Microbial Ecology of Phototrophic  2007 by Guus Roeselers www.roeselers.com Microbial Ecology of Phototrophic Biofilms

PROEFSCHRIFT

ter verkrijging van de graad van doctor aan de Technische Universiteit Delft, op gezag van de rector Magnificus prof. dr. ir. J. T. Fokkema, voorzitter van het College voor Promoties, in het openbaar te verdedigen op dinsdag 11 september 2007 om 15.00 uur.

door

Guus Roeselers

doctorandus in de Biologie geboren te Leeuwarden Dit proefschrift is goedgekeurd door de promotor: Prof. dr. ir. M. C. M. van Loosdrecht

Toegevoegd promotor: Dr. G. Muyzer

Samenstelling promotie commissie: Prof. K. C. A. M. Luyben, Technische Universiteit Delft, voorzitter Prof. dr. ir. M. C. M. van Loosdrecht, Technische Universiteit Delft, promotor Dr. G. Muyzer, Technische Universiteit Delft, toegevoegd promotor Prof. dr. J. G. Kuenen, Technische Universiteit Delft Prof. dr. W. Admiraal, Universiteit van Amsterdam Prof. dr. L. J. Stal, Universiteit van Amsterdam Prof. dr. H.J. Laanbroek, Universiteit Utrecht Dr. A. Wilmotte, Université de Liège, België

Het in dit proefschrift beschreven onderzoek is uitgevoerd bij de sectie Milieubiotechnologie, afdeling Biotechnolgie, Technische Universiteit Delft, Julianalaan 67, 2628 BC, Delft.

Dit onderzoek is gefinancierd door de Europese Unie (PHOBIA: QLK3-CT-2002- 01938) en de Technische Universiteit Delft.

ISBN: 978-90-9022164-9 Voor Clara en Stella

CONTENTS

Chapter 1 General introduction 9

Chapter 2 Phototrophic biofilms and their potential applications 17

Chapter 3 Diversity of phototrophic in microbial mats from 33 Arctic hotsprings (Greenland)

Chapter 4 Heterotrophic pioneers facilitate phototrophic 53 development

Chapter 5 On the reproducibility of microcosm experiments - 67 different community composition in parallel phototrophic biofilm microcosms

Chapter 6 Development of a PCR for the detection and 83 identification of cyanobacterial nifD genes

Chapter 7 Diversity and expression of cyanobacterial hupS genes 97 in pure cultures and in a nitrogen limited phototrophic biofilm

Chapter 8 Concluding remarks 111 Summary Samenvatting About the author List of publications Acknowledgements

1

General Introduction Chapter 1

INTRODUCTION

Biofilms Biofilms can be defined as layered structures of microbial cells and an extracellular matrix of polymeric substances, associated with surfaces and interfaces. Biofilms trap nutrients for growth of the enclosed microbial community and help prevent detachment of cells from surfaces in flowing systems. Moreover, microorganisms embedded in a biofilm are protected against toxic substances, predation, and for that matter host immune responses (Davey and O'Toole G, 2000). However, life in a biofilm has also some disadvantages compared to the planktonic mode of life. Cells growing in a biofilm often experience mass transfer limitation and competition for available resources may be intense. Therefore, microbial growth rates are often found to be lower in a biofilms than in suspension. Given their ubiquity and importance in the microbial world, it is hardly surprising that biofilms have received much attention from the scientific community. However, the morphology, chemistry, physiology, and ecology of naturally occurring biofilms are as diverse as their constituent microorganisms. For example, pearl-string-like archaeal biofilms found in sulfidic springs (Henneberger et al., 2006) are profoundly different from the plaque formed on air-exposed surfaces of teeth (Kolenbrander and London, 1993).

Phototrophic biofilms and microbial mats The structure, growth dynamics and physiology of heterotrophic biofilms have been extensively studied. But until recently phototrophic biofilms have received little attention. Phototrophic biofilms occur on contact surfaces in a range of terrestrial and aquatic environments. Phototrophic biofilms can best be described as surface attached microbial communities mainly driven by light as the energy source with a photosynthesizing component clearly present. Eukaryotic and generate energy and reduce carbon dioxide, providing organic substrates and . The photosynthetic activity fuels processes and conversions in the total biofilm community, including the heterotrophic fraction. Thick laminated multilayered phototrophic biofilms are usually referred to as microbial mats or phototrophic mats. Phototrophic biofilms and microbial mats have been described in extreme environments like thermal springs, hyper saline ponds (Sorensen et al., 2005), desert soil crusts (Garcia-Pichel and Pringault, 2001), and in lake ice covers in Antarctica (Taton et al., 2003). The 3.4-billion-year fossil record of benthic phototrophic communities, such as microbial mats and (Des Marais, 1990; Arp et al., 2001), indicates that these associations represent the Earth's oldest known ecosystems. These early ecosystems played a key role in the formation of the present Earth’s oxygenic atmosphere (Knoll, 1996; Hoehler et al., 2001).

Green algae, diatoms and cyanobacteria Algae are a large and diverse group of eukaryotic organisms containing chloroplasts capable of performing oxygenic . Although most algae are of microscopic size and hence are clearly microorganisms, some macroscopic species known as kelps are over 70 meters in length. Green algae (Chlorophyta) are a large group of algae quite closely related to green plants (Lewis and Mccourt, 2004). Their chloroplasts contain chlorophyll a and b, giving them a characteristic bright green color. Some species are flagellated and motile, other are immobile but often have a flagellated stage in their life cycle. Most species live in freshwater and marine environments, but some species are adapted to terrestrial environments such as the surface of rocks or trees. Some lichens are symbiotic associations of fungi and green alga.

10 General introduction

Diatoms (Bacillariophyta) are another major group of eukaryotic . Most diatoms are unicellular, although some form chains or simple colonies. They are characterized by their highly ornamented glasslike cell walls consisting of hydrated silicon dioxide. Diatoms have an extensive fossil record going back to the Cretaceous; some rocks are formed almost entirely of fossil diatoms. Many diatoms show gliding motility, giving them the ability to adhere and glide over a substratum. Benthic diatoms are an important component of the microphytobenthos that inhabit intertidal mudflats. They are also an important component of the biofouling community in marine environments. Cyanobacteria are Gram-negative, aerobic bacteria containing the photosynthetic pigment chlorophyll a and performing oxygenic photosynthesis, fixing CO2 through the reductive pentose phosphate cycle (Castenholz, 2001). This monophyletic group is morphologically and developmentally very diverse (Figure 1).

Figure 1. Micrograph of an Anabaena variabilis filament with a heterocyst. The

nitrogenase enzyme complex involved in the fixation of atmospheric N2 is extremely sensitive to oxygen. Heterocysts are cells specialized in the fixation

of N2 under oxic conditions, which differentiate at regular intervals along the filaments of some diazotrophic cyanobacteria.

Both unicellular and filamentous forms are known, and considerable variation within these morphological types occurs (Rippka et al., 1979; Castenholz, 2001). They are widely distributed in nature in terrestrial, freshwater, and marine habitats. In general they are more tolerant to extreme environmental conditions than eukaryotic algae. Many species grow attached to the surfaces of rocks and soils. Cyanobacteria can form dense microbial mats, especially in extreme environment with low grazing pressure such as hyper saline ponds (Caumette et al., 1994) and hot springs (Ward et al., 1998).

Cyanobacterial nitrogen fixation An experiment that revealed cyanobacterial nitrogen fixation was already performed by Beijerinck (Beijerinck, 1901). He covered garden soil with two solutions: water buffered with potasium phosphate and water containing essential nutrients, including nitrate.

After a few weeks of incubation in daylight, the soil with the KPO4 solution was covered with a film of Nostoc, Anabaena, and Cylindrospermum filaments, genera that can fix atmospheric nitrogen. The soil treated with all essential nutrients showed an entirely different community, including Oscillatoria (not known to be capable of nitrogen fixation), diatoms, and some green algae. Bejierinck was clearly selecting for nitrogen fixing genera with his nitrogen-poor solution.

11 Chapter 1

Biological N2 fixation involves the activity of the oxygen sensitive multi-component nitrogenase enzyme (Mancinelli, 1996). The conventional nitrogenase is encoded by the nifHDK genes (Figure 2). The nitrogenase subunit genes are like the 16S rRNA genes highly conserved and can be used as phylogenetic markers (Zehr et al., 1995).

