groups than the isolates of this phylogenetic lineage. The diversity within Actinobacteria we detected was greater than that described in other studies using either culture-dependent (Jensen and Fenical, 1995; Suzuki et al. , 1997; Köpke et al. , 2005; Maldonado et al. , 2005; Du et al. , 2006; Pathom-aree et al ., 2006) or culture–independent approaches (Fuhrman et al. , 1993; Gray and Herwig, 1996; Suzuki et al. , 1997; Rappé et al. , 1999; Urakawa et al. , 1999; Wilms et al. , 2006). This may be due to the various isolation procedures we applied, such as dilution series and different substrates and to the various habitats we sampled, but may also reflect the specific signature of the Wadden Sea ecosystem as a melting pot with marine as well as terrestrial impacts. The fact that we successfully enriched and isolated Actinobacteria from various habitats of the Wadden Sea with various biopolymers shows that these strains are capable of degrading a variety of polymeric substances. Some of these substances are typical for coastal

90 Kapitel IV Diversity and abundance of Gram-positive bacteria marine environments such as F. vesiculosus , cellulose, starch, chitin, and laminarin. Actinobacteria are well known to be capable to degrade various polymeric substances such as cellulose and lignin, but also rubber and polyester (Haider et al. , 1978; Godden and Penninckx, 1984; Jendrossek, 1997; Pranamuda et al. , 1999; Linos et al. , 2002). Their hydrolytic potential appears comparable or even greater than that of the Sphingobacteria and Flavobacteria group of the Bacteroidetes phylum (Reichenbach, 1992; Kirchman, 2002). It may explain why Actinobacteria prosper in the bulk water and sediment of the Wadden Sea and other coastal environments which are characterized by high concentrations of various biopolymers (Harvey and Mannino, 2001). In fact, Piza et al. (2004) found a surprising diversity of Actinobacteria in a Brazilian estuary subjected to high pollution during the last fifty years, and in particular at its brackish end. This high diversity obviously reflects the high potential of Actinobacteria to degrade complex organic substances including recalcitrant compounds and pollutants. In contrast to sequences affiliating to Actinobacteria , our sequences affiliating to Firmicutes were much less divers. Sequences of all bacterial isolates of this group clustered within the class Bacilli . The only clone obtained from the Firmicutes belongs to the class Clostridia . Firmicutes appear to be more prominent members of the subsurface sediment, possibly because of their ability to produce endospores. In a recent study a surprisingly high diversity of Firmicutes was found in subsurface sediments of the German Wadden Sea (Köpke et al., 2005). Since the early 1970s various Bacilli were isolated from marine habitats (e.g., Denis, 1971; Bonde, 1976; Stackebrandt et al., 1997; Urakawa et al., 1999; Siefert et al., 2000), but only very few marine isolates affiliated to the classes Lactobacillales , Clostridia, and Mollicutes (Finne and Matches, 1974; Timmis et al., 1974; Marty, 1986; Gray and Herwig, 1996). Hence, Bacilli seem to be the most abundant marine Firmicutes . A further feature, explaining the high diversity of Gram+ bacteria in coastal ecosystems may be their ability to grow at wide salinity ranges. From the 14 strains whose growth adaptation to salinity we tested, all except one were able to grow at salinities >5‰ NaCl and exhibited growth optima ranging to 20‰ or higher (Table 1). Four of the 14 isolates had as closest relative a strain also isolated from a marine habitat (GWS-BW-H301M, GWS- AG-H268, GWS-SE-H117, GWS-TZ-H232, Table 3) and two of the latter isolates grew only at salinities >5‰ NaCl. The fact that the next relatives of the other isolates were not of marine origin is no indication that our isolates growing at higher salinity ranges were not truly marine strains. Aeromicrobium marinum , the recently described marine species within the family Nocardioidaceae , clusters with terrestrial isolates but exhibits a requirement for seawater

91 Kapitel IV Diversity and abundance of Gram-positive bacteria typical of marine bacteria (Bruns et al. , 2003). Hence, these marine species clustering closely together with other non-marine species may indicate that they were introduced into coastal habitats from adjacent soil and freshwater habitats. This notion underlines the significance of the close interactions of the land sea transition zone for evolutionary processes and may further explain the great diversity of Gram+ bacteria occurring in estuarine and coastal environments. Four of the five phylotypes we obtained from the clone library and the DGGE-bands of the marine samples clustered closely together with other phylotypes of exclusively marine origin and only distantly related to phylotypes or isolates of freshwater or soil origin (Fig. 4). Because none of the isolates appeared as a prominent DGGE band with the Actinobacteria - specific PCR, the in situ dominant components of the Gram+ bacterial community presumably constituted of these yet uncultured phylotypes to a great extent. Obviously the community of Gram+ bacteria in coastal environments consists of two fractions, one including the cultivated strains and one including only not-yet cultivated phylotypes, mainly affiliated to the Marine Actinobacteria Clade (Rappé et al., 1999), and whose physiological traits are basically unknown. Whereas the indigenous community appears to be most important for the turnover of organic matter at ambient conditions, the cultivable fraction appears as a valuable resource for isolates capable of interesting catalytic pathways and for bioactive compounds (Grossart et al. , 2004; Bull et al. , 2005; Maldonado et al. , 2005). Our analysis of the community of Gram+ bacteria in the Wadden Sea shows that it was surprisingly divers, mainly consisting of various groups of Actinobacteria , and to a much lesser extent of Firmicutes . Phylotypes and isolates clustered to distinctly different groups. The phylogenetic affiliation of the phylotypes and the broad salinity range of most of the isolates indicates that the Gram+ bacterial community in the Wadden Sea is well adapted and indigenous to the marine environment. On the basis of the CARD-FISH results we estimate that Gram+ bacteria constitute around 5% of total bacteria in the water column of the Wadden Sea. Hence, they appear to be prominent members of the bacterial communities and, because of their high potential to degrade various biopolymers, are important in the turnover and decomposition of organic matter in this ecosystem.

92 Kapitel IV Diversity and abundance of Gram-positive bacteria

Experimental procedures

Sampling Samples were collected on 27 May and 25 October 1999 in the East Frisian Wadden Sea, Germany (station A, 53° 37´ N, 07° 08´ E; station B 53° 42´ N, 07° 43´ E; Fig 1). Water samples were taken at high tide with pre-rinsed 10 l-plastic jugs. Sediment cores from an intertidal mixed sand/mud flat were taken only in October with Plexiglas tubes (36 mm diameter) at low tide. Samples were brought to the lab on ice in cooling boxes and processed further within 2 h. Water samples for CARD-FISH and DGGE analysis were taken on 17 August 2005 at five sampling points along a salinity gradient from 0.3 to 31 psu close to station B (Fig. 1) and processed further within 3 hours. For DGGE analysis, 100 ml of water were filtered onto 5.0 µm polycarbonate filters (Nuclepore, Whatman) to obtain particle- associated bacteria and subsequently onto 0.2 µm polycarbonate filters to obtain free-living bacteria. Filters were stored at -20°C until nucleic acid extraction.

Isolation of bacteria For the October samples we applied the MPN (most probable number) technique (Trolldenier, 1993) with different substrates and subsequent isolation of bacteria (Stevens et al. , 2005b). For the May samples dilution series were used for isolation of bacteria. Therefore, 1 ml of bulk water was used as inoculum for 10-fold dilution series. Mineral media amended with various substrates and MB 2216 (Difco, Germany) were prepared as described previously (Stevens et al. , 2005b). The following substrates were added (0.1%): agar, alginate, casein, cellulose, chitin, laminarin, dried and pestiled Fucus vesiculosus (a brown algae growing copiously along the coast line of the Wadden Sea) , palmitate, starch and stearate. Growth was checked by turbidity and microscopically. Bacteria were isolated from various dilution steps on agar plates containing the same media as in the MPN assays amended with 1.5% agar. For further cultivation Marine Agar 2216 (Difco, Germany) was used. Additionally, three isolates were obtained from 1 l-rolling tanks filled with natural seawater and amended with 0.1% agar and alginate, respectively, and incubated for 100 days at 15 °C in the dark. Single colonies were transferred at least five times until considered as pure. The purity of the isolated strains was examined by DGGE analysis according to Brinkhoff and Muyzer (1997). Isolates from the same habitat and from assays with the same substrate with sequence similarities of ≥99% [as determined by a similarity matrix calculated with ARB (Ludwig et al. , 2004)] were

93 Kapitel IV Diversity and abundance of Gram-positive bacteria considered identical and only one sequence, either from the highest dilution step or from the highest number of sequenced bp, was used in this study and submitted to GenBank.

Salinity dependent growth In order to examine growth adaptation to the ambient salinity range 14 selected isolates from the major families of Gram+ bacteria (Table 1) were grown on MB agar plates and transferred to liquid MB medium of a salinity of 20‰ NaCl (w/v). These cultures were used to inoculate 5 ml liquid MB cultures in triplicate test tubes of salinities of 1, 5, 10, 20, 30 and 45‰ NaCl. Test tubes were incubated in the dark at 20°C and growth was monitored over 4 to 10 days by examining periodically the optical density at 660 nm spectrophotometrically directly in the test tubes. Growth rates were determined as the slope of the exponential growth phase of the log-plotted growth curves.

Fluorescence in situ hybridization (CARD-FISH) CARD-FISH analyses were carried out basically following the protocol of Pernthaler et al. (2004). Four replicates of one ml of each sample were filtered onto a 0.2 µm polycarbonate filter (25 mm diameter, Nuclepore, Whatman) and fixed with 3 ml of paraformaldehyde (4% w/v) for one hour. Subsequently, filters were embedded in low-gelling point agarose (0.2% w/v, Metaphor) and incubated in lysozyme (10 mg ml -1) and achromopeptidase (2 U µl -1) as described by Sekar et al. (2003). Cells were hybridized with the probes HRP-EUB338 (Amann et al. , 1990), HRP-HGC69a (Roller et al. , 1994) and HRP-NON338 (Wallner et al. , 1993) for 2 hours at 35°C. Amplification was performed at 37°C using FITC-labeled Tyramide (fluorescein-5-isothiocyanate, Invitrogen) for 30 min in the dark. Samples were incubated in 1x PBS (pH 7.3) amended with Triton X-100 (0.05%) for 30 min at room temperature in the dark to remove residual dye. Cells were counterstained using Vectashield with DAPI (4´,6´-diamidino-2-phenylindol, 1.5 µg ml-1; Vector Laboratories, Peterborough, UK) and kept frozen at -20°C until further processing. Enumeration was performed by epifluorescence microscopy (Axiolab, Zeiss, Germany) at 1,000 x magnification.

Nucleic acid extraction Genomic DNA was extracted after bead-beating by phenol-chloroform as described earlier (Stevens et al. , 2005a). Precipitation of nucleic acids was done overnight at -20°C using isopropanol. DNA pellets were resuspended in molecular grade water (Eppendorf, Hamburg, Germany) and stored at -20°C until further processing.

94 Kapitel IV Diversity and abundance of Gram-positive bacteria

PCR amplification of 16S rRNA gene fragments PCR amplifications were performed with an Eppendorf Mastercycler (Eppendorf) as described previously (Brinkhoff and Muyzer, 1997). 16S rRNA gene fragments of Actinobacteria were amplified using the primers S-C-Act-235-a-S-20 (CGCGGCCTATCAGCTTGTTG, forward) and S-C-Act-878-a-A-19 (CCGTACTCCCCAGGCGGGG, reverse, Stach et al. , 2003). A GC-clamp was added to the forward-primer for subsequent DGGE analysis (Muyzer et al. , 1998). Amplification was performed according to Stach et al. (2003) with the following modifications: Denaturing and annealing were extended from 45 to 60 s and elongation from 1 to 3 min. Touchdown from 72°C to 67°C in 0.5°C steps was done with two instead of one cycle per step and annealing at 68°C was extended from 15 to 20 cycles. Four µl of the amplification products were analyzed by electrophoresis in 2% (w/v) agarose gels and stained with ethidium bromide (1 µg ml -1) (Sambrook et al. , 1989). For subsequent sequencing analysis PCR products were purified by using the Qiaquick PCR purification kit (Qiagen Inc.).

DGGE analysis of PCR products DGGE was performed with the D-Code system (Bio-Rad Laboratories, Inc.). For the 16S rRNA gene fragments of Actinobacteria , a gradient of 35 to 85% denaturant was used. After electrophoresis, the gels were stained with SYBR Gold (Molecular Probes, Inc.) and photographed using a BioDoc Analyze Transilluminator (Biometra, Göttingen, Germany). Bands were excised with a scalpel sterilized with ethanol and transferred to sterile Eppendorf caps. Fifty µl of water (molecular grade, Eppendorf, Germany) were added and the samples were stored at –20°C.

Clone library construction A water sample of 250 ml collected in October at station B was filtered onto a 0.2 µm Nuclepore filter (47 mm diameter). The filter was immediately frozen at –80°C until DNA- extraction. Bacterial genomic DNA of the sample was isolated after bead beating, phenol- chloroform extraction, and isopropanol precipitation as described previously (Stahl et al., 1988; MacGregor et al. , 1997), but slightly modified. Lysozyme treatment was not applied, precipitation done at –20°C and molecular grade water (Eppendorf, Hamburg, Germany) was used for resuspension at 4°C over night. PCR amplification of almost complete 16S rRNA gene fragments was performed as previously described (Brinkhoff and Muyzer, 1997)

95 Kapitel IV Diversity and abundance of Gram-positive bacteria with primers GM3F (8F) and GM4R (1492R) (Muyzer and Ramsing, 1995). Amplification was done in triplicates and the products were pooled prior to cloning. For cloning the pGEM ®-T Vector System II (Promega, Madison, USA) was used according to the manufacturer’s instructions. Sequencing of the clones was performed as described previously (Stevens et al. , 2005a). Clone sequences were checked for chimera formation with the CHECK_CHIMERA software of the Ribosomal Database Project II (Maidak et al. , 2001).

Sequencing and phylogenetic analysis PCR amplification of 16S rRNA gene fragments of bacterial isolates and subsequent sequencing was performed as described before (Brinkhoff and Muyzer, 1997). Sequences were compared with similar sequences of reference organisms by BLAST search (http://www.ncbi.nlm.nih.gov/blast (Altschul et al. , 1998). Phylogenetic analysis was performed with the ARB software package [http://www.arb-home.de (Ludwig et al. , 2004)]. For tree calculation, only sequences with more than 1300 bp were considered using maximum-likelihood analysis. Shorter sequences were added later to the final tree using the maximum parsimony option of the ARB program. Alignment positions at which less than 50% of sequences of the entire set of data had the same residues were excluded from the calculations to prevent uncertain alignments within highly variable positions of the 16S rRNA gene fragments, which cause mistakes in tree topology (Ludwig et al. , 2004). A phylogenetic analysis of Gram+ bacteria of the Wadden Sea included sequences of the Gram+ isolates obtained in May and October 1999 (this study), of a clone library and Actinobacteria -specific DGGE bands (this study), and of previous studies of the water column of the Wadden Sea using DGGE analysis or cultivation-based methods (Bruns et al. , 2003; Brinkhoff et al. , 2004; Grossart et al. , 2004; Selje et al. , 2005; Stevens et al. , 2005a). Sequences obtained from isolates and clones in this study are available from GenBank under accession no. given in Table 4.

