Lehrstuhl für Siedlungswasserwirtschaft der Technischen Universität München Bacteria and Extracellular Polymeric Substances in Activated Sludge Scum Formation Elisabeth Müller Vollständiger Abdruck der von der Fakultät für Bauingenieur- und Vermessungswesen der Technischen Universität München zur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften (Dr. rer. nat.) genehmigten Dissertation. Vorsitzender: Univ.-Prof. Dr.-Ing. habil. Michael Manhart Prüfer der Dissertation: 1. Univ.-Prof. Dr.-Ing., Dr. h. c. Peter A. Wilderer, i. R. 2. Univ.-Prof. Dr. rer. nat. Karl-Heinz Schleifer 3. Univ.-Prof. Dr. rer. nat. Harald Horn Die Dissertation wurde am 20.04.2006 bei der Technischen Universität München eingereicht und durch die Fakultät für Bauingenieur- und Vermessungswesen am 02.06.2006 angenommen. Schön ist alles, was Himmel und Erde verbindet: der Regenbogen, die Sternschnuppe, der Tau, die Schneeflocke, doch am schönsten ist das Lächeln eines Kindes. Für meine Tochter Mira Madgalena For my daughter Mira Madgalena ACKNOWLEDGMENTS The present study was performed at the Institute of Water Quality Control, Technical University of Munich in cooperation with the Bavarian Environment Agency (formerly Bavarian Water Management Agency) under the supervision by Dr. Hilde Lemmer. Financial support of the research by the Bavarian State Ministry of the Environment, Public Health and Consumer Protection (StMUGV, formerly StMLU) is gratefully acknowledged. My sincere thanks and gratitude goes to Dr. Hilde Lemmer for providing the very interesting project, for her scientific and committed supervision, the many fruitful discussions, her support and encouragements during the last years, and especially during the last phase of my thesis. I record my special thanks to Prof. Dr. Peter Wilderer, first for the opportunity to perform this project at his institute and to give me free hand during my research work, second for his encouragement to write this doctoral thesis and his willingness to be my supervisor. Thanks are due to Prof. Dr. Karl-Heinz Schleifer and Prof. Dr. Harald Horn for their take over of the expertise’s. I wish to express my thanks to Prof. Dr. Linda Blackall for her interest in my research work, the many exciting discussions, and fruitful comments on my thesis, and to her colleagues for the hospitality and fantastic stay down under. The extensive practical work in the laboratory, which contributed to the success of this thesis, would not have been possible without the assistance of Dr. Natalie Zorn. I would like to thank her for this valuable cooperation. For the good time in our “women office“, which is unforgettable, the many scientific discussions and the critical reading of the manuscript I would like to thank Dr. Martina Hausner. I like to send my grateful thanks to Dipl.-Biol. Margit Schade for the good organization of our collective sampling procedure, the many discussion about “scum“ and “FISH“, the proofreading of my thesis and the beautiful atmosphere at the Bavarian Environment Agency, Department of Microbiology. For the stimulating discussions and the good cooperation I would like to thank Dr. Stefania Paris, Dipl.-Biol. Ioanna Letsiou, Dr. Michael Lebuhn and Dr. Stefan Bathe. I would also like to take this opportunity to thank all colleagues of the Institute of Water Quality Control and the Bavarian Environment Agency, Department of Microbiology, for the fair and beautiful working atmosphere. Last but not least my heartfelt thanks go to my family, Andy, Mira, Madga, Jana, Edith and my friends, Simone, Luca, Marc, Michi, Mick for their appreciation and their support during the last years and especially during the last period of completion of this thesis. ABSTRACT The formation of thick viscous scum layers is a persistent problem in wastewater treatment plants (WWTPs) observed all over the world. Three factors are necessary for scum formation: (i) dispersed gas bubbles, (ii) surface-active material and (iii) hydrophobic compounds. For instance, air bubbles in the aeration tank initialize scum development. The gas-water interface is stabilized by the adsorption of surface-active molecules. The presence of hydrophobic material provides stable adhesion to solid particles promoting persistent scum layers. Surface- active substances may enter the WWTP via primary effluent or be produced by various bacteria in the presence of hydrophobic carbon sources and under nutrient deficiency. Hydrophobic material might originate from the primary effluent or be available by an abundant growth of organisms with hydrophobic cell surfaces. One objective of this study was to characterize the relevant bacteria that are involved in scum formation. The identification of these bacteria was performed by classical microscopic