Diazotrophic cyanobacteria (cyanobacteria capable of N2 fixation) can be subdivided into three main groups (Stal, 1995). Group I consists of heterocystous cyanobacteria. Heterocysts are differentiated cells specialized in nitrogen fixation. Their thick cell wall contains special lipopolysacharides and forms a barrier for gases, limiting the entry of oxygen (Figure 1). Heterocysts have lost the capacity of oxygenic photosynthesis and provide the other cells in the filament with nitrogen for biosynthesis. Group II consist of filamentous and unicellular cyanobacteria that do not form heterocysts and only perform N2 fixation under anoxic conditions. Group III cyanobacteria comprise filamentous and unicellular cyanobacteria that are able to fix N2 under fully aerobic conditions. Their mechanism of protecting the nitrogenase complex from O2 inhibition is still not fully understood.

Uptake hydrogenases

Most nitrogenases are catalysts for hydrogen production as they liberate H2 during the reduction of nitrogen to ammonia (Mancinelli, 1996). In cyanobacteria, a NiFe-uptake hydrogenase recycles the H2 effused by the nitrogenase machinery (Figure 2). The recycling of hydrogen has been suggested to have three physiological functions. I) It generates ATP via the oxyhydrogen (Knallgas) reaction, II) it protects the sensitive nif complex from oxydative inactivation, and III) it supplies reducing equivalents (electrons) for N2 reduction and other cell functions (Bothe et al., 1977). production could be facilitated by the selection of strains, or the production of mutants, deficient in H2 uptake activity.

Figure 2. Enzyme complexes involved in cyanobacterial nitrogen fixation and hydrogen metabolism.

12 General introduction

The PHOBIA project The research described in this thesis was partly performed within the frame of the European Union funded project PHOBIA (Phototrophic biofilms and their potential applications: towards the development of a unifying concept). This 36-month project has been funded within the 5th EU Framework Programme 'Quality of Life', Key action 3 'The Cell Factory'. A consortium with the Centre for Environmental Research, Magdeburg (Germany), Institute of Chemical and Biological Technology, Oeiras (Portugal), University of Copenhagen (Danmark), The Netherlands Institute of Ecology (NIOO-KNAW), University of Rome “Tor Vergata” (Italy), and Delft University of Technology investigated the microenvironment, physiology, and spatial development of a number of aquatic phototrophic biofilms. The objectives of PHOBIA were (1) to comprehend how aquatic biofilms adhere to submerged surfaces, how the biofilms grow and function; (2) to make a unifying conceptual model for phototrophic biofilms, based on artificial neural networks. Such self-learning models can predict how environmental conditions structure the biofilm (both in spatial terms and in terms of biodiversity/phylogeny), and how the biofilm interacts with the environment and the substrata. The knowledge of these biofilms will provide a basis for applied research in the field of water and development of new antifouling agents. Within the PHOBIA project, various aspects of phototrophic biofilm growth were studied in separate work packages. Additional to biofilm structure and architecture, physical and chemical gradients, surface interactions, and biofilm strength, an analysis of the microbial species in the biofilm communities was made. This included (1) a taxonomic analysis of the major species, using light- and electron microscopy (University of Rome); (2) fingerprints of membrane phospholipid fatty acids (PFLA) that are specific for various groups of microbes (heterotrophic bacteria, cyanobacteria, green algae, diatoms), using GC-MS and fingerprints of group-specific photosynthetic pigments via HPLC-PDA (NIOO-KNAW); (3) phylogenetic analysis of the biofilm species (photo- and heterotrophs) using molecular techniques such as PCR-denaturing gradient gel electrophoresis (DGGE) (Delft University of Technology).

Phototrophic biofilm incubator In several chapters of this thesis results are presented that were obtained with aquatic phototrophic biofilms cultivated in a temperature controlled flow-lane incubator (Fig. 4) which was designed by collaborating partners in the PHOBIA consortium and constructed by the workshop at the Centre for Environmental Research in Magdeburg. This flow-lane incubator system contained four separate flow channels (LC1, LC2, LC3 and LC4) through which a volume of 4 L mineral medium circulated over a surface covered with 47 polycarbonate slides (76 x 25 x 1 mm). Polycarbonate slides were used as a substratum for biofilm adhesion. Each light chamber contained an adjustable light source and the circulation speed of the culture medium could be regulated precisely. Biofilm growth was monitored and recorded with three light sensors positioned directly under selected polycarbonate slides (Figure 3 and 4). Each light sensor contained three independent photodiodes. Decrease of subsurface light was used as an indicator for different biofilm growth stages (mean of three light sensors along 1.2 m length of each flow lane).

13 Chapter 1

Figure 3. Schematic representation of the phototrophic biofilm incubator.

Figure 4. A) Top view of the incubator flow lanes. B) Inside an incubator flow-lane. In the front: incident light sensor mounted on a polycarbonate slide. At the back: temperature sensor submerged in the inlet reservoir.

14 General introduction

OUTLINE

Chapter 2 provides a brief introduction in the ecology of phototrophic biofilms and discusses their actual and potential applications in , bioremediation, fish-feed production, biohydrogen production, and soil improvement and their role in biofouling. Chapter 3 describes the diversity of phototrophic bacteria in hot spring microbial mats found on the east coast of Greenland. Chapter 4 focuses on the successional changes in community composition of freshwater phototrophic biofilms growing under different light intensities. The role of heterotrophic bacteria in phototrophic biofilm development is further explored In Chapter 5 the incubator system is used for an unprecedented microcosm reproducibility experiment. Chapter 6 demonstrates that nifD gene sequences can be used to detect and identify diazotrophic cyanobacteria in natural communities. PCR products generated using primers homologous to conserved regions in the cyanobacterial nifD genes were subjected to DGGE and clone library analysis in order to determine the genetic diversity of diazotrophic cyanobacteria in environmental samples. Chapter 7 describes the development of PCR primers targeting conserved regions within the cyanobacterial hupS gene family. We analyzed hupS diversity and transcription in cultivated phototrophic biofilms by the direct retrieval and analysis of mRNA that was reverse transcribed, amplified with hupS specific primers, and cloned. In Chapter 8 the main findings presented in this thesis are summarized and evaluated.

REFERENCES

Arp, G., Reimer, A., and Reitner, J. (2001) Photosynthesis-induced biofilm calcification and calcium concentrations in Phanerozoic oceans. Science 292: 1701-1704. Beijerinck, M.W. (1901) On oligonitrophilous bacteria. In KNAW Proceedings volume 3; 1900-1901. Amsterdam, pp. 586-595. Bothe, H., Tennigkeit, J., and Eisbrenner, G. (1977) The utilization of molecular hydrogen by the blue- green alga Anabaena cylindrica. Arch Microbiol 114: 43-49. Castenholz, R.W. (2001) Phylum BX. Cyanobacteria. Oxygenic photosynthetic bacteria. In Bergey's Manual of Systematic Bacteriology. Boone, D.R., Castenholz, R.W., and Garrity, G.M. (eds). New York, USA: Springer, pp. 474-487. Caumette, P., Matheron, R., Raymond, N., and Relexans, J.C. (1994) Microbial Mats in the Hypersaline Ponds of Mediterranean Salterns (Salins-De-Giraud, France). Fems Microbiology Ecology 13: 273-286. Davey, M.E., and O'Toole G, A. (2000) Microbial biofilms: from ecology to molecular genetics. Microbiol Mol Biol Rev 64: 847-867. Des Marais, D.J. (1990) Microbial mats and the early evolution of life. Trends Ecol Evol 5: 140-144. Garcia-Pichel, F., and Pringault, O. (2001) Microbiology. Cyanobacteria track water in desert soils. Nature 413: 380-381. Henneberger, R., Moissl, C., Amann, T., Rudolph, C., and Huber, R. (2006) New insights into the lifestyle of the cold-loving SM1 euryarchaeon: natural growth as a monospecies biofilm in the subsurface. Appl Environ Microbiol 72: 192-199. Hoehler, T.M., Bebout, B.M., and Des Marais, D.J. (2001) The role of microbial mats in the production of reduced gases on the early Earth. Nature 412: 324-327. Knoll, A.H. (1996) Breathing room for early animals. Nature 382: 111-112. Kolenbrander, P.E., and London, J. (1993) Adhere today, here tomorrow: oral bacterial adherence. J Bacteriol 175: 3247-3252.