96 Kapitel IV Diversity and abundance of Gram-positive bacteria

ACKNOWLEDGEMENTS

We are grateful to Andrea Schlingloff for the sequencing, and to H.P. Grossart for the introduction into the rolling tank incubation method. This work was supported by grants from the Volkswagen Foundation within the Lower Saxonian priority Program Marine Biotechnology and by the Deutsche Forschungsgemeinschaft within the research group BioGeoChemistry of the Wadden Sea (FG-432, TP 5).

97 Kapitel IV Diversity and abundance of Gram-positive bacteria

REFERENCES

Allgaier, M., and Grossart, H.P. (2006) Diversity and seasonal dynamics of Actinobacteria in four lakes of northeastern Germany. Appl Environ Microbiol 72 : 3489-3497. Amann R., Krumholz, L., Stahl, D.A. (1990) Fluorescent-oligonucleotide probing of whole cells for determinative, phylogenetic, and environmental studies in microbiology. J Bacteriol 172 : 762-770 Altschul, S., Madden, T., Schaffer, A., Zhang, J.H., Zhang, Z., Miller, W., and Lipman, D. (1998) Gapped BLAST and PSI-BLAST: A new generation of protein database search proGrams. Faseb Journal 12 : A1326-A1326. Bonde, G.J. (1976) Marine Bacillus. J Appl Bacteriol 41 : R7-R8. Brinkhoff, T., and Muyzer, G. (1997) Increased species diversity and extended habitat range of sulfur-oxidizing Thiomicrospira spp. Appl Environ Microbiol 63 : 3789-3796. Brinkhoff, T., Bach, G., Heidorn, T., Liang, L., Schlingloff, A., and Simon, M. (2004) Antibiotic production by a Roseobacter Clade-affiliated species from the German Wadden Sea and its antagonistic effects on indigenous isolates. Appl Environ Microbiol 70 : 2560-2565. Bruns, A., Phillip, H., Cypionka, H., and Brinkhoff, T. (2003) Aeromicrobium marinum sp. nov., an abundant pelagic bacterium isolated from the German Wadden Sea. Int J Syst Evol Microbiol 53 : 1917-1923. Bull, A.T., Stach, J.E.M., Ward, A.C., and Goodfellow, M. (2005) Marine actinobacteria: perspectives, challenges, future directions. Antonie van Leeuwenhoek 87 : 65-79. Denis, F. (1971) Bacillus in marine environment - study of 120 strains. Cr Soc Biol 165 : 2404-&. Dittmann, S. (1999) The Wadden Sea ecosystem: stability, properties and mechanisms. Springer- New York, p. 336 pp. Du, H.L., Jiao, N.Z., Hu, Y.H., and Zeng, Y.H. (2006) Diversity and distribution of pigmented heterotrophic bacteria in marine sediments. FEMS Microbiol Ecol 57 : 92-105. Eilers, H., Pernthaler, J., Glöckner, F.O., and Amann, R. (2000) Culturability and in-situ abundance of pelagic bacteria from the North Sea. Appl Environ Microbiol 66 : 3044- 3051. Elifantz, H., Malmstrom, R.R., Cottrell, M.T., and Kirchman, D.L. (2005) Assimilation of polysaccharides and glucose by major bacterial groups in the Delaware estuary. Appl Environ Microbiol 71: 7799-7805. Finne, G., and Matches, J.R. (1974) Low-temperature-growing clostridia from marine sediments. Can J Microbiol 20 : 1639-1645. Fuhrman, J.A., McCallum, K., and Davis, A.A. (1993) Phylogenetic diversity of subsurface marine microbial communities from the Atlantic and Pacific Oceans. Appl Environ Microbiol 59 : 1294-1302. Fuchs, B.M., Woebken, D., Zubkov, M.V., Burkill, P., and Amann, R. (2005) Molecular identification of picoplankton populations in contrasting waters of the Arabian Sea. Aquat Microb Ecol 39: 145-157.

98 Kapitel IV Diversity and abundance of Gram-positive bacteria

Glöckner, F.O., Zaichikov, E., Belkova, N., Denissova, L., Pernthaler, J., Pernthaler, A., and Amann, R. (2000) Comparative 16S rRNA analysis of lake bacterioplankton reveals globally distributed phylogenetic clusters including an abundant group of Actinobacteria . Appl Environ Microbiol 66 : 5053-5065. Godden, B., and Penninckx, M.J. (1984) Identification and evolution of the cellulolytic microflora present during composition of cattle manure: on the role of Actinomycetes .sp. Ann. Microbiol. (Paris) 135B : 69-78. Goodfellow, M, and Haynes, J.A. (1984) Actinomycetes in marine sediments. In: Ortiz-Ortiz, L. (ed.), Biological, biochemical and biomedical aspects of actinomycetes. Academic Press, New York, pp. 453-472. Gray, J.P., and Herwig, R.P. (1996) Phylogenetic analysis of the bacterial communities in marine sediments. Appl Environ Microbiol 62 : 4049-4059. Grossart, H.P., Schlingloff, A., Bernhard, M., Simon, M., and Brinkhoff, T. (2004) Antagonistic activity of bacteria isolated from organic aggregates of the German Wadden Sea. FEMS Microbiol Ecol 47 : 387-396. Hagström, A., Pinhassi, J., and Zweifel, U.L. (2000) Biogeographical diversity among marine bacterioplankton. Aquat Microb Ecol 21 : 231-244. Hahn, M.W., Lunsdorf, H., Wu, Q.L., Schauer, M., Höfle, M.G., Boenigk, J., and Stadler, P. (2003) Isolation of novel ultramicrobacteria classified as actinobacteria from five freshwater habitats in Europe and Asia. Appl Environ Microbiol 69 : 1442-1451. Haider, K., Trojanowski, J., and Sundmann, V. (1978) Screening for lignin degrading bacteria by means of 14 C-labelled ligninds. Arch Microbiol 119 : 103-106. Harvey, H.R., and Mannino, A. (2001) The chemical composition and cycling of particulate and macromolecular dissolved organic matter in temperate estuaries as revealed by molecular organic tracers. Org Geochem 32 : 527-542. Haukka, K., Heikkinen, E., Kairesalo, T., Karjalainen. H., and Sivonen, K. (2005) Effect of humic material on the bacterioplankton community composition in boreal lakes and mesocosms. Environ Microbiol 7: 620-630. Jendrossek, ?? (1997) Bacterial degradation of natural rubber: a privilege of actinomycetes? FEMS Microbiol Lett 150 : 179-188. Jensen, P.R., and Fenical, W. (1995) The relative abundance and seawater requirements of Gram- positive bacteria in near-shore tropical marine samples. Microb Ecol 29 : 249-257. Kirchman, D.L. (2002) The ecology of Cytophaga -Flavobacteria in aquatic environments. FEMS Microbiol Ecol 39 : 91-100. Kirchman, D.L., Dittel, A.I., Malmstrom, R.R., and Cottrell, M.T. (2005) Biogeography of major bacterial groups in the Delaware estuary. Limnol Oceanogr 50 : 1697-1706. Köpke, B., Wilms, R., Engelen, B., Cypionka, H., and Sass, H. (2005) Microbial diversity in coastal subsurface sediments: a cultivation approach using various electron acceptors and substrate gradients. Appl Environ Microbiol 71: 7819-7830. Linos, A., Berekaa, M.M., Steinbüchel, A., Kim, K.K., Sproer, C., and Kroppensted, R.M. (2002) Gordonia westfalica sp. nov., a new rubber-degrading actinomycete. Int J Syst Evol Microbiol 52 : 1133-1139.

99 Kapitel IV Diversity and abundance of Gram-positive bacteria

Llobet-Brossa, E., Rossello-Mora, R., and Amann, R. (1998) Microbial community composition of Wadden Sea sediments as revealed by fluorescence -in-situ-hybridization. Appl Environ Microbiol 64 : 2691-2696. Ludwig, W., Strunk, O., Westram, R., Richter, L., Meier, H., Yadhukumar et al. (2004) ARB: a software environment for sequence data. Nuc Acids Res 32 : 1363-1371. Lunau, M., Lemke, A., Dellwig, O., and Simon, M. (2006) Physical and biogeochemical controls of microaggregate dynamics in a tidally affected coastal ecosystem. Limnol Oceanogr 51 : 847-859. MacGregor, B.J., Moser, D.P., Alm, E.W., Nealson, K.H., and Stahl, D. (1997) Crenarchaeota in Lake Michigan sediment. Appl Environ Microbiol 63 : 1178-1181. Maidak, B.L., Cole, J.R., Lilburn, T.G., Parker, C.T., Saxman, P.R., Farris, R.J. et al. (2001) The RDP-II (Ribosomal Database Project). Nuc Acids Res 29 : 173-174. Maldonado, L.A., Stach, J.E.M., Pathom-aree, W., Ward, A.C., Bull, A.T., and Goodfellow, M (2005) Diversity of cultivable actinobacteria in geographically widespread marine sediments. Antonie van Leeuwenhoek 87 : 11-18. Marty, D.G. (1986) Description of 23 cellulolytic and non-cellulolytic marine clostridia. Ann Inst Pasteur Mic 137A : 33-43. Mincer, T.J., Jensen, P.R., Kauffman, C.A., and Fenical, W. (2002) Widespread and persistent populations of a major new marine actinomycete taxon in ocean sediments. Appl Environ Microbiol 68 : 5005-5011. Moran, M.A., Rutherford, L.T., and Hodson, R.E. (1995) Evidence for indigenous Streptomyces populations in a marine environment determined with a 16S ribosomal- RNA probe. Appl Environ Microbiol 61 : 3695-3700. Morris, R.M., Vergin, K.L., Cho, J.C., Rappé, M.S., Carlson, C.A. and Giovannoni, S.J. (2005) Temporal and spatial response of bacterioplankton lineages to annual convective overturn at the Bermudas Atlantic Time-series Study site. Limnol Oceanogr 50: 1687- 1696. Morris, R.M., Longnecker, K., and Giovannoni, S.J. (2006) Pirellula and OM43 are among the dominant lineages identified in an Oregon coast diatom bloom. Environ Microbiol 8: 1361-1370. Muyzer, G., and Ramsing, N.B. (1995) Molecular methods to study the organization of microbial communities. Water Sci Technol 32 : 1-9. Muyzer, G., de Waal, E.C., and Uitterlinden, A.G. (1993) Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl Environ Microbiol 59 : 695-700. Muyzer, G., Brinkhoff, T., Nübel, U., Santegoeds, C., Schäfer, H., and Wawer, C. (1998) Denaturing gradient gel electrophoresis (DGGE) in microbial ecology. In: Akkermans ADL, van Elsas, J.D., Bruijn, F.J. (eds.) Molecular Microbial Ecology Manual. Kluwer Academic Publishers, Dordrecht, p 1-27. Pathom-aree, W., Stach, J.E.M., Ward, A.C., Horikoshi, K., Bull, A.T., and Goodfellow, M. (2006) Diversity of actinomycetes isolated from Challenger Deep sediment (10,898 m) from the Mariana Trench. Extremophiles 10 : 181-189. Pernthaler, A., Pernthaler, J., and Amann, R. (2004) Sensitive multi-color fluorescence In situ hybridization for the identification of environmental microorganisms. Molecular Microbial Ecology Manual (2 nd Edition) 3.11: 711-726.

100 Kapitel IV Diversity and abundance of Gram-positive bacteria

Piza, H.I., Prado, P.I., and Manfio, G.P. (2004). Investigation of bacterial diversity in Brazilian tropical estuarine sediments reveals high actinobacterial diversity. Antonie van Leeuwenhoek 86: 317-328. Poremba, K., Tillmann, U., and Hesse, K.J. (1999) Tidal impact on planktonic primary and bacterial production in the German Wadden Sea. Helgol Mar Res 53 : 19-27. Pranamuda, H., Chollakup, R., and Tokiwa, Y. (1999) Degradation of polycarbonate by a polyester-degrading strain, Amycolatopsis sp. strain HT-6. Appl Environ Microbiol 65 : 4220-4222. Rajendran, N., Matsuda, O., Urushigawa, Y., and Simidu, U. (1994) Characterization of microbial community structure in the surface sediment of Osaka Bay, Japan, by phospholipid fatty- acid analysis. Appl Environ Microbiol 60 : 248-257. Rajendran, N., and Nagatomo, Y. (1999) Seasonal changes in sedimentary microbial communities of two eutrophic bays as estimated by biomarkers. Hydrobiologia 393 : 117- 125. Rappé, M., Kemp, P., and Giovannoni, S. (1997) Phylogenetic diversity of marine coastal picoplankton 16S rRNA genes cloned from the continental shelf off Cape Hatteras, North Carolina. Limnol Oceanogr 42 : 811-826. Rappé, M.S., Gordon, D.A., Vergin, K.L., and Giovannoni, S.J. (1999) Phylogeny of actinobacteria small subunit (SSU) rRNA gene clones recovered from marine bacterioplankton. Syst Appl Microbiol 22 : 106-112. Ravenschlag, K., Sahm, K., Pernthaler, J., and Amann, R. (1999) High bacterial diversity in permanently cold marine sediments. Appl Environ Microbiol 65 : 3982-3989. Reichenbach, H. (1992) The Order Cytophagales . In: Starr, M. P., H. Stolp, H. G. Trüper, A. Balows, H. G. Schlegel (eds.). The Prokaryotes. 2nd edition, Vol.1, Springer press, N. Y. pp. 3632-3666. Roller, C., Wagner, M., Amann, R., Ludwig, W., Schleifer, K.-H. (1994) In situ probing of gram-positive bacteria with high DNA G+C content using 23S rRNA-targeted oligonucleotides. Microbiology 140 : 2849-2859. Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989) Northern Hybridisation. Molecular Cloning: a laboratory manual, 2nd ed, Cold Spring Harbour Laboratory Press; New York. Sekar R., Pernthaler, A., Pernthaler, J., Warnecke, F., Posch, T., and Amann, R. (2003) An improved protocol for quantification of freshwater Actinobacteria by fluorescence in situ hybridization. Appl Environ Microbiol 69 : 2928-2935. Selje, N., Brinkhoff, T., and Simon, M. (2005) Detection of abundant bacteria in the Weser estuary by culture-dependent and culture-independent approaches. Aquat Microb Ecol 39 : 17-34. Siefert, J.L., Larios-Sanz, M., Nakamura, L.K., Slepecky, R.A., Paul, J.H., Moore, E.R.B. et al. (2000) Phylogeny of marine bacillus isolates from the Gulf of Mexico. Curr Microbiol 41 : 84-88. Stach, J. E. M., Maldonado, L. A., Ward, A. C., Goodfellow, M., and Bull, A. T. (2003) New primers specific for Actinobacteria : application to marine and terrestrial environments. Environ. Microbiol . 5:828-841. Stackebrandt, E., Rainey, F.A., and WardRainey, N.L. (1997) Proposal for a new hierarchic classification system, Actinobacteria classis nov. Int J Syst Bacteriol 47 : 479-491.