sludge analysis and molecular biological methods to obtain information about their taxonomic affiliation. The knowledge about the phylogeny of scum bacteria might reveal physiological properties of these bacteria, which is helpful for the understanding of the scum formation process and the application of specific control measures. A further topic of this study was to investigate whether scum formation is a passive floating mechanism of activated sludge to the water surface or an active growth process of a specific scum biocenosis close to the water surface. The elucidation of the role of these scum bacteria and extracellular polymeric substances (EPS) in the formation of stable scum layers was the third objective of this work. EPS might perform two different functions in the scum process: (i) they serve as nutrient source and promote the growth of a specific bacterial biocenosis and (ii) they directly stabilize scum layers due to their hydrophobic sites. For this purpose, scum bacteria numbers quantified by fluorescence in situ hybridization (FISH) were correlated to sludge hydrophobicity and specific EPS components extracted from activated sludge and scum. A screening of various WWTPs with scum formation revealed that in most plants examined Microthrix parvicella dominated the activated sludge and scum biocenoses. Furthermore nocardioform actinomycetes could be characterized as important scum bacteria. These microorganisms were found at very low numbers in activated sludge whereas they were clearly enriched in scum layers. The filamentous type 1863 found in several WWTPs was clearly enriched in the scum fraction but never detected as the dominant microorganism within the biocenosis. The filamentous bacterial morphotypes 1851, 0041/0675, 0092, and Nostocoida limicola were identified only in some WWTPs as the dominant organisms. However, they were often found concomitantly in M. parvicella- and nocardioform actinomycetes-dominated biocenoses. The identification of filamentous bacterial morphotypes by classical microscopy often failed because of morphological and taxonomic variations. Therefore the FISH technique using specific rRNA-targeted oligonucleotide probes was applied to reveal reliable phylogenetic information on these described morphotypes. The effectiveness of the two methods, classical sludge analysis and FISH, for detection, identification, and quantification of filamentous organisms in activated sludge and scum was compared. Classical sludge analysis and FISH corresponded well in the case of filamentous M. parvicella, which was identified as “Candidatus Microthrix parvicella”, an unclassified member of the Actinobacteria. Type 1863 was characterized as Acinetobacter spp. and type 1851 was identified as a member of the Chloroflexi subdivision 3, closely related to Roseiflexus castenholzii. Classical microscopy overlooked nocardioform actinomycetes and type 1863 single cells, which FISH detected only. Furthermore the non-branched filamentous nocardioform actinomycetes growing as short filaments or in cell clusters were clearly identified by FISH only. FISH underestimated branched filamentous nocardioform actinomycetes and morphotypes 0041/0675 or 0092 because of insufficient cell wall permeability for fluorescently labeled rRNA-targeted oligonucleotide probes, and also because of their taxonomic variability and consequently the availability of adequate probes. FISH results for N. limicola morphotypes I and III are still insufficient because of both their low in situ numbers and due to their high taxonomic variability. However, morphotype N. limicola II was frequently identified as “Candidatus Nostocoida limicola” belonging to the Actinobacteria phylum. The identification of the nocardioform actinomycetes at genus and species level was limited by classical microscopy because the different species exhibit mostly the same morphotype. The use of FISH for this purpose was restricted due to the deficit of available genera- and species-specific probes. The application of the full-cycle rRNA approach, combined with nucleic acid fingerprinting methods, restriction fragment length polymorphism (RFLP) and denaturing gradient gel electrophoresis (DGGE), proved to be a successful approach to identify unknown nocardioform actinomycete species in different scum samples. A screening of the clone libraries with Actinobacteria-specific primers followed by RFLP profiling allowed to organize clone libraries into taxonomic groups and the identification of the dominant clones for each library. The analysis of environmental DNA by DGGE and the comparison with DGGE patterns of the dominant clones indicate that the dominant clone insert represented the dominant species within the Actinobacteria phylum of the microbial