15 Chapter 1

Lewis, L.A., and Mccourt, R.M. (2004) Green algae and the origin of land plants. American Journal of Botany 91: 1535-1556. Mancinelli, R.L. (1996) The nature of nitrogen: an overview. Life Support Biosph Sci 3: 17-24. Rippka, R., Deruelles, J., Waterbury, J.B., Herdman, M., and Stanier, R.Y. (1979) Generic assignments, strain histories and properties of pure cultures of cyanobacteria. J Gen Microbiol 111: 1–61. Sorensen, K.B., Canfield, D.E., Teske, A.P., and Oren, A. (2005) Community composition of a hypersaline endoevaporitic . Appl Environ Microbiol 71: 7352-7365. Stal, L.J. (1995) Physiological ecology of cyanobacteria in microbial mats and other communities. New Phytologist 131: 1-32. Taton, A., Grubisic, S., Brambilla, E., De Wit, R., and Wilmotte, A. (2003) Cyanobacterial diversity in natural and artificial microbial mats of Lake Fryxell (McMurdo Dry Valleys, Antarctica): a morphological and molecular approach. Appl Environ Microbiol 69: 5157-5169. Ward, D.M., Ferris, M.J., Nold, S.C., and Bateson, M.M. (1998) A natural view of microbial biodiversity within hot spring cyanobacterial mat communities. Microbiol Mol Biol Rev 62: 1353-1370. Zehr, J.P., Mellon, M., Braun, S., Litaker, W., Steppe, T., and Paerl, H.W. (1995) Diversity of heterotrophic nitrogen fixation genes in a marine cyanobacterial mat. Appl Environ Microbiol 61: 2527- 2532.

16 2

Phototrophic biofilms and their potential applications

G. Roeselers, M.C.M. van Loosdrecht, and G. Muyzer

Journal of Applied Phycology (in press) Chapter 2

ABSTRACT

Phototrophic biofilms occur on contact surfaces exposed to light in a range of terrestrial and aquatic environments. Oxygenic phototrophs like diatoms, green algae and cyanobacteria are the major primary producers that generate energy and reduce carbon dioxide, providing the system with organic substrates and oxygen. Photosynthesis fuels processes and conversions in the total biofilm community, including the metabolism of heterotrophic organisms. A matrix of polymeric substances secreted by phototrophs and heterotrophs enhances the attachment of the biofilm community. This review discusses the actual and potential applications of phototrophic biofilms in wastewater treatment, bioremediation, fish-feed production, biohydrogen production, and soil improvement and their role in biofouling.

INTRODUCTION

Phototrophic biofilms can best be described as surface attached microbial communities driven by light energy with a photosynthesizing component clearly present. Oxygenic phototrophic microorganisms such as benthic diatoms (centric, pennate, unicellular and filamentous), unicellular and filamentous cyanobacteria, and benthic green algae generate energy and reduce carbon dioxide, providing organic substrates and oxygen. The photosynthetic activity fuels processes and conversions in the total biofilm community. For example, heterotrophs derive their organic C and N requirements from excreted photosynthates and cell lysates, while nutrient regeneration is enhanced by heterotrophs (Bateson and Ward, 1988). The microorganisms produce extracellular polymeric substances (EPS) that hold the biofilm together (Flemming, 1993; Wimpenny et al., 2000). Thick laminated multilayered phototrophic biofilms are usually referred to as microbial mats or phototrophic mats (Guerrero et al., 2002; Roeselers et al., 2007; Stal et al., 1985; Ward et al., 1998). The top layer of microbial mats is typically dominated by oxygenic phototrophs, such as cyanobacteria (Castenholz, 2001a) and diatoms, with underlying or intermixed layers of anoxygenic phototrophs, i.e. green and purple sulfur bacteria (GSB and PSB) (Martinez- Alonso et al., 2005) and chloroflexi like bacteria (Castenholz, 2001b; Ruffroberts et al., 1994). Steep vertical redox and chemical gradients (~ microns to millimeters) that establish in phototrophic biofilms and mats enforce these stratifications in the microbial community (Figure 1). Light intensity decreases with depth, restricting phototrophic activity to the upper layer of the mat. Oxygenic photosynthesis results in a steep oxygen gradient that restricts most anoxygenic phototrophs and anaerobic chemotrophs to the lower parts of the mat. However, recent studies have also shown examples of anaerobes thriving in the oxic zone of microbial mats (Cypionka, 2000; Schaub and Van Gemerden, 1994). The utilization of CO2 during photosynthesis results in a pH gradient (Revsbech et al., 1983). Phototrophic biofilms and mats are formed on surfaces in a range of terrestrial and aquatic environments (Chan et al., 2003; Ferris et al., 1997; Ortega-Morales et al., 2000). The oldest fossilized phototrophic biofilm or mat like structures date back approximately 3.5 billion years (Des Marais, 1990). Currently, there is a growing interest in the application of phototrophic biofilms, for instance in wastewater treatment, (Craig et al., 1996; Schumacher and Sekoulov, 2002; Vymazal et al., 2001), bioremediation (Blanco et al., 1999; Chaillan et al., 2006; Cohen, 2002), aquaculture (Bender and Phillips, 2004; Phillips et al., 1994, van Dam et al., 2002),

18 Phototrophic biofilms and their potential applications

biohydrogen production (Prince and Kheshgi, 2005; Tsygankov et al., 1999), and in the development of antifouling agents (Bhadury and Wright, 2004; Callow and Callow, 2002). The present review will give a brief introduction in various actual and potential biotechnological applications of phototrophic biofilms.

Figure 1. This cross section reveals the stratified structure of a thick cultivated freshwater phototrophic biofilm. The dark top layer consists predominantly of Oscillatoria like cyanobacteria. Scale bar indicates 1 mm.

Wastewater treatment The application of oxygenic phototrophs in the treatment of waste streams that are relatively rich in nutrients and low in organic carbon has many advantages. In a -2 -1 heterotrophic biofilm O2 transfer by diffusion is limited to approximately 20 nmol cm min . However, the areal net oxygen production in an active phototrophic biofilm at a light intensity of 1000 mol photons m-2 s-1 is approximately a factor two higher (Epping and Kuhl, 2000). Hence, the oxygen that is produced by phototrophs can cover a great part of the oxygen demand of bacterial nitrification and the heterotrophic consumption of organic carbon. In adition, oxygenic phototrophs assimilate nutrients for building biomass with carbon dioxide as carbon source. In contrast to wastewater treatment by bacterial nitrification and denitrification, where a large part of the nitrogen escapes as N2 to the atmosphere, the nitrogenous compounds are in this case retained in algal biomass. Cyanobacteria can assimilate nitrogenous compounds like ammonium, nitrate, nitrite, urea and, amino acids. Diazotrophic cyanobacteria can also assimilate atmospheric nitrogen (N2) (Flores and Herrero, 2005). However, the reduction of dinitrogen gas to ammonia by nitrogenases, is a highly endorgenic reaction requiring metabolic energy in the form of ATP. Protection of the sensitive enzyme complex from inactivation by O2, and the replacement of the damaged enzyme add to this high metabolic cost.

19 Chapter 2

Table 1. Nitrogen and phosphorus removal rates obtained with different phototrophic biofilm based wastewater treatment systems. N-removal P- removal System Reference ratea rateb Algal turf scrubber (ATS). 1110 730 Craggs et al, 1996 Rectenwald and Drenner, Periphyton-fish system mesocosm. 108 27 2000 Secondary effluent clarifier 1900 160 Davis et al., 1990 ATS fed with 1% dairy manure. 720 330 Pizarro et al., 2002 Phototrophic biofilms in natural streams 648 117 Davis and Minshall, 1999 aAverage phosphorus removal rates (mg P m-2 day-1), bAverage nitrogen removal rates (mg N m-2 day-1)