101 Kapitel IV Diversity and abundance of Gram-positive bacteria

Stahl, D.A., Flesher, B., Mansfield, H.R., and Montgomery, L. (1988) Use of phylogenetically based hybridization probes for studies of ruminal microbial ecology. Appl Environ Microbiol 54 : 1079-1084. Stepanauskas, R., Moran, M.A., Bergamaschi, B.A., and Hollibaugh, J.T. (2003) Covariance of bacterioplankton composition and environmental variables in a temperate delta system. Aquat Microb Ecol 31 : 85-98. Stevens, H., Brinkhoff, T., and Simon, M. (2005a) Composition of free-living, aggregate- and sediment surface-associated bacterial communities in the German Wadden Sea. Aquat Microb Ecoly 38 : 15-30. Stevens, H., Stübner, M., Simon, M., Brinkhoff, T. (2005b) Phylogeny of Proteobacteria and Bacteroidetes from oxic habitats of a tidal flat ecosystem. FEMS Microbiol Ecol 54 : 351- 364. Suzuki, M.T., Rappé, M.S., Haimberger, Z.W., Winfield, H., Adair, N., Strobel, J., and Giovannoni, S.J. (1997) Bacterial diversity among small-subunit rRNA gene clones and cellular isolates from the same seawater sample. Appl Environ Microbiol 63 : 983-989. Timmis, K., Hobbs, G., and Berkeley, R.C. (1974) Chitinolytic clostridia isolated from marine mud. Can J Microbiol 20 : 1284-1285. Trolldenier, G. (1993) Bestimmung physiologischer Gruppen nach der MPN-Methode. Schinner F., Öhlinger R., Kandeler E., and Margesin R. (eds.) Bodenbiologische Arbeitsmethoden, 2nd edition. Springer-Verlag, Berlin, Germany: 30-32. Uphoff, H.U., Felske, A., Fehr, W., and Wagner-Döbler, I. (2001) The microbial diversity in picoplankton enrichment cultures: a molecular screening of marine isolates. FEMS Microbiol Ecol 35 : 249-258. Urakawa, H., Kita-Tsukamoto, K., and Ohwada, K. (1999) Microbial diversity in marine sediments from Sagami Bay and Tokyo Bay, Japan, as determined by 16S rRNA gene analysis. Microbiology 145 : 3305-3315. Wallner, G., Amann, R., and Beisker, W. (1993) Optimizing fluorescent in situ hybridization with rRNA-targeted oligonucleotide probes for flow cytometric identification of microorganisms. Cytometry 14: 136–143. Warnecke, F., Amann, R., and Pernthaler, J. (2004) Actinobacterial 16S rRNA genes from freshwater habitats cluster in four distinct lineages. Environ Microbiol 6: 242-253. Wilms, R., Köpke, B., Sass, H., Chang, T.S., Cypionka, H., Engelen, B. (2006) Deep biosphere-related bacteria within the subsurface of tidal flat sediments. Environ Microbiol 8: 709-719. Zobell, C.E., and Upham., H.C. (1944) A list of marine bacteria including descriptions of sixty new species. Bull Scripps Inst Oceanogr Univ Calif 5: 239-292.

102 Kapitel IV Diversity and abundance of Gram-positive bacteria

Table 1: Strain ID, affiliation and salinity range of optimal growth of Gram positive bacteria tested. Growth was examined in a salinity range from 1 to 45‰ NaCl (w/v). For more details on the strains see Table 3 and Fig 3.

Strain Affiliation Salinity range (‰ NaCl) GWS-BW-H60M Microbacteriaceae Gr I 1-30 GWS-BW-H301M Microbacteriaceae Gr II 5-30 GWS-AG-H268 Microbacteriaceae Gr II 5-30 GWS-BW-H45M Micrococcaceae 1-20 GWS-BW-H231 Micrococcaceae Gr V 1-30 GWS-BW-H260 Micrococcaceae Gr VI 10-20 GWS-BW-H82M Mycobacteriaceae 1-30 GWS-TZ-H135 Nocardiaceae 1-20 GWS-BW-H259 Nocardioidaceae 1-20 GWS-BW-H252 Nocardioidaceae 10-30 GWS-TZ-H118 Promicromonaspora citrea 1-45 GWS-BW-H222 Bacillus Gr XIII 1-5 GWS-SE-H117 Bacillaceae Gr XI 1-45 GWS-TZ-H232 Bacillus sp. 1-30

103 Kapitel IV Diversity and abundance of Gram-positive bacteria

Table 2: Cell numbers enumerated with the CARD-FISH probes EUB338 and HGC69a at five locations in the Wadden Sea near Neuharlingersiel. For exact locations see Fig. 1.

Station Salinity EUB338 HGC69a (psu) (% DAPI cell numbers) 1 (fresh water) 0.3 56.3 + 9.3 8.0 + 1.3 2 29.1 67.7 + 7.5 4.2 + 0.9 3 31.0 48.9 + 7.4 4.7 + 1.4 4 30.8 59.1 + 10.3 5.1 + 1.4 5 30.5 62.3 + 7.7 6.5 + 2.1

104 Kapitel IV Diversity and abundance of Gram-positive bacteria

Table 2: Clones and isolates obtained from the German Wadden Sea affiliating with Gram+ bacteria as well as their closest relatives determined by BLAST analysis (http://www.ncbi.nlm.nih.gov/blast). Sequences with a similarity ≥ 99% were grouped (sequence-groups I - XV), the given information pertains the longest obtained sequence of a sequence group. Given are phylogenetic affiliation, sequence / isolate ID and sequence group where applicable, summarized data on sequence or isolate, closest relative according to BLAST analysis and similarity of the 16S rRNA gene (%), and information concerning the closest relative. Remarks on the isolates give the MPN dilution steps (Dil. [10 x]), habitats, substrates, and isolation dates. CAS= casein, CEL= cellulose; CHI= chitin, FUV= Fucus vesiculosus ; LAM= laminarin, PAL= palmitate; STA= starch, MB = Marine Broth 2216, MB* = Marine Broth 2216 prepared with natural sea water (Grossart et al ., 2004). SW = autoclaved natural seawater amended with trace elements and vitamins (Selje et al ., 2005), BW= bulk water, AG= aggregate, SE= oxic sediment; TZ= oxic/anoxic transition zone of the sediment; rt= rolling tank.

Phylum / family Sequence / isolate ID Remarks on isolate / Closest relative (acc. number) [%] Habitat or environmental (Representative of seq.- summary of sequence group features of closest x group) (Dil. [10 ], habitat, relative substrate, date) Actinobacteria Microbacteriaceae GWS-SE-H242a -2, SE, PAL, Oct 99 Microbacterium sp. OS-6 99 coastal marsh (Galicia, (AJ296094) Spain) GWS-AG-H197 -3, AG, CHI, Oct 99 Microbacterium sp. V4.BP.11 98 marine bacterioplankton (AJ244677) (Mediterranea) GWS-TZ-H305 -2, TZ, FUV, Oct 99 Microbacterium esteraromaticum 95 soil (Y17231) GWS-TZ-H139 -1, ALG, TZ, Oct 99 Microbacterium testaceum SE034 97 endophytic, agronomic (AF474327) crop GWS-BW-H60M -8, -7, -6, -1; BW, SE, TZ, Microbacterium sp. VKM 99 plant nematode (Sequence-group I) AGA, CAS, CEL, STA; Ac-2050 (AB042084) May, Oct 99 GWS-SE-H300 -5, ALG, SE, Oct 99 Microbacterium sp. LB030 99 endophytic, prairie plant (AF474326) GWS-SE-H149 -2, CEL, SE, Oct 99 Gram+ bacterium strain 12-8 99 copiotrophic, urban soil (AB008510) GWS-BWrt-H97M -1, rt; BW, AG, AGA, CAS, Marine bacterium P_wp0234 98 deep sea (Sequence-group II) MB*, STA; May, Oct 99 (AY188942) sediment/degrading PAH GWS-SE-H243 -2, SE, PAL, Oct 99 Frigoribacterium faeni (Y18807) 98 psychrophilic, non- marine Clone GWS- K13 From clone-library, BW, Actinobacterium MWH-Dar4 98 0.2 µm filtered Oct 99 (AJ565416) freshwater Sanguibacteraceae GWS-AG-H192 -3; AG, CHI, Oct 99 Cellulomonas fermentans (X79458) 94 municipal dumping site Promicromonosporacea GWS-TZ-H118 -5, TZ, AGA, Oct 99 Cellulomonas sp. IFO16243 96 no information available e (AB023364) Micrococcaceae GWS-BW-H45M -5, -4, -1; BW, ALG, CEL, Arthrobacter nicotianae SB42 97 starter culture (cheese) (Sequence-group III) MB; May 99 (AJ315492) GWS-SE-H161 -2, SE, CEL, Oct 99 Bacterium PS32 (AF200218) 99 psychrophilic, marine GWS-BW-H126 -4, -2; BW, SE; ALG, CEL, Bacterium isolate SI-12 99 agricultural soil (Sequence-group IV) MB*; Oct 99 (AJ252579) GWS-BW-H15M -7, -2, -1; BW; CEL, FUV, Micrococcus luteus strain Ballarat 99 activated sludge (Sequence-group V) MB; May, Oct 99 (AJ409096) HP42 From aggregates Micrococcus sp. V4.MO.20 98 marine bacterioplankton (AJ244665) (Mediterranea) GWS-BWrt-H158 -4, -2, rt; BW; CEL, STA; Kocuria rosea (Y11330) 99 soil and water . (Sequence-group VI) May, Oct 99 Mycobacteriaceae GWS-BW-H82M -1, BW, MB, May 99 Mycobacterium sp. IP20010961 99 water supplies (AY163341) GWS-BW-H50M -1, BW, STA, May 99 Mycobacterium sp. TH-2003 98 associated with sepsis (AY266138)

105 Kapitel IV Diversity and abundance of Gram-positive bacteria

Table 3 cont. Phylum / family Sequence / isolate ID Remarks on isolate / Closest relative (acc. number) [%] Habitat or environmental summary of sequence group features of closest (Representative of seq.- group) (Dil. [10x], habitat, relative substrate, date) Nocardiaceae GWS-BWrt-H95M rt, AGA, May 99 Rhodococcus sp. UFZ-B520 98 aquifer / degrading (AF235011) chlorobenzene GWS-TZ-H135 -4, -1; BW, TZ; ALG; Oct Rhodococcus fascians KM6 100 humus (spruce stands) (Sequence-group VII) 99 (AJ011329) GWS-SE-H175 -1, SE, CHI, Oct 99 Rhodococcus sp. MBIC01430 99 no infomation available (AB088667) GWS-TZ-H309 -6, -5, BW, TZ, FUV, Oct Rhodococcus tukisamuensis 98 depolymerizing, from soil (Sequence-group VIII) 99 (AB067734) Pseudonocardiaceae Pseudonocardiaceae -1, BW, MB, Oct 99 Pseudonocardia alni IMSNU 99 root nodules of alders bacterium T4 20049 (AJ252823) GWS-BW-H127 -2,-1; AG, BW, AGA, ALG, Pseudonocardia alni IMSNU 99 root nodules of alders (Sequence-group IX) Oct 99 20049 (AJ252823) Nocardioidaceae GWS-BW-H99 -6, -5, -4, -1; AG, BW, SE; Uncult. actinobacterium 97 aposymbiotic pea aphids (Sequence-group X) AGA, CHI, FUV, PAL, (AB074621) STA; Oct 99 GWS-BW-H259 -4, BW, STA, Oct 99 Uncult. Nocardioides sp. GCPF40 98 nutrient-limited cave (AY129808) GWS-BW-H311M -1, BW, LAM, May 99 Nocardioides sp. MWH-CaK6 99 0.2 µm filtered (AJ565419) freshwater GWS-AG-H266 -4, AG, STA, Oct 99 Nocardioides sp. V4.BE.17 97 marine bacterioplankton (AJ244657) (Mediterranea) GP-1 -4, estuary: mar., Aug 99 Nocardioides OS4 (U61298) 98 oil shale column (oxic zone) Aeromicrobium marinum -8, BW, MB, Oct 99 Aeromicrobium fastidiosum 97 herbage (Z7820) GWS-BW-H252 -2, BW, PAL, Oct 99 Nocardioides sp. NCFB3005 97 No information available (X76178) GWS-BW-H89M -4, BW, ALG, May 99 Nocardioides sp. 2.20 (AJ299233) 98 freshwater biofilm GWS-BW-H84M -4 BW, STA, May 99 Nocardioides jensenii KCTC 9134 97 soil (AF005006) uncertain actinomycetes GP-5 -7, estuary: brack., Unident. bacterium strain rJ7 97 activated sludge (0.5 g Aug 99 (AB021325) phenol) GP-6 -6, estuary: brack., Aug 99 Unident. bacterium strain rJ7 96 activated sludge (0.5 g (AB021325) phenol) Clone GWS-K46 From clone library, BW, Unident. bacterium strain rJ7 96 activated sludge (0.5 g Oct 99 (AB021325) phenol) Clone GWS-K39 From clone library, BW, Uncultured bacterium 90 Gulf of Mexico gas Oct 99 AT425_EubY10 (AY053479) hydrates Clone GWS-K11 From clone library, BW, Unidentified bacterium clone K2- 98 Hawaiian archipelago Oct 99 30-12 (AY344421) GWS-FL-8 DGGE band May-Aug 99 Uncultured actinobacterium clone 99 Marine sediment SAa03 (AY124414) Clone GWS-K72 From clone library, BW, Uncultured actinomycete clone 95 Deep sea sediment Oct 99 BD2-10 (AB015539) Clone GWS-K105 From clone library, BW, Uncultured bacterium clone E17 95 Deep sea sediment Oct 99 (AJ966591 Clone GWS-K112 From clone library, BW, Uncultured actinomycete OCS155 98 Coastal NW Pacific Oct 99 (AF001652) Clone GWS-DG1 DGGE band, Aug 05 Uncultured actinobacterium clone 99 Temperate river PRD18H10 (AY948072) Clone GWS-DG2 DGGE band, Aug 05 uncultured bacterium, clone AV9- 98 Subtropical lake 158 (AM181875) Clone GWS-DG3 DGGE band, Aug 05 Uncultured bacterium clone 98 Oregon coast NH10_01 (DQ372838)

106 Kapitel IV Diversity and abundance of Gram-positive bacteria

Table 3 cont.

Phylum / family Sequence / isolate ID Remarks on isolate / Closest relative (acc. number) [%] Habitat or environmental (Representative of seq.- summary of sequence group features of closest x group) (Dil. [10 ], habitat, relative substrate, date) Firmicutes Bacillaceae GWS-SE-H117 -7; AG, SE; AGA, MB*; "Bacillus baekryungensis " 99 seawater (Korea) (Sequence-group XI) Oct 99 (AF541965) HP 8 From aggregates, MB* Bacillus sp. KMM3737 99 seawater (Korea) (AY228462) GWS-BW-H68M -1; BW; MB*, STA; May Bacillus pumilus OM-F6 98 No information available (Sequence-group XII) 99 (AB020208) HP 10 From aggregates, MB* Bacterium KA64 (AY345445) 95 Hawaiian archipelago GWS-BW-H220M -6, -1; BW, MB; May, Oct Bacillus licheniformis Mo1 99 GTN degrading (Sequence-group XIII) 99 (AF372616) GWS-TZ-H114 -2, TZ, AGA, Oct 99 Bacillus sp. HT-1 (AF463535) 96 hamster feces Planococcaceae GP14 -5, estuary: marine, August “Planococcus psychrotolerantus ” 99 No information available (Sequence-group XIV) 99 (AF324659) GWS-TZ-H232 -1, TZ, MB, Oct 99 "Planococcus rifitiensis " M8 99 mineral water in Italy (AJ493659) GWS-SE-H236 -8, -4; SE, TZ; ALG, MB; Bacillus sp. Fa25 (AY131220) 99 strawberry plants (Sequence-group XV) Oct 99 Peptostreptococcaceae Clone GWS-K48 from clone library, BW, Oct Fusibacter paucivorans 92 oil-producing well (Clostridia ) 99 (AF050099)

Table 4: Accession numbers in GenBank of isolates and clones of this study.