The capability of nutrient removal in the absence of organic carbon has often been used for wastewater treatment in algal pond systems (Davis et al., 1990; Garcia et al., 2000). A major disadvantage using suspended algae is the high secondary organic pollution caused by algae biomass in the effluent of the ponds (Racault, 1993). Biomass can be removed by filtration, sedimentation with centrifugation or with decantation but most of these methods are costly. By using immobilized phototrophic biofilms the problem of separation of suspended algal biomass and water can be avoided and the nitrogenous compounds retained in algal biomass can be harvested and used as a in (Schumacher and Sekoulov, 2002). An interesting feature of some cyanobacteria is that they can accumulate inorganic phosphorus and store them internally as polyphosphates (Kromkamp, 1987). However, this aspect has hardly been explored in the context of wastewater treatment. The photosynthetic activity in phototrophic biofilms results in an increasing pH due to the change of the carbon dioxide equilibrium in water. This increase in pH causes precipitation of dissolved phosphates, in addition to phosphorus removal by assimilation. This photosynthesis induced pH increase has also shown potential for the reduction of faecal coliform bacteria in wastewater streams (Schumacher et al., 2003). European Union regulations have led to more stringent effluent standards for sewage treatment facilities located in ecologically sensitive areas. Phototrophic biofilms can be applied for the additional nutrient removal from secondary effluents of wastewater treatment plants. Nutrient removal in so-called constructed wetlands, which show potential for small scale wastewater treatment in developing countries, depends also to a large extend on the activity of epiphytic phototrophic biofilms growing on reed stems (Larsen and Greenway, 2004; Ragusa et al., 2004). Table 1. shows the nitrogen and phosphorus removal rates obtained with several phototrophic biofilm based wastewater polishing systems. For process control and system optimization, it is important to define the right operational conditions. Craggs et al. (1996) described several parameters that determine the efficiency of nutrient removal in an experimental phototrophic biofilm system. An important parameter is the applied flow velocity. At higher flow velocities there is a trade-off between reduced colonization and shear stress versus increased metabolism in the established biofilms by reduced boundary layers and increased water mixing. Two parameters that are interconnected are water depth and light intensity. Algal production generally declines with water depth because increased water depths result in reduced light penetration (Craggs et al., 1996; Havens et al., 1996). However, in shallow

20 Phototrophic biofilms and their potential applications

water with a limited flow-velocity, the reduced debit can lead to conditions were the biofilm becomes nutrient limited instead of light limited. Hence, optimal water depths will also depend on nutrient loads and local or seasonal light conditions.

Figure 2. Discharge of the municipal wastewater treatment facility (WWTF) of Sint Maartensdijk (the Netherlands) is polished in a system (I= inlet, O = outlet). Nitrate is primarily assimilated by epiphytic phototrophic biofilms growing on reed stems.

Removal of heavy metals Biosorption consists of several mechanisms, mainly ion exchange, chelation, adsorption, and diffusion through cell walls. These “passive” mechanisms can take place at the cellular level and at the microbial community level. The active mode of metal uptake and concentration is called bioaccumulation. This process is dependent on the cellular metabolism. Many oxygenic phototrophic microorganisms have the capability to sorb or accumulate metals in one way or another, and there is considerable potential for the application of algal biofilms in the detoxification of wastewaters polluted with heavy metals (Bender et al., 1994; Mehta and Gaur, 2005). Extracellular polysaccharides that are negatively charged at elevated pH levels generated by oxygenic photosynthesis, may account for the metal-binding properties of phototrophic biofilms (Bender et al., 1994;

21 Chapter 2

Bhaskar and Bhosle, 2006; Liu et al., 2001; Wang et al., 1998). Parker et al. (2000) showed that mucilage sheaths isolated from the cyanobacteria Microcystis aeruginosa and Aphanothece halophytica exhibit strong affinity for heavy metal ions such as copper, lead and zinc. In addition to biosorption and bioaccumulation, the elevated pH inside photosyntetically active biofilms may favor removal of metals by precipitation (Liehr et al., 1994). Major advantages of metal removal by biosorption include, low cost, and high efficiency of heavy metal removal from diluted solutions. However, in order not to simply displace heavy metal pollution, methods will have to be developed to extract heavy metals easily from biomass (Kratochvil and Volesky, 1998). Water hardness is a crucial factor that influences metal uptake efficiency because cations such as Ca2+ and Mg2+ compete with trace metals for binding sites on cell membranes and extracellular polysaccharides (Fortin et al., 2007). Meylan et al., (2003) showed that different concentrations of dissolved manganese affected the intracellular accumulation of zinc and copper by phototrophic biofilms. In addition to cation concentrations, metal uptake is affected by light intensity, pH, biofilm density, the presence of metal binding humic substances, and the tolerance of individual algal species to specific heavy metals. (Fortin et al., 2007; Vymazal, 1984)

Oil degradation The volumes of petroleum-based products transported across the world are enormous and the risk of oil spillage is significant. The volume of spills usually exceeds the inherent remediation capacity for any given environment, resulting in a significant ecological impact (Cohen, 2002). It has been suggested that microbial mats can play a role in the biodegradation of oil. In the years after the massive oil spills during the first Gulf war of 1991, it was observed that dense mats of cyanobacteria formed on contaminated beaches (Sorkhoh et al., 1992). It has been shown that in particular Oscillatoria spp. are able to cope with heavy oil pollution (Abed et al., 2006; Cohen, 2002; van Bleijswijk and Muyzer, 2004). Although there is no direct evidence that cyanobacteria are directly involved in the degradation of petroleum products, they probably facilitate degradation by sulfate-reducing bacteria (Edwards et al., 1992) and aerobic heterotrophs (Benthien et al., 2004; Cohen, 2002; Sorkhoh et al., 1995). Previous studies have shown that the addition of nitrogen supplements enhances microbial assimilation of carbon from oil (Coffin et al., 1997).

Cyanobacterial N2 fixation could provide sufficient nitrogen compounds for heterotrophic oil degradation. Free radicals formed during oxygenic photosynthesis could indirectly enhance photochemical oil degradation (Nicodem et al., 1997). Microcosm studies examining the initial response of phototrophic biofilms exposed to petrochemical compounds revealed signs of acute toxicity (Nayar et al. 2004). Phototrophic biofilms are ubiquitous and dominant primary producers forming the base of aquatic food webs. Therefore, it has been suggested that phototrophic biofilms are applicable as sensitive bioindicators of petrochemical pollution and for ecotoxicology tests (Nayar et al. 2004).

Agriculture When phototrophic biofilms are used for polishing of nitrate or ammonium containing wastewater streams, the nitrogen that is retained in biomass can be used as a in agriculture. Biomass applied in the remediation of waste streams containing hazardous metals or recalcitrant organic pollutants is not directly applicable as a fertilizer.

22 Phototrophic biofilms and their potential applications

Cyanobacteria can also be applied for in situ soil fertilization via N2 fixation. Much work has been done on the fertilization of rice paddy fields with nitrogen fixing cyanobacteria (Ariosa et al., 2004; Habte and Alexander, 1980; Lem and Glick, 1985). In addition, EPS produced by algae and cyanobacteria can improve the soil water- holding capacity and prevent erosion (Barclay and Lewin, 1985; Rao and Burns, 1990). Mazor et al. (1996) showed that addition of 0.5 mg of Microcoleus sp. EPS per g sand retained approximately 30% of the water-holding capacity of the sand after 24 h of desiccation at 55°C while sand samples without EPS dried out completely. Elevated soil salinity, which is increasing world wide, has a major impact on soil quality and agricultural production. In many coastal areas, salinity is an inherent situation, but inefficient water management, i.e. excess recharging of groundwater and accumulation through concentration often leads to secondary salinization of farmlands. Early 1950 a biological approach to the problem of saline soils using cyanobacteria was proposed (Singh, 1950). It has been shown that inoculation of soil surfaces with a suspension of halotolerant cyanobacteria leads to a salinity reduction (Apte and Thomas, 1997; Kaushik and Venkataraman, 1982). This amelioration of soil salinity is probably caused by a temporal entrapment of Na+ ions in cyanobacterial EPS sheaths, resulting in a restricted Na+ influx in the plant roots (Ashraf et al., 2006). Permanent removal of Na+ from the soil may not be possible, since Na+ is released back into the soil subsequent to the death and decay of the cyanobacteria.