Accession-no. Accession-no. Accession-no. Accession-no.

AY332093-AY332098 AY332125 AY332163 AY332202

AY332101 AY332129-AY332131 AY332164 AY332211

AY332104 AY332134 AY332170 AY332214

AY332105 AY332140 AY332173 AY332220

AY332108 AY332144 AY332183 AY332221

AY332111-AY332113 AY332146 AY332185 AY370612-AY370633

AY332118 AY332148 AY332193 EF088451-EF088496

AY332121 AY332149 AY332197

AY332122 AY332152-AY332154 AY332200

107 Kapitel IV Diversity and abundance of Gram-positive bacteria

Figure legends

Fig. 1: Locations of sample collection for the isolates (A, B), the clone library (B), and the DGGE and CARD-FISH analyses (1-5) in the Wadden Sea (lower panel). Station 1 is a fresh water drainage channel and stations 2-5 are marine. Between stations 1 and 2 is a sluice.

Fig. 2: Numbers of Gram+ strains obtained in May and October 1999 from various dilution steps of dilution cultures of bulk water samples (May and October), of suspended aggregates (October), the sediment surface (October), and the oxic-anoxic transition zone in the sediment (October).

Fig. 3: Banding patterns of an Actinobacteria -specific DGGE analysis of free-living (FL) and particle-associated bacterial communities (PA) collected at stations 1 to 5 in the vicinity of station B in the Wadden Sea. For location of station B see Fig. 1. Arrows indicate bands excised for sequencing.

Fig. 4: Maximum likelihood trees of all Gram+ isolates and clones obtained from the Wadden Sea (bold) showing the affiliation within the Actinobacteria (A) and the Firmicutes (B). Sequences <1300 bp were added with maximum parsimony. The scale bars indicate 10% sequence divergence. The Marine Actinobacteria Clade was adopted from Rappé et al. (1999) and the MAR 1 cluster from Mincer et al. ( 2002). Isolates from May are marked with an "M" at the end of the name. If available, dilution step and substrate were added to the accession number (CAS = casein, CEL = cellulose; CHI = chitin, FUV = Fucus vesiculosus ; LAM = laminarin, PAL = palmitate; STA = starch, MB = Marine Broth 2216, MB* = Marine Broth 2216 prepared with natural sea water (Grossart et al. , 2004), SW = autoclaved natural seawater amended with trace elements and vitamins (Selje et al. , 2005). The sub-habitat can be derived from the name of the May and October isolates (BW = bulk water, AG = aggregates, SE = sediment surface, TZ = oxic /anoxic transition zone of the sediment). Numbers on branches with pooled sequences indicate the number of sequences used to calculate the cluster.

108 Kapitel IV Diversity and abundance of Gram-positive bacteria

North Sea N

B A

River Weser

River Ems 10 km

5 5 33 44

2 2

1 1 50 m

FIG. 1. Stevens et al.

109 Kapitel IV Diversity and abundance of Gram-positive bacteria

-9

-8 -7

-6

-5

-4

-3

-2 Dilution step (log 10) -1

0 0 5 10 15 Number of isolates

FIG. 2. Stevens et al.

1 3 5 2 4

PA FL PA FL PA FL PA FL PA FL

3

1 2

FIG. 3. Stevens et al.

110 Kapitel IV Diversity and abundance of Gram-positive bacteria

FIG. 4A, Stevens et al.

111 Kapitel IV Diversity and abundance of Gram-positive bacteria

FIG. 4A cont., Stevens et al.

112 Kapitel IV Diversity and abundance of Gram-positive bacteria

FIG. 4B, Stevens et al.

113

V. High regional variability of bacterial communities in the German Bight, North Sea

114 Kapitel V Regional variability of bacterial communities in the German Bight

High regional variability of bacterial communities

in the German Bight, North Sea

Beate Rink, Thorsten Brinkhoff, Katja Ziegelmüller, Meinhard Simon *

Institute for Chemistry and Biology of the Marine Environment (ICBM),

University of Oldenburg, D-26111 Oldenburg, Germany

Running title: Bacteria in the North Sea

Key words: North Sea, free-living and attached bacteria, Roseobacter , DGGE, phytoplankton

______

* Corresponding author. Institute for Chemistry and Biology of the Marine Environment (ICBM), University of Oldenburg, PO Box 2503, D-26111 Oldenburg, Germany, Phone: +49-441-798-5361. Fax: +49-441-798-3438. E-mail: [email protected]

115 Kapitel V Regional variability of bacterial communities in the German Bight

ABSTRACT

The German Bight of the North Sea is characterized by near shore tidal flat regions with high loads of suspended matter and estuarine inputs of organic matter and pelagic off shore regions. Due to tidal and wind-induced currents its hydrography is highly dynamic. In order to examine how these highly dynamic properties affect the regional distribution and composition of the bacterioplankton we conducted two surveys in June in two consecutive years during which we assessed the composition of the free-living (FL, 0.2-5.0 µm fraction) and particle- associated (PA, >5.0 µm fraction) bacterial communities on the background of hydrographic (salinity, temperature) and biogeochemical properties (suspended matter, particulate organic carbon, chlorophyll, phytoplankton composition). The composition of the bacterial communities was determined by denaturing gradient gel electrophoretic (DGGE) analysis of 16S rRNA gene fragments PCR-amplified by Bacteria -, α-Proteobacteria and Bacteroidetes - specific primer sets and subsequent sequencing of excised bands. The results showed that the FL-bacterial community was rather evenly distributed in the German Bight irrespective of the regional hydrographic and biogeochemical differences. Several prominent bands, identified as phylotypes affiliated to the Roseobacter clade of α- Proteobacteria , persisted throughout all 10 stations visited. The PA bacterial community exhibited distinct differences among the various stations. These differences were not simply attributed to properties of the near shore tidal flat regions and to the more homogeneous hydrographic situation of the off shore region. They were rather site-specific, obviously reflecting local conditions of the phytoplankton present and its growth phase and the resuspended particles in the tidal flat regions. The results of the PA bacterial community showed that unspecific PCR-amplifications were obtained by the Bacteria – (chloroplasts) and α-Proteobacteria –specific primer sets ( δ-Proteobacteria ), biasing the results to a certain extent. Because one primer applied for amplifying α-proteobacterial 16S rRNA gene fragments (ALF968) is frequently used as a probe in fluorescence in situ hybridization (FISH) analyses, its application leads to overestimates of α-Proteobacteria in samples containing δ- Proteobacteria .

116 Kapitel V Regional variability of bacterial communities in the German Bight

INTRODUCTION

In aquatic environments, complex communities of free-living and particle-associated heterotrophic bacteria are the main decomposers of dissolved (DOC) and particulate organic carbon (POC) and play a key role in the global carbon cycle (Cotner & Biddanda 2002). It has been shown that origin and composition of organic matter affect the composition and biomass production of bacterial communities (Covert & Moran 2001, Crump et al. 2003, Lebaron et al. 1999) and that phylogenetic bacterioplankton groups exhibit distinct preferences for low and high molecular weight carbon sources (Cottrell & Kirchman 2000). Marine ecosystems are structured into the open sea and coastal environments including the polyhaline estuarine regions, and the environmental conditions in these habitats comprise different physical and biogeochemical properties which may affect the ambient microbial communities. In the coastal regions of the southern North Sea, the German Bight, high loads of dissolved and particulate inorganic and organic matter are introduced from the tidal flats and the rivers Weser and Elbe, thus providing organic and inorganic nutrients as well as refractory organic matter (Loewe et al. 2005). An easterly current follows the southern coastal line along the East Frisian Islands and the Weser estuary and encounters the polyhaline plume of the Elbe estuary. Both water masses circulate along the north Frisian coast, often building separate layers of distinct salinities or, in the case of persisting strong winds, a salinity gradient. Because of the shallow water depth (10 – 40 m), the variable river discharge and because of often variable winds the extent and position of the different water masses may change considerably even on a short-term scale. Various aspects of the North Sea bacterioplankton have been the scope of quite a few investigations, e.g. the culturability of pelagic bacteria (Eilers et al. 2001), seasonal and interannual dynamics and abundance of specific phylogenetic groups (Eilers et al. 2000, Gerdts et al. 2004) and the composition of bacterial communities as a function of bacterial respiration and growth (Reinthaler et al. 2005). The spatial distribution of the bacterial community composition, e.g. in and off shore gradients including the tidal flat areas and pelagic regions has not been considered. It is not known how the diversity of the bacterial communities varies within and among the various water bodies mentioned above, possibly as a function of a patchy distribution of phytoplankton blooms and the strong tidal currents, or whether the community composition remains unaffected. The strong currents within the German Bight and the generally shallow water depth may also lead to a well mixed situation, preventing the establishment of pronounced regional differences of the bacterial community

117 Kapitel V Regional variability of bacterial communities in the German Bight composition. Obtaining insight into such regional distributions of the composition of bacterial communities is also important for designing sampling strategies for future investigations, linking the community composition to hydrographical and biogeochemical processes, not only in the German Bight, but also in other shallow coastal marine regions exhibiting strong currents, and estuarine and terrestrial inputs of dissolved and particulate matter. We investigated the bacterial communities at various near shore and off shore stations in the German Bight together with properties to characterize the suspended particulate matter (SPM) and phytoplankton. We used denaturing gradient gel electrophoresis (DGGE) of 16S rRNA gene fragments and applied primer sets specific for Bacteria , α-Proteobacteria and the Bacteroidetes phylum.

MATERIALS AND METHODS

Study area and sampling. Surface water samples were collected at various locations in the German Bight from 11 to 13 June 2002 and from 24 to 27 June 2003 (Fig. 1, Table 1) on board RV Heincke with a 10 L Niskin bottle. For analysis of suspended matter (SPM) dry weight (DW), particulate organic carbon (POC), and chlorophyll a (Chl a) 500 to 1000 ml of sample water were filtered in duplicates on precombusted and preweighed glass fiber filters (GF/F, Whatman) and stored at –20°C in the dark until further processing in the lab within four weeks. For enumeration of bacteria 100 ml of seawater were fixed with formaldehyde (2% v/v) and stored at 4°C until further processing within four weeks. Phytoplankton cells were fixed with Lugol’s solution as described elsewhere (Utermöhl 1958). For DGGE analysis, 250 ml of sample water were fractionated by filtration on polycarbonate-filters (Nuclepore) with pore sizes of 5.0 µm (particle-associated bacteria) and subsequently of 0.2 µm (free-living bacteria) and stored at –20°C until further processing within four months. Temperature and salinity were recorded by a built-in probe of RV Heincke. Enumeration of bacteria and algae. Bacteria were enumerated by epifluorescence microscopy after staining with DAPI (4´-6-diamidino-2-phenylindole) on black 0.2 µm Nuclepore filters at 1000x magnification (Porter & Feig 1980). We did not differentiate between free-living (FL) and particle-associated (PA) cells, mainly because it was rather difficult to reliably enumerate particle-associated cells in the near shore stations with high concentrations of SPM. When numbers were assessed the reliable desorption technique of PA bacteria by Lunau et al. (2005) was not yet available. Phytoplankton cells were counted by

118 Kapitel V Regional variability of bacterial communities in the German Bight inverted microscopy (Utermöhl 1958) and phytoplankton species were identified according to Drebes (1974). Phytoplankton pigments, SPM and POC. For chlorophyll analysis filters were extracted at 75°C in 90% ethanol and concentrations of Chl a were determined by standard procedures (Parsons et al. 1984). For phaeopigment determination, samples were acidified with HCl (2N) prior to spectrophotometric analysis. For determination of DW, filters were dried for 1 hour at 110°C and weighed on a micro-balance (Sartorius, Germany). In 2002, DW was corrected for salt according to Lunau et al. (2006) and in 2003 filters were rinsed with distilled H 2O. POC was determined with a FlashEA 1112 CHN-analyzer (Thermo Finnigan). Nucleic acid extraction, primer sets and PCR amplification of 16S rRNA gene fragments. Genomic DNA was extracted with phenol-chloroform as described in Rink et al. (2006a) with slight modifications. DNA was precipitated at –20°C overnight using isopropanol and resuspended in molecular grade water. Samples were stored at –20°C until further processing. For the amplification of 16S rRNA gene fragments, primer sets were used targeting eubacterial DNA (primer pair GC 341F, Muyzer et al. 1993; 907RM, Muyzer et al. 1998), the Bacteroidetes phylum (primer pair GC CF319aF, Jaspers et al. 2001, and 907RM) and α-Proteobacteria (primer pair GC 341F and ALF 968R, Rink et al. 2006a). Specificity of the primer sets and the applied PCR conditions are described by Rink et al. (2006a). Amplification products were analyzed by electrophoresis in 1.5% (w/v) agarose gels and stained with ethidium bromide (1 µg ml -1) (Sambrook et al. 1989). For subsequent sequence analysis PCR products were purified by using the Qiaquick PCR purification kit (Qiagen Inc., Chatsworth, California). DGGE analysis of PCR products and cluster analysis. DGGE was performed with an INGENYphorU system (Ingeny International BV, Leiden, The Netherlands) following the protocol of Rink et al. (2006a). After electrophoresis, the gels were stained with SYBR Gold (Molecular Probes, Inc.) and documented using a BioDoc Analyze Transilluminator (Biometra, Göttingen, Germany). Bands were excised, suspended in 50 µl of water (molecular grade, Eppendorf, Germany) and centrifuged for 2 min. at 3,000 rpm. Samples were stored at –20°C and 1 µl was used as template in subsequent PCR reactions. A cluster analysis of the DGGE banding patterns was performed using the software GelCompare II, Version 2.5 (Applied Maths, St. Martens-Latem, Belgium). We applied 5 to 20% background subtraction depending on the signal-to-noise ratio of the corresponding gel. Patterns were compared curve-based using Pearson correlation as similarity coefficient and UPGMA (unpaired group

119 Kapitel V Regional variability of bacterial communities in the German Bight method of analysis) to generate the dendrogram. We used the position tolerance optimization option of the software to fit the curves to the best possible matching. Cloning. 16 DGGE bands were cloned using the pGEM ®-T Vector System II (Promega, Madison, USA) following the instruction manual. At least five clones per DGGE band with inserts were picked after blue-white-screening and amplified with the specific DGGE primers. Fragment length of the inserts was screened by agarose gel electrophoresis and positive inserts were examined for their specific height using DGGE. Adequate clones were amplified for sequencing using the primers pUC/M13f and pUC/M13r (Sambrook et al. 1989). Sequencing and phylogenetic analysis. PCR products were sequenced with an Automated DNA Sequencer (Model 4200, LI-COR Inc.) using the primers 341F and 907RM, labeled with IRDye TM 800, and the DYEnamic Direct cycle sequencing kit (Amersham Life Science Inc.). Clones were sequenced by Geneart (Regensburg, Germany) using the primer M13f. At least 400 bp were determined for all sequences and the phylogenetic affiliation was compared to those in GenBank using the BLAST function of the NCBI server (http://www.ncbi.nlm.nih.gov). The phylogenetic trees were constructed using the ARB software package (http://www.arb-home.de, Ludwig et al. 2004). The backbone tree was calculated with the maximum likelihood method using sequences with a minimum of 1300 bp length including type strains of the selected phylogenetic groups. To avoid uncertain alignments, positions were excluded at which less than 50% of all sequences showed the same residues. Sequences with less than 1300 bp were added to the backbone tree with the maximum parsimony method using the same filter. 16S rRNA gene sequences of seven type strains belonging to Cyanobacteria were used as outgroup. Nucleotide sequence accession number. The sequences obtained in this study are available from GenBank under accession no. DQ911759 to DQ911821.

120 Kapitel V Regional variability of bacterial communities in the German Bight

RESULTS

Hydrography. In both years, the off shore stations with depths >17 m (sta 1, 8, 9, 10) were characterized by lower water temperatures and higher salinities than the near shore stations (sta 2-7, Fig. 2). At the latter stations water depths is only 8-11 m except at station 2 which exhibits more off shore than near shore properties (see below). In general, temperature and salinity were higher in 2003 as compared to 2002, when the survey was carried out two weeks earlier. In 2002, surface water temperatures ranged from 13.7°C (sta 1, 8) to 16.5°C (sta 7) and in 2003 from 14.4°C (sta 10) to 17.6°C (sta 4). Salinity ranged from 29.6 (sta 7) to 33.2 psu (sta 9) in 2002 and from 31.8 (sta 6) to 35.5 psu (sta 9) in 2003. Abundance and composition of SPM. In both years, SPM concentrations were higher at the near shore than at the off shore stations (Fig. 3A, 3B). In 2002, SPM ranged from 4 to 5 mg DW l -1 at the off shore stations and station 2 to 7.6 to 16 mg DW l -1 at the near shore stations with the highest value at station 5. In 2003, respective values at the off shore stations (sta 1, 2, 8, 9, 10) were between 2 and 4 mg DW l -1 and at the near shore stations 6.5 to 15.4 mg DW l -1. POC varied from 0.10 mg l -1 at station 9 to 0.78 mg l -1 at station 6 with generally higher values at the near shore than at the off shore stations (Fig. 3B). At stations 9, 10 and 2 POC constituted around 10% of DW but at the near shore stations generally <6%. Phytoplankton. Chl a concentrations in 2002 were lower than in 2003 and varied from 0.8 to 4.5 µg l -1 (Fig. 3C, 3D). In 2002, concentrations at the near shore stations were higher than at the offshore stations except at station 8. In 2003, highest concentrations occurred at stations 3 and 8 with up to 6.3 µg Chl a l-1 without any clear-cut difference between the near and off shore stations. Patterns of phaeopigment concentrations, which are only available for 2003, were quite different from that of Chl a (Fig. 3D), indicating that the growth phase and thus the physiological status of the phytoplankton was quite variable among the stations. Highest proportions of phaeopgments, close to or exceeding those of Chl a, occurred at stations 6, 7 and 9. In 2002, Leptocylindricus danicus dominated the phytoplankton at the off shore stations and Rhizosolenia imbricata at the near shore stations (data not shown). At the North Frisian coast (sta 6, 7), Guinardia delicatula and Guinardia flaccida were also present to substantial proportions. In 2003, the phytoplankton at all stations was highly dominated by Rhizosolenia imbricata except at station 10 which showed a more diverse composition including substantial proportions of Guinardia spp. (Fig. 3E). Also at station 8, Guinardia spp. were the second most abundant taxa. Phytoplankton cell numbers generally reflected concentrations of Chl a with highest values at stations 3 and 8. At station 6 and 7 the high

121 Kapitel V Regional variability of bacterial communities in the German Bight sediment load in the samples and low phytoplankton cell numbers prevented a reliable enumeration. Bacterial abundance. In 2002, bacterial abundance varied between 1.8 x 10 6cells ml -1 at station 1 and 3.5 x 10 6cells ml -1 at station 7. Numbers did not covary with Chl a or SPM. In 2003, the off shore stations exhibited low bacterial numbers and highest numbers were recorded at stations 3 and 6 together with high concentrations of Chl a and SPM. DGGE and cluster analysis. Pronounced differences were detected between DGGE banding patterns of FL and PA bacterial communities. These differences were substantiated by a cluster analysis (Fig. 4). The application of the Bacteria -specific primer set yielded 7 to 13 DGGE bands of the FL bacterial community in 2002 and 15 to 24 bands in 2003. In the former year, the lowest and highest number of bands occurred at stations 2 and 1 and in the latter year at stations 1 and 6. Two prominent bands persisted throughout all stations in both years, identified as clones GB02-e8-FL, GB02-e9-FL, GB03-e16-FL and GB03-e17-FL (Table 2). Other bands occurred only at a few or individual stations such as at stations 1 and 10 (GB03-e15-FL), 6 and 7 (GB03-e24-FL, GB03-e25-FL), and 5 (GB03-e23-FL). There was no band detected only at the off shore or near shore stations. In 2003, stations 6 and 7, however, exhibited distinctly different patterns than the other stations. In the PA bacterial community in both years, the number of bands exceeded that of the FL bacterial community. In 2002, between 10 and 17 bands were detected with lowest and highest numbers at stations 10 and 1, respectively. In 2003, band numbers ranged between 12 and 28 with lowest and highest numbers at stations 1 and 6. The variability of the PA bacterial community among the different stations was more pronounced than that of the FL bacterial community. No single band was detected at all stations. The banding patterns of the PA bacterial community were highly biased by chloroplast-derived 16S rRNA gene fragments. Two of the 7 bands of the 2002 samples and 8 of the 14 bands of the 2003 samples sequenced turned out as chloroplast-like phylotypes. The cluster analysis substantiated the different banding patterns of the FL- and PA bacterial communities detected by the Bacteria -specific primer set (Fig. 4). The banding patterns of the PA bacterial communities exhibited a lower similarity (>55% Pearson correlation) than those of the FL bacterial communities (>76% Pearson correlation). Even though micro-clusters occurred, near shore and off shore stations did not exhibit distinct clusters. Stations 6 and 7, however, formed a separate cluster in both bacterial communities in 2003.

122 Kapitel V Regional variability of bacterial communities in the German Bight

The DGGE analysis of the FL- and PA associated bacterial communities applying the Bacteroidetes -specific primer set was only done in 2003. The results also revealed pronounced differences between both communities with a higher diversity in the PA bacterial community. In the FL bacterial community between 7 (sta 8) and 13 bands (sta 1) were detected, and in the PA bacterial community between 9 (sta 7) and 17 bands (sta 5). There was no band which was detected at all stations, neither in the FL- nor in the PA bacterial community. However, several bands occurred at distinct stations, such as at stations 1-5 and 8 (GB03-c5-PA, Table 3), at stations 2-5 (GB03-c8-FL, Table 2), and at stations 1, 8 and 10 (GB03-c12-FL). In the PA bacterial community, stations 2-5 and 10 clustered together, as did stations 6 and 7 (Fig. 4). The off shore stations 1, 8 and 9 branched deeply separated. In the FL fraction, no specific sub-clusters were detected. The α-Proteobacteria –specific DGGE banding patterns revealed lowest band numbers of all target groups with 5 to 7 bands in the FL bacterial community and 7 to 16 bands in the PA bacterial community. In the former community, two conspicuous bands (GB-a14-FL, GB- a15-FL, Table 2) dominated the banding patterns at all stations and the similarity between the banding patterns was very high, as confirmed by the cluster analysis (Fig. 4).The banding patterns of the PA bacterial community were much more diverse. Only stations 6 and 7 clustered together. Phylogenetic affiliation. Sequencing of the excised DGGE bands obtained from the Bacteria -specific amplicons showed that 12 bands of a total of 38 contained chloroplast- derived 16S rRNA gene fragments. Ten of them were detected in the PA fraction. In addition, 13 sequences affiliated to α-, 1 to β, 2 to γ-Proteobacteria and 10 to the Bacteroidetes - phylum. All sequences of α-Proteobacteria obtained from the Bacteria -specific DGGE gels affiliated to the Roseobacter clade except GB02-e3-PA (station 4), which was related most closely to Acidiphilium aminolytica (Tables 2, 3, Fig. 5A). The phylotypes occurring at all stations in both years, GB02-e8-FL, GB03-e16-FL, GB02-e9-FL and GB03-e17-FL, affiliated to the NAC11-7 cluster detected in the North Atlantic (Gonzalez et al. 2000) and the WM11- 36 cluster identified in the polyhaline section of the Weser estuary (Selje & Simon 2003). The two clones which affiliated to γ-Proteobacteria (GB03-e7-PA, GB03-e10-PA) were detected in 2003 in the PA bacterial community of station 6 and were closely related to clones from the East Frisian Wadden Sea (GWS-AG-6, GWS-SE-4, Fig. 5C). Sequences of DGGE bands obtained with the Bacteroidetes specific primer set revealed 2 unspecific amplifications. DGGE band GB03-c1-PA (sta 1, PA fraction) was related to a

123 Kapitel V Regional variability of bacterial communities in the German Bight

Firmicutes species (Table 3) and band GB03-c6-PA to chloroplast-derived 16S rRNA genes. The sequences of two DGGE bands obtained from the PA fraction of station 6 (North Frisian coast, GB03-c3-PA and GB03-c4-PA) were closely related to 16S rRNA gene fragments detected as a diatom-associated bacterium and in coastal bacterioplankton (GWS-AG-8, GWS-c2-FL, Fig. 5D), respectively. Clone GB03-c2-PA (sta 6) was closely related to strain T15, isolated from a high dilution step of a dilution culture from the East Frisian Wadden Sea (Brinkhoff et al. 2004). Sequencing of 16S rRNA gene fragments obtained from the α-Proteobacteria specific DGGE gel showed unspecific amplification in the PA bacterial community. Nine of the 13 sequences obtained were identified as δ-Proteobacteria predominantly originating from sediments and only 4 as α-Proteobacteria (Fig. 5A and B, Table 3). Two of them affiliated to the Roseobacter clade, to the RCA cluster (GB03-a5-PA) and to Sulfitobacter pontiacus (GB- a4-PA), and the other two to Rhodobacterales (GB-a7-PA) and to the genus Sphingomonas (GB03-a1-PA), respectively. The 4 sequenced bands of the FL bacterial community of the α- Proteobacteria specific DGGE gel affiliated to the clusters RCA (GB03-a14-FL) and WM11- 36 (GB03-a15-FL, GB03-a16-FL) of the Roseobacter clade and one closely to a phylotype retrieved from the German Wadden Sea and related to Acidiphilum aminolytica . (GB03-a17- FL).

DISCUSSION

We found a surprisingly high variability of the DGGE banding patterns both of FL and PA bacterial communities of either primer set applied in the German Bight of the North Sea which, however, only partially reflected the clear differences of salinity and SPM concentrations between the near shore and off shore stations. Similarly, banding patterns did not reflect patterns in the distribution of Chl a or phytoplankton composition. We did find distinct differences between banding patterns of PA and FL-bacterial communities, substantiated by the cluster analysis. A number of phylotypes in both communities only occurred at certain stations near shore or off shore, indicating that hydrographic and biogeochemical differences did affect the composition of the bacterial communities to a certain extent. The only stations which exhibited clearly different banding patterns and formed a distinct subcluster in 2003 were stations 6 and 7 close to the North Frisian coast. When samples were collected at these shallow stations rather strong wind (Beaufort scale 6-7) prevailed leading to high sediment resuspension. The unspecific detection of sediment-

124 Kapitel V Regional variability of bacterial communities in the German Bight associated δ-Proteobacteria in these samples is a further indication of the sediment resuspension. Several prominent DGGE bands of the FL bacterial communities amplified with the Bacteria - and α-Proteobacteria -specific primer sets, however, were present at all stations, indicating that several populations of this community persisted in the German Bight, irrespective of the given hydrographic and biogeochemical conditions. The variability within the PA bacterial community among the various stations was greater than that of the FL bacterial community, indicating that site-specific properties affected more the former than the FL bacterial community. However, unspecific amplification of chloroplast-derived 16S rRNA gene fragments by the Bacteria -specific primer set and of δ-proteobacterial 16S rRNA gene fragments by the α-Proteobacteria -specific primer set contributed to this variability and biased the banding patterns and cluster analysis of the PA bacterial community. Our investigation was carried out during several days in June of two consecutive years, thus covering only a short period of the annual development of the German Bight. There are consistent reports of general annual patterns of the hydrography and biological development at individual stations such as at Helgoland Roads, at Norderney (East Frisian Wadden Sea) and Büsum and Sylt (North Frisian Wadden Sea) (BSH 2002 and 2003, Loewe et al. 2005). Even though they exhibit a general seasonal trend, the patterns of each station do vary. Further, the individual stations show short-term deviations from the seasonal trends, also for June 2002 and 2003, presumably because of wind- and current-induced movements of water masses with different physico-chemical and biological properties. Our surveys in both years encountered two different biological situations, as shown by the different Chl a concentrations and composition of the phytoplankton, and also reflected by the generally higher number of DGGE bands detected both in the FL and the PA bacterial community in 2003. The physiological state of the phytoplankton among the near shore and off shore station varied, as indicated by the variable proportions of phaeopigments relative to total chlorophyll. Hence it seems not surprising that we did not find consistent patterns of the composition of the bacterial communities in the near and off shore regions in a rather dynamic regional coastal sea. As mentioned, our DGGE and cluster analysis was biased by unspecific PCR amplifications. Unspecific amplifications by the Bacteria -specific primer set applied has been reported previously (e.g. Selje & Simon 2003, Stevens et al. 2005a) and amplification of chloroplast-derived 16S rRNA gene fragments in the PA bacterial community with this primer set appears a general problem for samples containing phytoplankton cells. Unspecific amplification with primer ALF968r ( α-Proteobacteria ) has also been reported by Rink et al.