Aquaculture Effluent discharges of intensive fish production systems may cause significant nutrient pollution. Fish farmers have a stake in regulating nutrient pollution, because poor water quality can reduce aquaculture productivity. On a small scale phototrophic biofilm based systems can be used to reduce ammonia and nitrate concentrations in aquaculture effluents (Bender and Phillips, 2004). In fish aquacultures, commercial feeds, consisting mostly of fishmeal and oil, may account for more than 50% of the total production costs (Elsayed and Teshima, 1991). Only about 15-30% of the nutrient input in feed-driven pond systems is converted into harvestable products (Gross et al. 2000). Therefore, there is a growing interest for substitution of commercial feeds with alternative protein sources. The cost savings and the reduction of ecological impact by using phototrophic biofilms for fish feed production and the possible simultaneous effluent treatment may be significant (Elsayed and Teshima, 1991; Naylor et al., 2000; Phillips et al., 1994). Tilapia (Oreochromis niloticus) can consume microalgae as a major, even exclusive source of its feed requirements. Due to their omnivorous diet and rapid growth, species of tilapia are highly suitable for aquaculturing in fertized pond systems. In fertilized ponds, organic and inorganic fertilizers are used to increase productivity. Nutrients are incorporated into algal biomass and, through a complex food web ultimately incorporated into fish biomass. Azim et al. (2003) showed that by adding substrate for biofilm adherence to fertilized aquaculture ponds; the conversion of nutrients into harvestable products could be optimized. Tilapia growth was significantly higher and nitrogen retention doubled in substrate ponds compared with control ponds. The potential of fish production based on phototrophic biofilms was reviewed in detail by van Dam et al. (2002). Cyanobacteria that produce toxic secondary metabolites may cause problems to the expanding aquaculture industry. E.g. Aphanizomenon, and Microcystis like species produce so-called microcystins that can accumulate in fish tissue used for human

23 Chapter 2 consumption (Magalhaes et al., 2001; Wiegand and Pflugmacher, 2005). The occurrence of toxins has often been related to rapid planctonic cyanobacterial biomass development during algal blooms but this link is much less common in benthic assemblages (Blaha et al., 2004).

Biohydrogen Hydrogen is a clean alternative to fossil fuels as its combustion only generates water as a byproduct. Biological production of hydrogen could provide a renewable source of energy. Cyanobacteria are highly promising microorganisms for biological photohydrogen production. Two cyanobacterial enzymes are capable of hydrogen production. The bidirectional hydrogenase complex can either produce or oxidize H2 in the presence of suitable electron donor or acceptor. The physiological role of bidirectional hydrogenases is still unclear and the enzyme is absent in a significant number of strains. Hydrogen evolution is also catalyzed by nitrogenases (Mancinelli, 1996; Tamagnini et al., 2002; Zehr and Turner, 2001). The nitrogenase machinery releases at least one mol

H2 per mol N2 reduced to ammonia, which represents a significant loss of energy. However, most diazotrophic cyanobacteria posses an enzyme called uptake hydrogenase that serves to recycle some of the electrons lost in the form of H2. Several studies with bioreactors have demonstrated the feasibility of cyanobacterial biohydrogen production (Lindberg et al., 2004;

Schutz et al., 2004; Tsygankov et al., 1999). As mentioned, N2 reduction has a high ATP requirement and this reduces the potential conversion of solar energy considerably. An advantageous aspect of this is that ATP hydrolysis provides a relatively strong thermodynamic driving force pushing hydrogen evolution, which is not true for bidirectional hydrogenases (Prince and Kheshgi, 2005). This allows the generation of a higher partial H2 pressure in potential bioreactors. Ideally, hydrogen producing cyanobacteria should invest a minimal amount of ATP in growth, have a high metabolism and should be restricted in place. Cyanobacterial biofilms or mats are of great interest because they meet up to these requirements. The efficiency of hydrogen production could be increased by constructing defined biofilm assemblages containing a range of desired cyanobacterial species or genetically modified strains with a reduced uptake hydrogenase activity (Lindberg et al., 2004; Tsygankov et al., 1999). In order to select an “optimal genetic background” for the construction of genetically engineered cyanobacteria, future studies should focus on the natural molecular variation of strains that have the potential to produce hydrogen. Currently, this technology is in its infancy and as yet not ready for commercial adaptation and exploitation.

Anoxygenic phototrophs and sulfide removal Sulfide-containing waste streams are usually treated in reactors by chemotrophic sulfide-oxidizing microorganisms using either oxygen or nitrate as the ultimate electron acceptor. In many bioreactors, sulfide is transformed into sulfate by aerobic sulfur-oxidizing bacteria, such as different species of the genus Thiobacillus. A disadvantage of using aerobic sulfur-oxidizing bacteria is that sulfide removal cannot be combined with sewage treatment by anaerobic digestion. Sulfide oxidation has to take place in a separate reactor in order to avoid exposure of strictly anaerobic methanogens to inhibitory levels of oxygen. Hence, anaerobic oxidation by phototrophic sulfur bacteria (Chloroflexi, GSB and PSB) has been proposed as an alternative method for sulfide removal (Kim et al., 1990). There is a particular interest in reactors using immobilized biofilms because suspended microbial biomass is easily washed out from the system whenever growth rates are disturbed. Several laboratory scale studies with monospecies and multispecies biofilms

24 Phototrophic biofilms and their potential applications

of anoxygenic phototrophic bacteria showed promising results (Ferrera et al., 2004; Kobayashi et al., 1983; Syed and Henshaw, 2003). Only very few processes based on substratum-irradiated biofilms have been employed for large scale treatment of sulfide-containing waste streams (Hurse and Keller, 2004; Jensen and Webb, 1995).

Phototrophic biofilm nuisances From a human point of view, phototrophic biofilms can develop in wrong places and at the wrong times (Flemming, 2002). Biofouling is known to cause widespread problems in industrial fluid processing applications. Since phototrophic biofilms require light they can cause problems at surfaces exposed to daylight. Phototrophic biofouling causes technical failure or damage to power plant cooling systems, aquaculture systems, fishing nets, ship hulls, marine infrastructures and historical buildings (Ortega-Morales et al., 2000). It constitutes a serious problem for seagoing ships due to increased resistance and fuel costs that can rise by as much as 40 percent (Townsin, 2003). In order to prevent or control phototrophic biofouling, efforts have been made to develop anti-fouling agents. The most common compounds in anti-fouling agents are tri-butyl-tin (TBT). TBT is highly effective but it is also toxic to non-target organisms and it is not biodegradable. Since, the use of the TBT based coatings will be completely banned by January 2008, there is an urgent need for sustainable alternatives. Surprisingly, some marine algae could serve in the prevention of biofouling. Certain cyanobacteria and eukaryotic algae produce biogenic compounds such as lipopeptides, amides, alkaloids, terpenoids, lactones, pyrroles and steroids with antibacterial, anti-algal, and antifungal properties, which could be applied in the development of environmentally friendly anti-fouling agents (Bhadury and Wright, 2004; Xu et al., 2005). Successional community changes during the colonization of new habitats or after environmental disturbances have been widely studied in the macro-ecology, but also for communities of planktonic and benthic microalgae (Helbling et al., 2005; Schäfer et al., 2001), and more recently in bacterial communities (Ferris et al., 1997; Martiny et al., 2003; Roeselers et al., 2004). Previous studies have shown that heterotrophic bacteria play an important role as early colonizers during phototrophic biofilm development (Chan et al., 2003; Roeselers et al., 2007). A valuable strategy for the development of anti-fouling agents could lie in the identification and characterization of pioneer microorganisms responsible for the initial surface colonization that leads to biofilm formation (Zhang et al., 2006). A critical remark at this point could be that the despite of the promising results presented by scientists, yet there has hardly been any development of commercially applicable alternatives for TBT.

Conclusions and future perspectives The field of microbiology has come to accept the universality of biofilms. Researchers in the fields of clinical, food and water, and environmental microbiology have begun to investigate microbiologic processes from a biofilm perspective. There are multiple examples of genotypic and physiological differences between microorganisms growing planktonic or in biofilms. Until recently, applied phycological studies have focused mainly on the planktonic mode of life. In this review we have discussed potential and actual applications of phototrophic biofilm biotechnology in the development of clean energy systems, wastewater treatment, bioremediation, fish feed production and soil fertilization.