125 Kapitel V Regional variability of bacterial communities in the German Bight

(2006a) for samples from the East Frisian Wadden Sea and by Overmann et al. (2005). The latter authors found amplification of Actinobacteria in freshwater samples and possible detection of few γ-Proteobacteria . Unspecific amplification of δ-Proteobacteria by the ALF968r primer predominantly occurred in the PA bacterial community at the shallow near shore stations (3-7) with high concentrations of resuspended SPM. Marine sediments contain high numbers of sulfate reducing bacteria affiliated to δ-Proteobacteria (Llobet-Brossa et al. 1998, Musat et al. 2006) and also myxobacteria which affiliate to this subclass of Proteobacteria as well (Stevens et al. 2005a). Hence, when these bacteria are present in the samples to detectable amounts, their 16S rRNA genes are amplified by this primer. In the FL bacterial community, no unspecific amplification occurred, indicating that this community did not include δ-Proteobacteria in proportions high enough to be amplified. In conclusion, a reliable assessment of the composition of PA α-Proteobacteria by the subclass-specific primer set was not possible. Our finding of unspecific amplification of the ALF968r primer has important implications for the interpretation of data obtained by fluorescence in situ hybridization (FISH) applying this oligonucleotide as a probe (ALF968). In habitats with high proportions of δ-Proteobacteria they may be included in the detection of α-Proteobacteria and thus lead to overestimating this subclass. Also the Bacteroidetes –specific primer set (Jaspers et al. 2001) resulted in unspecific amplification in three of 12 cases. Even though this is also a bias of the DGGE results, it appears not as critical as that with the α-Proteobacteria –specific primer set, but emphasizes the importance for sequencing of prominent bands applying group-specific primer sets in DGGE analyses. These biases of unspecific amplification predominantly affected the DGGE banding pattern of the PA bacterial community obtained by the Bacteria –specific (Fig. 4B) and α- Proteobacteria –specific primer sets in 2003 (Fig. 4D). The other banding patterns were only marginally affected. Hence our general findings of greater differences of the composition of the PA bacterial community among the various stations and a greater diversity as compared to the FL bacterial community is unaffected by these biases. Our findings are consistent with previous reports from single stations of the same region (Gerdts et al. 2004, Stevens et al. 2005a). However, similarly as these two reports, we did not enumerate the PA bacteria and thus can not directly assess their significance relative to that of the FL bacteria. Particle- associated bacteria have been enumerated in the SPM-rich East Frisian Wadden Sea by Lunau et al. (2006). In June and July of various years, numbers of PA bacteria vary from 23 to 32% of total bacteria, and we assume that these numbers are representative for the near shore

126 Kapitel V Regional variability of bacterial communities in the German Bight stations in general. Comparable numbers are not available for the off shore region of the German Bight. This region is characterized by low numbers of aggregates (Riebesell 1991), and PA bacteria in such neritic seas usually constitute <10% of total bacterial biomass and activity (Simon et al. 2002). Therefore, we assume that PA bacteria are of much lower significance in the off shore than in the near shore region of the German Bight. The phylogenetic lineages we identified have previously been shown to constitute the bacterial communities in the German Bight to a great extent. Applying FISH, Eilers et al. (2000 and 2001) found that at Helgoland Roads Bacteroidetes and α- and γ-Proteobacteria constitute 18–30, 15–25 and 6–9% of DAPI-stainable bacteria, respectively, at various seasonal situations including June. These groups also constitute the bacterial communities in the Wadden Sea to a great extent, with rather equal proportions in the FL (each group 10– 20% of DAPI stainable bacteria) and PA fractions (each group 15–40%) (Rink et al. 2006b). Our results, in agreement with previous studies (Eilers et al. 2000, Selje et al. 2004, Stevens et al. 2005a, Zubkov et al. 2002), show that members of the Roseobacter clade and in particular of distinct subclusters (RCA, WM11-36, NAC11-7) are important components of the FL bacterial community (Fig. 5A). As indicated by the prominent DGGE bands persisting at all stations (GB02-e8-FL, GB02-e9-FL, GB-a14-FL, GB-a15-FL, GB03-e16-FL, GB03-e17-FL) and the high similarity of the banding patterns, the FL α-Proteobacteria sub-community was most evenly distributed in the German Bight and least affected by hydrographic and biogeochemical differences of the various regions. The occurrence of closely related phylotypes in other neritic and oceanic regions (Gonzalez et al. 2000, Selje et al. 2004) emphasizes that bacteria of this lineage are important components of the marine bacterioplankton globally. The FL sub–community of the Bacteroidetes phylum exhibited a greater variability and diversity than that of α-Proteobacteria , as shown by the group-specific DGGE banding patterns and cluster analysis and the occurrence of more phylotypes at distinct stations and of none at all stations. The phylotypes were retrieved from the bands amplified by both the Bacteria -specific and Bacteroidetes –specific primer sets. This notion indicates the significance of site-specific factors for controlling the composition of this sub-community. The phylotypes affiliated to rather different clusters of the Bacteroidetes phylum but in most cases were closely related to other phylotypes previously detected in the Wadden Sea, the German Bight or North Sea during phytoplankton blooms (Table 2, Fig. 5D, Rink et al. 2006a, Stevens et al. 2005a and 2005b, Zubkov et al. 2002, J. Pernthaler et al. unpubl.). Close relationships between phylotypes and isolates of the Bacteroidetes phylum and distinct

127 Kapitel V Regional variability of bacterial communities in the German Bight phytoplankton and algal populations have been reported (Schäfer et al. 2002, Grossart et al. 2005, Rooney-Varga et al. 2005). Further, bacteria of this phylum are known to degrade complex organic polymers (Cottrell & Kirchman 2000, Kirchman 2002). Therefore, it appears that the phylotypes affiliated to Bacteroidetes reflect more the specific substrate conditions related to the phytoplankton community of different growth stages and composition and to other substrate sources at the various stations. Particles and suspended aggregates provide much more diverse micro-habitats than the surrounding water. Depending on the given conditions and the water depth, they may include phytoplankton-derived organic matter as well as organic and inorganic matter resuspended from the sediment. Therefore, it appears not surprising that the PA bacterial community was more diverse than the FL bacterial community and contained more site-specific phylotypes reflecting the local environmental conditions. The PA bacterial community contained quite a few phylotypes affiliated to various clusters of the Bacteroidetes phylum (Fig. 5D, Table 3), but also various phylotypes affiliated to α- and γ-Proteobacteria and distinct from those in the FL bacterial community. The PA bacterial phylotypes affiliated to α-Proteobacteria were related mainly to various clusters of the Roseobacter clade which differ from those containing the FL bacterial phylotypes. Others affiliated to Sphingomonas and Acidiphilum aminolyticum (Fig. 5A, Table 2). The two γ-Proteobacteria phylotypes were detected in 2003 at stations 6 and 7 in the North Frisian Wadden Sea and closely related to phylotypes retrieved from the East Frisian Wadden Sea (Fig. 5B, Stevens et al. 2005b). One phylotype fell into the SAMMIC-cluster (Surface Attached Marine MICrobes, Stevens et al. 2005b). Members of this cluster are uncultured, globally distributed and always associated to suspended aggregates in coastal systems or to sediments. The unspecific amplification of the tentatively α- Proteobacteria –specific primer set also retrieved exclusively phylotypes of sediment- associated δ-Proteobacteria (Fig. 5B, Table 3). In conclusion, our results show a variable diversity and distribution patterns of the bacterial community in the German Bight of the North Sea. The FL bacterial community is rather similar in this coastal sea and harbors several widely distributed members, affiliated to the Roseobacter clade of α-Proteobacteria and obviously little affected by site-specific environmental conditions. However, other members occurred only at specific locations, obviously as the result of site-specific environmental conditions of the tidal flat areas and the phytoplankton communities. The PA bacterial community was more divers and reflected mainly local environmental features, also related to the specific environmental conditions of

128 Kapitel V Regional variability of bacterial communities in the German Bight the tidal flat areas with high SPM concentrations and intense resuspension and of the phytoplankton communities.

ACKNOWLEDGEMENTS We appreciate the hospitality and cooperation of the captain and crew of RV Heincke . We thank B. Kuerzel for dry weight and chlorophyll analyses and A. Luek for phytoplankton and bacterial cell counts. This work was supported by the Deutsche Forschungsgemeinschaft (DFG) within the research group “BioGeoChemistry of the Wadden Sea” (FG 432 TP5).

129 Kapitel V Regional variability of bacterial communities in the German Bight

REFERENCES

Brinkhoff T, Bach G, Heidorn T, Liang L, Schlingloff A, Simon M (2004) Antibiotic production by a Roseobacter clade affiliated species from the German Wadden Sea and its antagonistic effects on indigenous isolates. Appl Environ Microbiol 70:2560-2565 BSH (2002) MURSYS Reportsystem 2002. Bundesamt für Seeschifffahrt und Hydrographie, Hamburg (www.bsh.de) BSH (2003) MURSYS Reportsystem 2003. Bundesamt für Seeschifffahrt und Hydrographie, Hamburg (www.bsh.de) Cotner JB, Biddanda BA (2002) Small players, large role: Microbial influence on biogeochemical processes in pelagic aquatic ecosystems. Ecosystems 5:105-121 Cottrell MT, Kirchman DL (2000) Natural assemblages of marine Proteobacteria and members of the Cytophaga-Flavobacteria cluster consuming low- and high-molecular weight dissolved organic matter. Appl Environ Microbiol 66:1692-1697 Covert JS, Moran MA (2001) Molecular characterization of estuarine bacterial communities that use high- and low-molecular weight fractions of dissolved organic carbon. Aquat Microb Ecol 25:127-139 Crump BC, Kling GW, Bahr M, Hobbie JE (2003) Bacterioplankton community shifts in an arctic lake correlate with seasonal changes in organic matter. Appl Environ Microbiol 69:2253-2268 Drebes G (1974) Marines Phytoplankton. Georg-Thieme-Verlag, Stuttgart Eilers H, Pernthaler J, Glöckner FO, Amann R (2000) Culturability and in situ abundance of pelagic bacteria from the North Sea. Appl Environ Microbiol 66:3044-3051 Eilers H, Pernthaler J, Peplies J, Glöckner FO, Gerdts G, Amann R (2001) Isolation of novel pelagic bacteria from the German Bight and their seasonal contributions to surface picoplankton. Appl Environ Microbiol 67:5134-5142 Gerdts G, Wichels A, Döpke H, Klings K-W, Gunkel W, Schütt C (2004) A long-term study of microbial parameters near Helgoland (German Bight, North Sea): Historical view and future perspectives. Helgoland Mar Res 58: 230-242 Gonzalez JM, Simo R, Massana R, Covert J, Casamayor E, Pedros-Alio C, Moran MA (2000) Bacterial community structure associated with a dimethylsulfonioproponiate-producing North Atlantic algal bloom. Appl Environ Microbiol 66:4237-4246 Grossart HP, Levold F, Allgaier M, Simon M, Brinkhoff T (2005) Composition of bacterial communities associated with marine diatoms. Environ Microbiol 7:960-973 Jaspers E, Nauhaus K, Cypionka H, Overmann J (2001) Multitude and temporal variability of ecological niches as indicated by the diversity of cultivated bacterioplankton. FEMS Microbiol Ecol 36:153-164 Kirchman DL (2002) The ecology of Cytophaga-Flavobacteria in aquatic environments. FEMS Microb Ecol 39:91-100 Lebaron P, Servais P, Trousselier M, Courties C, Vives-Rego J, Muyzer G, Bernard L, Guindulain T, Schäfer H, Stackebrandt E (1999) Changes in bacterial community structure in seawater mesocosms differing in their nutrient status. Aquat Microb Ecol 19:255-2267

130 Kapitel V Regional variability of bacterial communities in the German Bight

Loewe P, Schmolke S, Becker G, Brockmann U Dick S, Engelke C, Frohse A, Horn W, Klein H, Müller-Navarra S, Nies H, Schmelzer N, Schrader D, Schulz A, Theobald N, Weigelt S (2005) Berichte des Bundesamtes für Seeschiffahrt und Hydrographie. Nr. 38/2005, Hamburg (www.bsh.de) Ludwig W, Strunk O, Westram R, Richter L, Meier H, Yadhukumar, Buchner A, Lai T, Steppi S, Jobb G, Forster W, Brettske I, Gerber S, Ginhart AW, Gross O, Grumann S, Hermann S, Jost R, Konig A, Liss T, Lussmann R, May M, Nonhoff B, Reichel B, Strehlow R, Stamatakis A, Stuckmann N, Vilbig A, Lenke M, Ludwig T, Bode A, Schleifer KH (2004) ARB: a software environment for sequence data. Nucleic Acid Res 32:1363-1371 Lunau M, Lemke A, Walther K, Martens-Habbena W, Simon M (2005) An improved method for counting bacteria in samples with high proportions of particle-associated cells by epifluorescence microscopy. Environ Microbiol 7:961-968 Lunau M, Lemke A, Dellwig O, Simon M (2006) Physical and biogeochemical controls of microaggregate dynamics in a tidally affected coastal ecosystem. Limnol Oceanogr 51:847-859 Musat N, Werner U, Knittel K, Dodenhof T, van Beusekom JE, De Beer D, Dubilier N, Amann R (2006) Microbial community structure of sandy intertidal sediments in the North Sea, Sylt-Romø Basin, Wadden Sea. Syst Appl Microbiol 4:333-348 Muyzer G, de Waal EC, Uitterlinden AG (1993) Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction- amplified genes coding for 16S rRNA. Appl Environ Microbiol 59:695-700 Muyzer G, Brinkhoff T, Nübel U, Santegods C, Schäfer H, Wawer C (1998) Denaturing gradient gel electrophoresis (DGGE) in microbial ecology. Molecular Microbial Ecology Manual, p 1-27. Kluwer Academic Publishers, Dordrecht Parsons TR, Maita Y, Lalli CM (1984) A manual of chemical and biological methods for seawater analysis, p101-112. Pergamon Press, New York Porter K, Feig Y (1980) The use of DAPI for identifying and counting aquatic microflora. Limnol Oceanogr 25:943-948 Reinthaler T, Winter C, Herndl GJ (2005). Relationship between bacterioplankton richness, respiration, and production in the southern North Sea. Appl Environ Microbiol 71:2260- 2266 Riebesell U (1991) Particle aggregation during a diatom bloom. II. Biological aspects. Mar Ecol Prog Ser 69:281-291 Rink B, Seeberger S, Martens T, Duerselen CD, Simon M, Brinkhoff T (2006a) A phytoplankton bloom in a coastal ecosystem affects the composition of bacterial communities. Aquat Microb Ecol, submitted Rink B, Martens T, Fischer D, Lemke A, Grossart HP, Simon M, Brinkhoff T (2006b) Tidal effects on coastal bacterioplankton. Manuscript to be submitted to Aquat Microb Ecol Rooney-Varga JN, Giewat MW, Savin MC, Sood S, LeGresley M, Martin JL (2005) Links between phytoplankton and bacterial community dynamics in a coastal marine environment. Microb Ecol 49:163-175 Sambrook J, Frisch EF, Maniatis T (1989) Northern hybridisation. In : Molecular Cloning: a laboratory manual, 2nd edn., Cold Spring Harbour Laboratory Press, New York

131 Kapitel V Regional variability of bacterial communities in the German Bight

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 Selje N, Simon M (2003) Composition and dynamics of particle-associated and free-living bacterial communities in the Weser estuary, Germany. Aquat Microb Ecol 30:221-236 Selje N, Simon M, Brinkhoff T (2004) A newly discovered Roseobacter cluster in temperate and polar oceans. Nature 427:445-448 Simon M, Grossart HP, Schweitzer B, Plough H (2002) Microbial ecology of organic aggregates in aquatic ecosystems. Aquat Microb Ecol 28:175-211 Stevens H, Brinkhoff T, Simon M (2005a) Composition and seasonal dynamics of free-living, aggregate- and sediment surface-associated bacterial communities in the German Wadden Sea. Aquat Microb Ecol 38:15-30 Stevens H, Stübner M, Simon M, Brinkhoff T (2005b). Phylogeny of Proteobacteria and Bacteroidetes from oxic habitats of a tidal flat ecosystem. FEMS Microbiol Ecol 54:351-365 Utermöhl H (1958) Zur Vervollkommnung der quantitativen Phytoplanktonmethodik. Mitt Int Verh Theor Angew Limnol 9:1-38 Zubkov MV, Fuchs BM, Archer SD, Kiene RP, Amann R, Burkill PH (2002) Rapid turnover of dissolved DMS and DMSP by defined bacterioplankton communities in the stratified euphotic zone of the North Sea. Deep-Sea Res II 49:3017-3038

132 Kapitel V Regional variability of bacterial communities in the German Bight

Table 1: Location of sampling stations, water depth and days of sampling.

Station-no. Latitude Longitude Water depth Days of sampling (°N) (°E) (m) June ‘02 June ‘03 1 54° 07.98’ 7° 04.64’ 32 13 26 2 53° 49.69’ 7° 15.31’ 18 12 26 3 53° 48.33’ 7° 38.45’ 8 12 27 4 53° 52.95’ 8° 05.24’ 8 13 24 5 53° 59.58’ 8° 03.52’ 8 13 24 6 54° 13.91’ 8° 20.66’ 11 11 25 7 54° 32.08’ 8° 10.98’ 9 11 25 8 54° 36.88’ 7° 42.26’ 17 11 25 9 54° 38.44’ 6° 56.41’ 36 11 26 10 54° 28.15’ 7° 15.05’ 29 11 26

133 Kapitel V Regional variability of bacterial communities in the German Bight

Table 2: Phylogenetic affiliation, band-identification (ID), accession number and origin (station) of the free-living bacterial 16S rRNA gene phylotypes retrieved in this study, and their closest related phylotypes, similarity to them and origin.