25 Chapter 2

In the context of potential applications, the most important features of these biofilm systems are versatility and adaptability, i.e. they have a broad spectrum of capabilities. This makes it possible to link different end uses within the same process; for example nitrate and phosphate removal combined with production of fish feeds (Bender and Phillips, 2004). The complexity in terms of species richness is an important aspect determining the metabolic biodiversity, and adaptability of phototrophic biofilms (Boles et al. 2004; Girvan et al. 2005). In addition, applications of pure culture or defined community biofilms seem less attractive due to the high costs associated with control of the culture performance and equipment sterilization and isolation to prevent contaminations. Hence, the focus of application development should be on the use of open mixed culture systems. Therefore, a clear understanding of the ecology of phototropic biofilm communities is essential in order to optimize their cultivation for specific biotechnological applications. Depending on the application it can be essential to select for phototropic biofilms containing specific species and strains, e.g. strains with a high polyunsaturated fatty acid content, strains that do or do not secrete harmful secondary metabolites, strains producing EPS with high metal sorption capacities, or strains with a high tolerance for petroleum-based compounds. The efficiency and reliability of phototrophic biofilm applications depend to a great extent on the possibility to select and maintain desired community compositions. Future studies should focus on successional changes during biofilm development (Chan et al., 2003; Roeselers et al., 2007), susceptibility to viral and grazing pressure (Simek and Chrzanowski, 1992; Thingstad, 2000), and the mechanisms that determine the structural and functional responses to abrupt perpetuations, and seasonal fluctuations on community composition and productivity (Kaufman, 1982). Molecular ecological techniques that allow detailed "in situ" characterization of community compositions and activities provide an important tool for future research. Non- culture based molecular methods such as DGGE, clone library analysis, quantitative PCR and stable isotope probing can be used to obtain the phylogeny, relative abundance and genetic activity of individual members of a biofilm community (Omoregie et al., 2004; Roeselers et al., 2006; Steunou et al., 2006). In particular functional genomics approaches will offer important clues about phototrophic biofilm biology. An important consideration for applications based on phototrophic activity in general is that they require much surface. These systems are primarily fueled by light, which differs from all other resources because it cannot be mixed. The unidirectional nature of photons requires that biofilms are cultivated on surfaces exposed to direct solar radiation. Therefore, high land prices could be a major hurdle for applications in for instance the treatment of municipal wastewater in densely populated areas. Combining different end uses within processes could compensate for the cost of these relatively large footprints. Phototrophic biofilms would also be suitable for the development of inexpensive treatment methods for developing countries, where land values are relatively low and where the bulk of domestic and industrial wastewater is still discharged without any treatment.

ACKNOWLEDGMENTS

This work was supported by the project PHOBIA (QLK3-CT-2002-01938) funded under European Union Framework V. We thank dr. Henk Jonkers for careful reading of the manuscript.

26 Phototrophic biofilms and their potential applications

REFERENCES

Abed, R. M., Al-Thukair, A. and de Beer, D. (2006). Bacterial diversity of a cyanobacterial mat degrading petroleum compounds at elevated salinities and temperatures. FEMS Microbiol Ecol 57: 290- 301. Apte, S. K. and Thomas, J. (1997). Possible amelioration of coastal soil salinity using halotolerant nitrogen-fixing cyanobacteria. Plant and Soil 189: 205-211. Ariosa, Y., Quesada, A., Aburto, J., Carrasco, D., Carreres, R., Leganes, F. and Fernandez Valiente, E. (2004). Epiphytic cyanobacteria on Chara vulgaris are the main contributors to N(2) fixation in rice fields. Appl Environ Microbiol 70: 5391-5397. Ashraf, M., Hasnain, S. and Berge, O. (2006). Effect of exo-polysaccharides producing bacterial inoculation on growth of roots of wheat (Triticum aestivum) plants grown in a salt-affected soil. IJEST 3: 43-51. Azim, M.E., Verdegem, M.C.J., Singh, M., van Dam, A.A., and Beveridge, M.C.M. (2003) The effects of periphyton substrate and fish density on water quality, phytoplankton, periphyton and fish growth. Aquac Res 34:685-695. Barclay, W. R. and Lewin, R. A. (1985). Microalgal polysaccharide production for the conditioning of agricultural soils. Plant and Soil 88: 159-169. Bateson, M. M. and Ward, D. M. (1988). Photoexcretion and fate of glycolate in a hot spring cyanobacterial mat. Appl Environ Microbiol 54: 1738-1743. Bender, J., Rodriguez-Eaton, S., Ekanemesang, U. M. and Phillips, P. (1994). Characterization of metal-binding bioflocculants produced by the cyanobacterial component of mixed microbial mats. Appl Environ Microbiol 60: 2311-2315. Bender, J. and Phillips, P. (2004). Microbial mats for multiple applications in aquaculture and bioremediation. Bioresour Technol 94: 229-238. Benthien, M., Wieland, A., García de Oteyza, T., Grimalt, J. O. and Kühl, M. (2004). Oil-contamination effects on a hypersaline microbial mat community (Camargue, France) as studied with microsensors and geochemical analysis. OPHELIA 58: 135-150. Bhadury, P. and Wright, P. C. (2004). Exploitation of marine algae: biogenic compounds for potential antifouling applications. Planta 219: 561-578. Bhaskar, P. V. and Bhosle, N. B. (2006). Bacterial extracellular polymeric substance (EPS): a carrier of heavy metals in the marine food-chain. Environ Int 32: 191-198. Blaha, L., Sabater, S., Babica, P., Vilalta, E. and Marsalek, B. (2004). Geosmin occurrence in riverine cyanobacterial mats: is it causing a significant health hazard? Water Sci Technol 49: 307-312. Blanco, A., Sanz, B., Llama, M. J. and Serra, J. L. (1999). Biosorption of heavy metals to immobilised Phormidium laminosum biomass. J Biotechnol 69: 227–240. Boles, B.R., Thoendel, M., and Singh, P.K. (2004) Self-generated diversity produces "insurance effects" in biofilm communities. Proc Natl Acad Sci U S A 101: 16630-16635 Callow, M. E. and Callow, J. E. (2002). Marine biofouling: a sticky problem. Biologist (London) 49: 10- 14. Castenholz, R. W. (2001a). Phylum BX. Cyanobacteria. Oxygenic photosynthetic bacteria. In Bergey's Manual of Systematic Bacteriology, pp. 474-487. Edited by D. R. Boone, R. W. Castenholz and G. M. Garrity. New York, USA: Springer. Castenholz, R. W. (2001b). Class I: Chloroflexi. In Bergey's Manual of Systematic Bacteriology, pp. 427. Edited by D. R. Boone, R. W. Castenholz and G. M. Garrity. New York: Springer-Verlag. Chaillan, F., Gugger, M., Saliot, A., Coute, A. and Oudot, J. (2006). Role of cyanobacteria in the biodegradation of crude oil by a tropical cyanobacterial mat. Chemosphere 62: 1574-1582. Chan, B. K., Chan, W. K. and Walker, G. (2003). Patterns of biofilm succession on a sheltered rocky shore in Hong Kong. Biofouling 19: 371-380. Coffin, R. B., Cifuentes, L. A. and Pritchard, P. H. (1997). Assimilation of oil-derived carbon and remedial nitrogen applications by intertidal food chains on a contaminated beach in Prince William Sound, Alaska. Mar Environ Res 44: 27-39. Cohen, Y. (2002). Bioremediation of oil by marine microbial mats. Int Microbiol 5: 189-193. Craggs, R. J., Adey, W. H., Jessup, B. K. and Oswald, W. J. (1996). A controlled stream mesocosm for tertiary treatment of sewage. Ecol Eng 6: 149-169.

27 Chapter 2

Craig, R. J., Adey, W. H., Jenson, K. R., St. John, M. S., Green, F. B. and Oswald, J. (1996). Phosphorus removal from wastewater using an algal turf scrubber. Water Sci Technol 33: 191–198. Cypionka, H. (2000). Oxygen respiration by desulfovibrio species. Annu Rev Microbiol 54: 827-848. Davis, L. S., Hoffmann, J. P. and Cook, P. W. (1990). Production and Nutrient Accumulation by Periphyton in a Waste-Water Treatment Facility. Journal of Phycology 26: 617-623. Davis, J.C., and Minshall, G.W. (1999) Nitrogen and phosphorus uptake in two Idaho (USA) headwater wilderness streams. Oecologia 119:247-255. Des Marais, D. J. (1990). Microbial mats and the early evolution of life. Trends Ecol Evol 5: 140-144. Edwards, E. A., Wills, L. E., Reinhard, M. and Grbic-Galic, D. (1992). Anaerobic degradation of toluene and xylene by aquifer microorganisms under sulfate-reducing conditions. Appl Environ Microbiol 58: 794-800. Elsayed, A. F. M. and Teshima, S. I. (1991). Tilapia nutrition in aquaculture. Rev Aquat Sci 5: 247-265. Epping, E. and Kuhl, M. (2000). The responses of photosynthesis and oxygen consumption to short- term changes in temperature and irradiance in a cyanobacterial mat (Ebro Delta, Spain). Environ Microbiol 2: 465-474. Ferrera, I., Sanchez, O. and Mas, J. (2004). A new non-aerated illuminated packed-column reactor for the development of sulfide-oxidizing biofilms. Appl Microbiol Biotechnol 64: 659-664. Ferris, M. J., Nold, S. C., Revsbech, N. P. and Ward, D. M. (1997). Population structure and physiological changes within a hot spring microbial mat community following disturbance. Appl Environ Microbiol 63: 1367-1374. Flemming, H. C. (1993). Biofilms and Environmental-Protection. Water Sci Technol 27: 1-10. Flemming, H. C. (2002). Biofouling in water systems--cases, causes and countermeasures. Appl Microbiol Biotechnol 59: 629-640. Flores, E. and Herrero, A. (2005). Nitrogen assimilation and nitrogen control in cyanobacteria. Biochem Soc Trans 33: 164-167. Fortin, C., Denison, F.H. and Garnier-Laplace, J. (2007) Metal-phytoplankton interactions: Modeling the effect of competing ions (H+, Ca2+, and Mg2+) on uranium uptake. Environmental Toxicology and Chemistry 26: 242-248. Garcia, J., Mujeriego, R. and Hernandez-Marine, M. (2000). High rate algal pond operating strategies for urban wastewater nitrogen removal. J of Appl Phycol 12: 331-339. Girvan, M.S., Campbell, C.D., Killham, K., Prosser, J.I., and Glover, L.A. (2005) Bacterial diversity promotes community stability and functional resilience after perturbation. Environ Microbiol 7: 301-313 Gross, A., Boyd, C.E., and Wood, C.W. (2000) Nitrogen transformations and balance in channel catfish ponds. Aquacultural Engineering 24: 1-14 Guerrero, R., Piqueras, M. and Berlanga, M. (2002). Microbial mats and the search for minimal ecosystems. Int Microbiol 5: 177-188. Habte, M. and Alexander, M. (1980). Nitrogen fixation by photosynthetic bacteria in lowland rice culture. Appl Environ Microbiol 39: 342-347. Havens, K. E., East, T. L., Meeker, R. H., Davis, W. P. and Steinman, A. D. (1996). Phytoplankton and periphyton responses to in situ experimental nutrient enrichment in a shallow subtropical lake. J Plankton Res 18: 551-566. Helbling, E. W., Barbieri, E. S., Marcoval, M. A., Goncalves, R. J. and Villafane, V. E. (2005). Impact of solar ultraviolet radiation on marine phytoplankton of Patagonia, Argentina. Photochem Photobiol 81: 807-818. Hurse, T. J. and Keller, J. (2004). Reconsidering the use of photosynthetic bacteria for removal of sulfide from wastewater. Biotechnol Bioeng 85: 47-55. Jensen, A. B. and Webb, C. (1995). Treatment of H2s-Containing Gases - a Review of Microbiological Alternatives. Enzyme Microb Technol 17: 2-10. Kaushik, B. D. and Venkataraman, G. S. (1982). Reclamative capacity of blue-green algae in saline and sodic soils. In Proceedings of the National Symposium on Biological Nitrogen Fixation, pp. 378–389. Bombay: Department of Atomic Energy. Kaufman, L.H. (1982) Stream aufwuchs accumulation - disturbance frequency and stress resistance and resilience. Oecologia 52: 57-63 Kim, B. W., Kim, I. K. and Chang, H. N. (1990). Bioconversion of hydrogen sulfide by free and immobilized cells of Chlorobium thiosulfatophilum. Biotechnol Lett 12: 381–386.

28 Phototrophic biofilms and their potential applications

Kobayashi, H. A., Stenstrom, M. and Mah, R. A. (1983). Use of Photosynthetic Bacteria for Hydrogen- Sulfide Removal from Anaerobic Waste Treatment Effluent. Water Research 17: 579-587. Kratochvil, D. and Volesky, B. (1998). Biosorption of Cu from ferruginous wastewater by algal biomass. Water Research 32: 2760-2768. Kromkamp, J. (1987). Formation and Functional-Significance of Storage Products in Cyanobacteria. N Z J Mar Freshwater Res 21: 457-465. Larsen, E. and Greenway, M. (2004). Quantification of biofilms in a sub-surface flow wetland and their role in nutrient removal. Water Sci Technol 49: 115-122. Lem, N. W. and Glick, B. R. (1985). Biotechnological uses of cyanobacteria. Biotechnol Adv 3: 195-208. Liehr, S. K., Chen, H. J. and Lin, S. H. (1994). Metals removal by algal biofilms. Water Sci Technol 30: 59-68. Lindberg, P., Lindblad, P. and Cournac, L. (2004). Gas exchange in the filamentous cyanobacterium Nostoc punctiforme strain ATCC 29133 and Its hydrogenase-deficient mutant strain NHM5. Appl Environ Microbiol 70: 2137-2145. Liu, Y., Lam, M. C. and Fang, H. H. (2001). Adsorption of heavy metals by EPS of activated sludge. Water Sci Technol 43: 59-66. Magalhaes, V. F., Soares, R. M. and Azevedo, S. M. (2001). Microcystin contamination in fish from the Jacarepagua Lagoon (Rio de Janeiro, Brazil): ecological implication and human health risk. Toxicon 39: 1077-1085. Mancinelli, R. L. (1996). The nature of nitrogen: an overview. Life Support Biosph Sci 3: 17-24. Martinez-Alonso, M., Van Bleijswijk, J., Gaju, N. and Muyzer, G. (2005). Diversity of anoxygenic phototrophic sulfur bacteria in the microbial mats of the Ebro Delta: a combined morphological and molecular approach. FEMS Microbiol Ecol 52: 339-350. Martiny, A. C., Jorgensen, T. M., Albrechtsen, H. J., Arvin, E. and Molin, S. (2003). Long-term succession of structure and diversity of a biofilm formed in a model drinking water distribution system. Appl Environ Microbiol 69: 6899-6907. Mazor, G., Kidron, G. J., Vonshak, A. and Abeliovich, A. (1996). The role of cyanobacterial exopolysaccharides in structuring desert microbial crusts. FEMS Microb Ecol 21: 121-130. Mehta, S. K. and Gaur, J. P. (2005). Use of algae for removing heavy metal ions from wastewater: progress and prospects. Crit Rev Biotechnol 25: 113-152. Meylan, S., Behra, R., and Sigg, L. (2003) Accumulation of copper and zinc in periphyton in response to dynamic variations of metal speciation in freshwater. Environ Sci Technol 37:5204-5212. Naylor, R. L., Goldburg, R. J., Primavera, J. H. and other authors (2000). Effect of aquaculture on world fish supplies. Nature 405: 1017-1024. Nayar, S., Goh, B.P.L., and Chou, L.M. (2004) The impact of petroleum hydrocarbons (diesel) on periphyton in an impacted tropical estuary based on in situ microcosms. J Exp Mar Bio Ecol 302:213- 232 Nicodem, D. E., Fernandes, M. C. Z., Guedes, C. L. B. and Correa, R. J. (1997). Photochemical processes and the environmental impact of petroleum spills. Biogeochemistry 39: 121-138. Omoregie, E. O., Crumbliss, L. L., Bebout, B. M. and Zehr, J. P. (2004). Determination of nitrogen- fixing phylotypes in Lyngbya sp. and Microcoleus chthonoplastes cyanobacterial mats from Guerrero Negro, Baja California, Mexico. Appl Environ Microbiol 70: 2119-2128. Ortega-Morales, O., Guezennec, J., Hernandez-Duque, G., Gaylarde, C. C. and Gaylarde, P. M. (2000). Phototrophic biofilms on ancient Mayan buildings in Yucatan, Mexico. Curr Microbiol 40: 81-85. Parker, D. L., Mihalick, J. E., Plude, J. L., Plude, M. J., Clark, T. P., Egan, L., Flom, J. J., Rai, L. C. and Kumar, H. D. (2000). Sorption of metals by extracellular polymers from the cyanobacterium Microcystis aeruginosa f. flos-aquae strain C3-40. J Appl Phycol 12: 219-224. Pizarro, C., Kebede-Westhead, E., and Mulbry, W. (2002) Nitrogen and phosphorus removal rates using small algal turfs grown with dairy manure. J Appl Phycol 14:469–473. Phillips, P., Russell, A., Bender, J. and Muñoz, R. (1994). Management plan for utilization of a floating microbial mat with its associated detrital gelatinous layer as a complete tilapia Oreochromis niloticus feed system. Bioresource Technol 47: 239-245. Prince, R. C. and Kheshgi, H. S. (2005). The photobiological production of hydrogen: potential efficiency and effectiveness as a renewable fuel. Crit Rev Microbiol 31: 19-31.