Phylogen. class Band ID Acc. no. Station Closest relative (acc. no.) Similarity Origin (%)

α-Proteobacteria GB02-e8-FL DQ911764 1 Uncultured alpha proteobacterium clone 96 environmental sample, Loch Fyne, CONW88 (AY828363) Scotland GB02-e9-FL DQ911765 1 Uncultured Roseobacter sp. clone EF100- 98 environmental sample, Monterey 65C12 (AY627371) Bay, California GB03-e16-FL DQ911810 1 Uncultured Roseobacter NAC11-7 98 associated with a DMSP- (AF245635) producing North Atlantic algal bloom GB03-e17-FL DQ911811 1 Uncultured Roseobacter sp. clone EF100- 99 environmental sample, Monterey 65C12 (AY627371) Bay, California GB03-e18-FL DQ911812 2 Uncultured alpha proteobacterium clone 94 bacterioplankton of Ria de Aveiro, RAN-63 (AY499446) Portuguese estuary GB03-e19-FL DQ911813 2 Uncultured alpha proteobacterium clone 98 environmental sample, Loch Fyne, CONW83 (AY828403) Scotland GB03-e22-FL DQ911816 3 Uncultured Roseobacter sp. clone EF100- 99 environmental sample, Monterey 65C12 (AY627371) Bay, California GB03-e26-FL DQ911820 9 Uncultured Roseobacter NAC11-7 99 associated with a DMSP- (AF245635) producing North Atlantic algal bloom GB03-e27-FL DQ911821 10 Uncultured alpha proteobacterium clone 96 environmental sample, Loch Fyne, CONW83 (AY828403) Scotland GB03-c10-FL DQ911777 9 Uncultured Rhodobacteraceae bacterium 97 Oxygen minimum zone, Chile clone ESP450-K6III-60 (DQ810729) GB03-c11-FL DQ911778 9 Uncultured alpha proteobacterium isolate 99 German Wadden Sea DGGE band GWS-TC-e9-FL (DQ911830) GB03-a14-FL DQ911793 1 Uncultured alpha proteobacterium isolate 100 German Wadden Sea, June 2000 DGGE band GWS-FL-2 (AY274228) GB03-a15-FL DQ911794 1 Uncultured Roseobacter sp. clone EF100- 97 environmental sample, Monterey 65C12 (AY627371) Bay, California GB03-a16-FL DQ911795 4 uncultured alpha proteobacterium CHAB-I- 99 marine bacterial assemblage 5 (UAL240910) during confinement GB03-a17-FL DQ911796 6 uncultured alpha proteobacterium CHAB-I- 99 marine bacterial assemblage 5 (UAL240910) during confinement

β-Proteobacteria GB03-e25-FL DQ911819 7 Uncultured marine bacterium clone 99 marine bacterioplankton, San SPOTSAPR01_5m110 (DQ009366) Pedro Ocean Time Series Bacteroidetes GB02-e10-FL DQ911766 5 Uncultured Flavobacteria bacterium 16S 97 Helgoland Roads; coastal North rRNA gene, clone NorSea33 (AM279185) Sea GB03-e15-FL DQ911809 1 Uncultured Bacteroidetes bacterium isolate 96 German Wadden Sea DGGE band GWS-c3-FL (DQ080946) GB03-e21-FL DQ911815 3 Uncultured Flavobacteria bacterium 16S 97 Helgoland Roads; coastal North rRNA gene, clone NorSea76 (AM279177) Sea GB03-e23-FL DQ911817 5 Uncultured bacterium isolate DGGE band 95 environmental sample, associated D37 (AF466926) with dinoflagellate A. tamarense GB03-e24-FL DQ911818 7 Uncultured Flavobacteria bacterium 16S 96 Helgoland Roads; coastal North rRNA gene, clone NorSea76 (AM279177) Sea GB03-c7-FL DQ911774 3 Uncultured Flavobacteria bacterium 16S 96 Helgoland Roads; coastal North rRNA gene, clone NorSea76 (AM279177) Sea GB03-c8-FL DQ911775 4 Uncultured Flavobacteria bacterium 16S 99 Helgoland Roads; coastal North rRNA gene, clone NorSea58 (AM279203) Sea GB03-c9-FL DQ911776 5 Uncultured Sphingobacteria bacterium 16S 99 Helgoland Roads; coastal North rRNA gene, clone NorSea49 (AM279196) Sea GB03-c12-FL DQ911779 10 Uncultured bacterium clone CTD5B 97 environmental sample, (AF469385) subseafloor, after deep-sea volcanic eruption

134 Kapitel V Regional variability of bacterial communities in the German Bight

Table 3: Phylogenetic affiliation, band-identification (ID), accession number and origin (station) of the particle-associated bacterial 16S rRNA gene phylotypes retrieved in this study, and their closest related phylotypes, similarity to them and origin.

Phylogen. class Band ID Acc. no. Station Closest relative (acc. no.) Similarity Origin (%)

α-Proteobacteria GB02-e3-PA DQ911759 4 Uncultured proteobacterium OCS126 99 bacterioplankton, continental shelf (AF001638) off Oregon, USA; SAR116 cluster GB02-e4-PA DQ911760 5 Sulfitobacter sp. DG1020 (AY258095) 98 strain associated with the dinoflagellate Gymnodinium catenatum GB02-e6-PA DQ911762 10 Roseobacter sp. MED008 (AY136104) 98 environmental sample, eastern Mediterranean Sea GB02-e7-PA DQ911763 10 Roseobacter sp. RED1 (AY136122) 97 environmental sample, Gulf of Eilat, Red Sea GB03-a1-PA DQ911780 1 Sphingomonas sp. Pd-S-(s)-m-D-1(2) 99 Endophytic at Rice Plants (Oryza (AB242948) sativa) GB03-a4-PA DQ911783 4 Uncultured marine bacterium clone 98 Antarctica: near Anvers Island AntCL2C1 (DQ906745) GB03-a5-PA DQ911784 5 Uncultured Rhodobacteraceae bacterium 99 10e-5 dilution step, marine sample, clone RCA-H28 (DQ489286) Weser Estuary, Germany GB03-a7-PA DQ911786 6 Uncultured alpha proteobacterium, clone 96 epibiotic bacteria in the accessory T63ANG236 (AJ633963) nidamentalglands of squids

δ-Proteobacteria GB03-a2-PA DQ911781 3 Uncultured delta proteobacterium clone 98 environmental sample, intertidal YS-UMF1_C112 (DQ901575) sediment, Korea GB03-a3-PA DQ911782 3 Uncultured delta proteobacterium, isolate 99 German Wadden Sea, sediment DGGE band 160NF32 (AM072605) GB03-a6-PA DQ911785 5 Uncultured delta proteobacterium clone 100 Sediment surface associated, North Belgica2005/10-130-17 (DQ351762) Sea GB03-a8-PA DQ911787 7 Uncultured delta proteobacterium clone 93 environmental sample, intertidal YS-UMF1_C112 (DQ901575) sediment, Korea GB03-a9-PA DQ911788 7 Unidentified proteobacterium OM27 99 marine coastal picoplankton, (U70713) continental shelf, off Cape Hatteras, North Carolina GB03-a10-PA DQ911789 7 Uncultured bacterium DGGE gel band FD 96 Faroe Deep of the central Baltic 20 (DQ385045) Sea (2 m) GB03-a11-PA DQ911790 7 Uncultured delta proteobacterium clone 98 Sediment surface associated, North Belgica2005/10-130-15 (DQ351760) Sea GB03-a12-PA DQ911791 7 Uncultured delta proteobacterium clone 98 Sediment surface associated, North Belgica2005/10-ZG-2 (DQ351798) Sea GB03-a13-PA DQ911792 8 Uncultured delta proteobacterium clone 93 farm soil adjacent to a silage AKYH967 (AY922176) storage bunker

γ-Proteobacteria GB03-e7-PA DQ911802 6 Uncultured gamma proteobacterium isolate 98 aggregate-associated, German DGGE band GWS-TC-e4-PA (DQ911825) Wadden Sea GB03-e10-PA DQ911804 7 Uncultured delta Proteobacterium, isolate 96 coastal subsurface sediment DGGE band IIIA3 (AJ889159) German Wadden Sea

Bacteroidetes GB02-e5-PA DQ911761 9 Marine Eubacterial sp. (aggregate agg32) 99 aggregate-attached, marine (L10944) bacterial assemblages GB03-e4-PA DQ911799 6 Uncultured bacteroidetes bacterium, isolate 98 Tidal flat sediment (1 m) German DGGE band 100G15 (AJ880446) Wadden Sea GB03-e5-PA DQ911800 6 Uncultured bacterium SB-42-DB 97 satellite bacterium of Dytilum (AJ319829) brightwellii GB03-e14-PA DQ911808 9 Marine Eubacterial sp. (aggregate agg32) 98 aggregate-attached, marine (L10944) bacterial assemblages GB03-c2-PA DQ911769 6 Flavobacteriaceae bacterium T15 98 strain, 10e-5 dilution step, German (AY177723) Wadden Sea GB03-c3-PA DQ911770 6 Uncultured Bacteroidetes bacterium clone 96 coastal bacterioplankton sample of PI_4q10f (AY580698) Plum Island Sound Estuary GB03-c4-PA DQ911771 6 Uncultured bacterium SB-42-DB 98 satellite bacterium of Dytilum (AJ319829) brightwellii GB03-c5-PA DQ911772 8 Flavobacteriaceae bacterium T15 98 strain, 10e-5 dilution step, German (AY177723) Wadden Sea

Firmicutes GB03-c1-PA DQ911768 1 Uncultured bacterium clone Napoli-4B-79 96 marine sediment, Napoli mud (AY592793) volcano, Eastern Mediterranean

135 Kapitel V Regional variability of bacterial communities in the German Bight

Figure Legends

Fig. 1. Study area and locations of the sampling stations in the German Bight, Southern North Sea.

Fig. 2. Scatter plot of temperature versus salinity of the sampling stations in the German Bight in June 2002 and 2003. For exact locations and dates see Table 1.

Fig. 3. Suspended matter dry weight (SPM) in 2002 (A) and 2003 (B), POC in 2003 (B), Chlorophyll a in 2002 (C) and 2003 (D), phaeopigments in 2003 (D), phytoplankton cell numbers in 2003 (E) bacterial numbers in 2002 (F) and 2003 (G) at various stations in the German Bight. Samples were collected in June of both years. Stations 8-10 and 1 are off shore, stations 2-7 near shore. For exact locations and dates see Table 1. Missing bars: data not available.

Fig. 4. Cluster analyses of the DGGE banding patterns of particle-attached and free-living Bacteria (EUB02, A; EUB03, B), Bacteroidetes (CFB03, C) and α-Proteobacteria (ALF03, D) retrieved from samples at various stations in the German Bight in June 2002 (02) and 2003 (03). PA: Particle associated bacterial community.

Fig. 5. Phylogenetic trees of α-Proteobacteria (A), δ-Proteobacteria (B), β- and γ-Proteobacteria (C) and the Bacteroidetes phylum (D) calculated with Maximum-Likelihood based on 16S rRNA gene fragments. Sequences obtained in this study are highlighted in bold.

136 Kapitel V Regional variability of bacterial communities in the German Bight

Fig. 1. Rink et al.

36 2002 34 9 10 8 5 32 1 4 3 2 30 6 7

28 36 9 2003 Salinity (psu) 10 1 2 34 4 8 3 7 5 32 6

30

28 13 14 15 16 17 18 Temperature (°C)

Fig. 2. Rink et al.

137 Kapitel V Regional variability of bacterial communities in the German Bight

2002 2003

) 16 1,6

-1 A B SPM POC

12 1,2 ) -1

8 0,8

4 0,4 L (mg POC

SPM dry weight (mg L (mg weight dry SPM 0 0,0

) C D Chl -1 6 6

Phaeo )

-1 (µg L

a 4 4

2 2 (µg Phaeo L Chlorophyll Chlorophyll 0 0

) 1200 E -1 Leptocylindrus danicus 1000 Thalassionema nitzschioides Pseudonitzschia pungens 800 Guinardia flaccida 600 Guinardia delicatula Rhizosolenia spp 400 Rhizosolenia imbricata 200 Phytoplankton (cells ml 0

) 6

-1 F G 5 4 cells mL cells

6 3

2

1

(10 Bacteria 0 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 Stations Stations

Fig. 3. Rink et al.

138 Kapitel V Regional variability of bacterial communities in the German Bight

St. 4 St. 5 A St. 1 St. 10 Free-living St. 6 St. 8 St. 3 St. 9 St. 7 St. 2 St. 10 PA EUB 02 St. 5 St. 1 St. 8 St. 7 Particle-attached St. 9 St. 4 St. 2 St. 6 St. 3 Std. left Std. right 25 50 75 100

100 St. 6 St. 7 B 100 St. 5 84 95 St. 3 80 St. 2 Particle-attached 83 94 St. 10 87 St. 1 St. 8 St. 9 100 St. 10 EUB 03 95 St. 1 94 St. 3 68 100 St. 8 67 St. 5 75 100 St. 2 96 Free-living St. 4 76 St. 9 100 St. 6 St. 7 100 Std. right Std. left

St. 6 St. 7 C St. 4 Particle-attached St. 5 St. 10 St. 2 St. 3 St. 8 St. 1 St. 9 CFB 03 St. 6 St. 5 St. 2 St. 3 St. 7 St. 4 Free-living St. 1 St. 9 St. 8 St. 10

St. 7 St. 6 D St. 8 Particle-attached St. 10 St. 2 St. 3 St. 5 St. 9 St. 4 St. 7 ALF 03 St. 6 St. 1 St. 2 St. 4 Free-living St. 5 St. 3 St. 9 St. 8 St. 10 Pearson correlation [0.0%-100.0%] Fig. 4. Rink et al. 139 Kapitel V Regional variability of bacterial communities in the German Bight

A

RCA

WM11-36 Roseobacter clade

NAC11-7

Fig. 5A. Rink et al.

140 Kapitel V Regional variability of bacterial communities in the German Bight

B

C SAMMIC

gamma

beta

Fig. 5B, C. Rink et al.

141 Kapitel V Regional variability of bacterial communities in the German Bight

D

Fig. 5D. Rink et al.

142 Kapitel V Regional variability of bacterial communities in the German Bight

Supplementary Figure: DGGE banding patterns which are the basis for the cluster analysis. Numbers and arrows indicate excised and sequenced bands and asterisks chloroplast-derived 16S rRNA gene phylotypes.

143

VI. Schlussbetrachtung und Ausblick

144 Kapitel VI Schlussbetrachtung und Ausblick

Die vorliegende Arbeit umfasst detaillierte Untersuchungen zur Struktur und Abundanz des Bakterioplankton im ostfriesischen Wattenmeer sowie der Deutschen Bucht. Besondere Berücksichtigung fand die Entwicklung und Anwendung spezifischer Nachweise für Bakteriengruppen, die durch methodische Einschränkungen in der Vergangenheit als bisher unterrepräsentiert angesehen werden mußten. Die Ergebnisse dieser Arbeit wurden im Rahmen der DFG geförderten Forschergruppe „BioGeoChemie des Watts“ gewonnen und stellen eine wesentliche Grundlage für zukünftige Untersuchungen bezüglich des Stoffumsatzes abundanter Bakteriengruppen dar.

Das Bakterioplankton im Wattenmeer kann in die Kompartimente freilebend, Aggregat- und Sediment- assoziiert unterteilt werden (vgl. Stevens et al. , 2005,). Die DGGE-Auswertung von Probenahmen in monatlichen Abständen ergab eine relative Stabilität der Bakteriengemeinschaften im Wattenmeer (Stevens et al., 2005). Diese veränderten sich nur geringfügig zu extremen Ereignissen, z. B. während und nach Phytoplanktonblüten und bei Sturm. Da diese Ereignisse fester Bestandteil der Jahreszeiten sind und regelmäßig wiederkehren, wurden sie als stabile Merkmale des Ökosystems Wattenmeer gewertet und die Schlussfolgerung war, dass das Bakterioplankton praktisch keinen Änderungen unterliegt. Die Aussagen von H. Stevens sind, basierend auf den damals erhobenen Daten, durchaus zutreffend, bedürfen allerdings durch die Ergebnisse der vorliegenden Arbeit nun der Ergänzung. In Kapitel II wurde deutlich, dass die große Stabilität der Bakteriengemeinschaft bei genauerer Betrachtung über den Zeitraum einer Phytoplanktonblüte moduliert wird. Man kann annehmen, dass durch Ausscheidungen der Diatomeen und auch deren Absterben bestimmte Bakteriengruppen in einem kurzen Zeitintervall stark beeinflusst werden. Dies betrifft vor allem Bakterien des Phylums Bacteroidetes und der Roseobacter -Gruppe innerhalb der alpha- Proteobakterien , die mit Änderungen der Phytoplanktonzusammensetzung korreliert waren. Es zeigte sich auch, dass die Bakteriengemeinschaften der unterschiedlichen Kompartimente (freilebend und Aggregat-assoziiert) auf unterschiedliche Weise beeinflusst werden. Eine überraschend hohe Artenvielfalt von Bakterien der Roseobacter -Gruppe war auf Aggregaten zu beobachten, die annehmen lässt, dass diese Gruppe für die Zersetzung partikulären organischen Materials (POM) bisher unterschätzt wurde. Ebenso deutlich wurde, dass frei lebende Bacteroidetes sehr schnell von Änderungen in der Umgebung beeinflusst werden und ihre Rolle im Abbau von gelöstem organischem Material (DOM) wahrscheinlich als wesentlich größer eingeschätzt werden kann als bisher angenommen. Die Bedeutung der

145 Kapitel VI Schlussbetrachtung und Ausblick vorliegenden Arbeit liegt daher in der Beschreibung des Potentials einzelner Bakteriengruppen, die im Ökosystem Wattenmeer extremen Einflüssen unterliegen und sowohl Stabilität als auch Flexibilität besitzen. Die Auswirkung der einzelnen Bakteriengruppen auf den gesamten Stoffumsatz ist daher komplex zu betrachten und sollte in weiteren Untersuchungen besonderen Stellenwert einnehmen.

Hinweise über Veränderungen bakterieller Aktivität wurden sogar während eines Tidenzyklus beschrieben (Grossart et al. 2004). Diese Beobachtung ließ die Frage entstehen, auf welcher Zeitskala Reaktionen der Bakteriengemeinschaft nachweisbar sind. In Kapitel II war ersichtlich, dass Bakteriengemeinschaften in wöchentlichen Zeitabständen starken Änderungen unterliegen können. Die Abhängigkeit der Änderungen in der Bakteriengemeinschaft von saisonalen Variablen, z. B. Sturmereignisse im Herbst und die Phytoplanktonblüte im Frühjahr und Sommer, war bereits beschrieben worden (Stevens et al. 2005). Daher wurde in Kapitel III anhand von Probenahmen zu den Kenterpunkten bei Hoch- und Niedrigwasser sowie zum Strömungsmaximum und hoch sensitiver Nachweismethoden der Einfluss der Tide auf die Zusammensetzung der Bakteriengemeinschaften zu unterschiedlichen Jahreszeiten untersucht. Die Ergebnisse zeigen, dass wenige Änderungen innerhalb der Bakteriengemeinschaften sogar im Gezeitenwechsel zu den Kenterpunkten stattfinden. Einige Phylotypen waren stets nachweisbar und wurden schon beschrieben, z. B. das RCA-Cluster und das WAC I-Cluster (Selje et al., 2004, Stevens et al, 2005), andere Phylotypen jedoch erschienen nur kurzzeitig zu bestimmten Ereignissen. Diese Änderungen waren im Wesentlichen durch den kombinierten Einsatz von RNA-basierter PCR und spezifischen Primern für Bacteroidetes , alpha-Proteobakterien und Roseobacter detektierbar, wodurch die Bedeutung dieser Bakterien für den gesamten Stoffumsatz als eher gering einzuschätzen ist. Die Frage nach ihrer Funktion bleibt dennoch interessant, da es sich innerhalb der Bakteriengemeinschaft um funktionell wichtige Prozesse handeln könnte, die zu bestimmten Zeitpunkten relevant sind. Darüber hinaus war auffällig, dass die Bandenmuster der aktiven Bakterien (RNA-basiert) stark abwichen von den Bandenmustern der als häufig anzunehmenden Bakterien (DNA-basiert). Dies lässt den Rückschluss zu, dass die Anwendung RNA-basierter Methoden bei weiteren Untersuchungen im Wattenmeer dringend notwendig ist, um Korrelationen zwischen Aktivität und Phylogenie herstellen zu können. Große Bedeutung hat die Anwendung der FISH- und CARD-FISH Methode, durch die erstmalig die Abundanz der verschiedenen Bakteriengruppen in der Wassersäule des

146 Kapitel VI Schlussbetrachtung und Ausblick

Wattenmeeres und damit auch der jeweilige potentielle Anteil der Gruppen am Stoffumsatz gezeigt werden konnte. Die Verknüpfung zu spezifischen tidalen Ereignissen war trotz der hochauflösenden CARD-FISH nicht erkennbar, jedoch zeigte sich, dass in den hoch produktiven Frühjahrs- und Sommermonaten wesentlich höhere Varianz innerhalb der Bakteriengruppen auftrat, was ebenfalls auf erhöhte Aktivität der Organismen schließen hinweist. Dies bedarf ebenfalls weiterer Untersuchungen, z. B. in Kombination mit der Micro- FISH Methode, um diese Aussagen stützen zu können.

Kapitel IV zeigte, dass die Actinobakterien im Wattenmeer unerwartet hohe Abundanz und Artenvielfalt aufweisen. Die Frage nach der ökologischen Bedeutung dieser Organismen für Küstengebiete rückt daher immer mehr in den Vordergrund und stellt eine interessante Perspektive für weitere Forschungsarbeiten dar. Auch für diese Gruppe zeigte sich, dass die Einteilung in die Kompartimente frei lebend und aggregat-assoziiert essentiell ist, um weitere Rückschlüsse ziehen zu können. Beide Fraktionen zeigten Unterschiede in der Zusammensetzung der Bakteriengemeinschaft und sind daher auch gesondert zu betrachten. Besonders interessant ist die klare Unterscheidung von Actinobakterien , die im Süßwasser leben, und den halotoleranten marinen Vertretern im Wattenmeer. Actinobakterien wurden bisher durch Produktion von Sekundärstoffen in biotechnologischen Fragestellungen fokussiert, aber auch im ökologischen Sinne ist diese Eigenschaft untersuchenswert. Besonders auf Aggregaten können diese als Sekundärstoffe bezeichneten Botenstoffe besondere Bedeutung für Kommunikation oder Abwehr innerhalb der Bakteriengemeinschaften einnehmen. Als Hauptursache für die Artenvielfalt und Abundanz der Actinobakterien in diesem schwebstoffreichen Grenzbereich zwischen Land und Meer kann jedoch der Abbau komplexer organischer Verbindungen, z. B. Huminstoffe, angenommen werden.

Der Austausch von Wasserkörpern zwischen dem Wattenmeer und der offenen Nordsee und damit verbundene Stoffflüsse sind lange Zeit Mittelpunkt sedimentologischer und weiterer geochemischer Untersuchungen gewesen. Inwiefern Aggregat-assoziierte oder auch frei lebende Bakterien durch diese Prozesse beeinflusst sind und welchen Einfluss standortgebundene Bedingungen auf die Bakteriengemeinschaften haben, zeigen die Ergebnisse aus Kapitel V .

147 Kapitel VI Schlussbetrachtung und Ausblick

Trotz standortspezifischer Unterschiede, die durch die Bandenmuster der partikel-assoziierten Bakteriengemeinschaften reflektiert wurden, war eine homogene Verteilung frei lebender Bakteriengemeinschaften zu beobachten. Die phylogenetischen Stammbäume ergaben große Ähnlichkeit der partikulären Bakteriengemeinschaften des ostfriesischen und nordfriesischen Wattenmeeres. Es konnten Indikatororganismen ausgemacht werden, die ausschließlich Aggregat-assoziiert leben und somit den Verlauf partikulären Materials aus den Wattenmeeren und von resuspendiertem Sediment in der Wassersäule aufzeigen. Überraschend waren auch hier die eindeutige Trennung der Kompartimente, und die große Stabilität innerhalb der alpha-Proteobakterien bzw. der Roseobacter -Gruppe. Nicht veröffentlichte Daten, die im Rahmen dieser Arbeit erhoben wurden, ergaben für die frei lebenden Roseobacter nahezu identische DGGE-Bandenmuster, unabhängig von der Jahreszeit und vom Standort. Die ökologische Bedeutung dieser Gruppe für den Umsatz von gelöstem organischen Kohlenstoff (DOC) ist daher als essentiell anzunehmen und bedarf dringend einer weiteren Erforschung.

Zusammenfassend ergibt sich daher aus dieser Arbeit ein komplexes Bild mit vertieften Erkenntnissen über die Zusammensetzung der Bakteriengemeinschaften im ostfriesischen Wattenmeer und der Deutschen Bucht. Die sensitiven Untersuchungsmethoden konnten zeigen, welche Einflussfaktoren in diesem als persistent bezeichneten Ökosystem eine wesentliche Rolle einnehmen und bilden daher die Grundlage für weitere Forschungsarbeiten. Aktivitätsmessungen und Erhebung des genetischen Potentials für Stoffwechselwege stellen daher eine geeignete Ergänzung dar, die durch weitere Arbeiten im Rahmen des Forschergruppenverbundes erhoben werden können. Intensive Datenerhebung von bakterieller Biomasseproduktion findet derzeit in stündlichen bis saisonalen Zeitskalen statt, die gleichzeitige Etablierung der Micro-CARD-FISH stellt eine weitere, höchst effektive Möglichkeit zur Aktivitätsmessung im Wattenmeer dar. Die Überprüfung der Ergebnisse über experimentelle Ansätze ist derzeit ebenfalls in der Ausführung, und die Verknüpfung zum natürlichen System wird über mathematische Modelle gelenkt. Die Metagenomik bleibt vorerst ein zukunftsträchtiger Ausblick, wird aber im Zuge der Fortschritte in der molekularen Ökologie ein logischer Schritt zur weiteren Erforschung der Bakteriengemeinschaften in der Wassersäule des Wattenmeeres und der Deutschen Bucht darstellen.

148

Erklärung

Hiermit bestätige ich, dass ich die vorliegende Dissertation selbständig verfasst und nur die angegebenen Quellen und Hilfsmittel verwendet habe.

Weiterhin erkläre ich, dass diese Dissertation weder in ihrer Gesamtheit noch in Teilen einer anderen wissenschaftlichen Hochschule zur Begutachtung in einem Promotionsverfahren vorliegt oder vorgelegen hat.

Oldenburg, den

Danksagung

Mein besonderer Dank gilt Prof. Meinhard Simon für das Vertrauen, die Unterstützung und die große Flexibilität, vor allem zum Ende der Fertigstellung dieser Arbeit. Diese Eigenschaften sind keineswegs selbstverständlich und ich bin dankbar, dass ich diese positive Erfahrung machen durfte! Vielen Dank an dieser Stelle auch an Prof. Heribert Cypionka, der unkompliziert die Aufgabe des Gutachters übernommen hat. Für ihn habe ich mich besonders um Prägnanz bemüht - ich hoffe, es ist gelungen.

Für Zielstrebigkeit, fachliche Diskussion und einfach immer wohltuenden freundlichen Rat möchte ich Dr. Thorsten Brinkhoff danken. Er hat mich mehr als einmal vor dem Verzetteln bewahrt und ist mir ein großes Vorbild im Umgang mit Kollegen. Vielen Dank auch für die große Flexibilität zum Ende meiner Arbeit!

Der gesamten Arbeitsgruppe, besonders Andrea Schlingloff und Birgit Kürzel, möchte ich danken für die Geduld, Offenheit und Hilfsbereitschaft. Heike Stevens bin ich mehr als dankbar für 2500000000 motivierende Gespräche und ihre erfrischende Art, das Leben positiv zu sehen. Und für die Fahrt im Schäbi Pop durch Yucatan! Katja Walther und Beate Köpke danke ich für Fachgespräche zu jeder Tages- und Nachtzeit und für die Freundschaft, die eine sehr wichtige Begleitung für mich war und ist.

Die größte Unterstützung fand ich bei meiner Familie, die mir jederzeit zur Seite stand und mir gezeigt hat, wie wichtig dieser Halt für mich ist. Ihr habt alle einen großen Teil zu dieser Arbeit beigetragen! Das werde ich nie vergessen und daher widme ich Euch meine Arbeit aus tiefstem Dank.

Meinem Freundeskreis, für den das Wort „Doktorarbeit“ mittlerweile fest mit Begrüßungs- und Abschiedsformeln verknüpft ist, danke ich für unendliche Ausdauer und Geduld, für seelischen Ausgleich, für tolle Gespräche, und fürs Dasein. Es ist geschafft!!! Ihr könnt wieder ans Telefon gehen!

Lebenslauf

Beate Rink, geboren am 23.01.1974 in Bremerhaven

Schulausbildung

1980-1984 Grundschule Friedrich-Ebert-Schule, Bremerhaven 1984-1986 Orientierungsstufe Wilhelm-Raabe-Schule, Bremerhaven 1986-1990 Gymnasium Wilhelm-Raabe-Schule, Bremerhaven 1990-1993 Gymnasiale Oberstufe Geschwister-Scholl-Schule, Bremerhaven Leistungskurse: Latein, Kunst Abschluß: Abitur

Berufsausbildung

1993-1995 Ausbildung zur Industriekauffrau Fa. Schottke GmbH & Co. KG, Bremerhaven Abschluß: Industriekauffrau 1995-1997 Biologiestudium Universität Regensburg 1997-2001 Biologiestudium Carl-von-Ossietzky-Universität, Oldenburg Hauptfächer: Mikrobiologie, Genetik, Biochemie Abschluß: Diplom Thema Diplomarbeit: „Besiedelung und Abbau von Fucus -detritus durch heterotrophe marine Bakterien“ 2001-2006 Wissenschaftliche Angestellte am Institut für Chemie und Biologie des Meeres Carl-von-Ossietzky-Universität, Oldenburg Arbeitsgruppe „Biologie geologischer Prozesse“, Prof. Meinhard Simon