29 Chapter 2

Racault, Y. (1993). Pond malfunction: case study of three plants in the south-west of France. Water Sci Technol 28: 183–192. Ragusa, S. R., McNevin, D., Qasem, S. and Mitchell, C. (2004). Indicators of biofilm development and activity in constructed wetlands microcosms. Water Res 38: 2865-2873. Rao, D. L. N. and Burns, R. G. (1990). The effect of surface growth of blue-green-algae and bryophytes on some microbiological, biochemical, and physical soil properties. Biol Fertil Soils 9: 239-244. Rectenwald, L.L., and Drenner, R. W. (2000) Nutrient removal from wastewater effluent using an ecological water treatment system. Environ Sci Technol 34:522-526. Revsbech, N. P., Jorgensen, B. B., Blackburn, T. H. and Cohen, Y. (1983). Microelectrode Studies of the Photosynthesis and O-2, H2s, and Ph Profiles of a Microbial Mat. Limnol Oceanogr 28: 1062-1074. Roeselers, G., van Loosdrecht, M. C. M. and Muyzer, G. (2004). Biodiversity of phototrophic biofilms. In European Symposium on Environmental Biotechnology; ESEB 2004, pp. 161-164. Roeselers, G., Zippel, B., Staal, M., van Loosdrecht, M. and Muyzer, G. (2006). On the reproducibility of microcosm experiments - different community composition in parallel phototrophic biofilm microcosms. FEMS Microbiol Ecol 58: 169-178. Roeselers, G., Norris, T., Castenholz, R., Rysgaard, S., Glud, R., Kuhl, M. and Muyzer, G. (2007). Diversity of phototrophic bacteria in microbial mats from Arctic hot springs (Greenland). Environ Microbiol 9:26-38 Roeselers, G., van Loosdrecht, M. C. M. and Muyzer, G. (2007). Heterotrophic pioneers facilitate phototrophic biofilm development. Microb Ecol. DOI: 10.1007/s00248-007-9238-x Ruffroberts, A. L., Kuenen, J. G. and Ward, D. M. (1994). Distribution of cultivated and uncultivated cyanobacteria and Chloroflexus-like bacteria in hot-spring microbial mats. Appl Environ Microbiol 60: 697-704. Schäfer, H., Bernard, L., Courties, C. and other authors (2001). Microbial community dynamics in Mediterranean nutrient-enriched seawater mesocosms: changes in the genetic diversity of bacterial populations. FEMS Microbiol Ecol 34: 243-253. Schaub, B. E. M. and Vangemerden, H. (1994). Simultaneous phototrophic and chemotropic growth in the purple sulfur bacterium Thiocapsa roseopersicina M1. FEMS Microbiol Ecol 13: 185-195. Schumacher, G. and Sekoulov, I. (2002). Polishing of secondary effluent by an algal biofilm process. Water Sci Technol 46: 83-90. Schumacher, G., Blume, T. and Sekoulov, I. (2003). Bacteria reduction and nutrient removal in small wastewater treatment plants by an algal biofilm. Water Sci Technol 47: 195-202. Schutz, K., Happe, T., Troshina, O., Lindblad, P., Leitao, E., Oliveira, P. and Tamagnini, P. (2004). Cyanobacterial H(2) production - a comparative analysis. Planta 218: 350-359. Simek, K., and Chrzanowski, T.H. (1992) Direct and indirect evidence of size-selective grazing on pelagic bacteria by fresh-water nanoflagellates. Appl Environ Microbiol 58: 3715-3720 Singh, R. N. (1950). Reclamation of 'Usar' lands in India through blue-green algæ. Nature 165: 325- 326. Sorkhoh, N., Al-Hasan, R., Radwan, S. and Höpner, T. (1992). Self-cleaning of the Gulf. Nature 359: 109. Sorkhoh, N. A., al-Hasan, R. H., Khanafer, M. and Radwan, S. S. (1995). Establishment of oil- degrading bacteria associated with cyanobacteria in oil-polluted soil. J Appl Bacteriol 78: 194-199. Stal L. J., Vangemerden, H. and Krumbein, W. E. (1985). Structure and Development of a Benthic Marine Microbial Mat. FEMS Microbiol Ecol 31: 111-125. Steunou, A. S., Bhaya, D., Bateson, M. M., Melendrez, M. C., Ward, D. M., Brecht, E., Peters, J. W., Kuhl, M. and Grossman, A. R. (2006). In situ analysis of nitrogen fixation and metabolic switching in unicellular thermophilic cyanobacteria inhabiting hot spring microbial mats. Proc Natl Acad Sci U S A 103: 2398-2403. Syed, M. A. and Henshaw, P. F. (2003). Effect of tube size on performance of a fixed-film tubular bioreactor for conversion of hydrogen sulfide to elemental sulfur. Water Res 37: 1932-1938. Tamagnini, P., Axelsson, R., Lindberg, P., Oxelfelt, F., Wunschiers, R. and Lindblad, P. (2002). Hydrogenases and hydrogen metabolism of cyanobacteria. Microbiol Mol Biol Rev 66: 1-20 Thingstad, T.F. (2000) Elements of a theory for the mechanisms controlling abundance, diversity, and biogeochemical role of lytic bacterial viruses in aquatic systems. Limnol Oceanogr 45: 1320-1328 Townsin, R. L. (2003). The ship hull fouling penalty. Biofouling 19 Suppl: 9-15.

30 Phototrophic biofilms and their potential applications

Tsygankov, A. A., Borodin, V. B., Rao, K. K., and Hall, D. O. (1999). H(2) photoproduction by batch culture of Anabaena variabilis ATCC 29413 and its mutant PK84 in a photobioreactor. Biotechnol Bioeng 64: 709-715. van Bleijswijk, J., and Muyzer, G. (2004). genetic diversity of oxygenic phototrophs in microbial mats exposed to different levels of oil pollution. OPHELIA 58: 157-164. van Dam, A.A., Beveridge, M.C.M., Azim, M.E., and Verdegem, M.C.J. (2002) The potential of fish production based on periphyton. Rev Fish Biol Fish 12: 1-31 Vymazal, J. (1984) Short-term uptake of heavy metals by periphyton algae. Hydrobiologia 119: 171- 179. Vymazal, J., Sladedek, V. and Stach, J. (2001). Biota participating in wastewater treatment in a horizontal flow constructed wetland. Water Sci Technol 44: 211-214. Wang, T. C., Weissman, J. C., Ramesh, G., Varadarajan, R. and Benemann, J. R. (1998). Heavy metal binding and removal by phormidium. Bull Environ Contam Toxicol 60: 739-744. Ward, D. M., Ferris, M. J., Nold, S. C. and Bateson, M. M. (1998). A natural view of microbial biodiversity within hot spring cyanobacterial mat communities. Microbiol Mol Biol Rev 62: 1353-1370. Wiegand, C. and Pflugmacher, S. (2005). Ecotoxicological effects of selected cyanobacterial secondary metabolites: a short review. Toxicol Appl Pharmacol 203: 201-218. Wimpenny, J., Manz, W. and Szewzyk, U. (2000). Heterogeneity in biofilms. FEMS Microbiol Rev 24: 661-671. Xu, Q., Barrios, C. A., Cutright, T. and Newby, B. M. (2005). Assessment of antifouling effectiveness of two natural product antifoulants by attachment study with freshwater bacteria. Environ Sci Pollut Res Int 12: 278-284. Zehr, J. P. and Turner, P. J. (2001). Nitrogen fixation: nitrogenase genes and gene expression. Methods Microbiol 30: 271–285. Zhang, K., Choi, H., Dionysiou, D. D., Sorial, G. A. and Oerther, D. B. (2006). Identifying pioneer bacterial species responsible for biofouling membrane bioreactors. Environ Microbiol 8: 433-440.

31

3

Diversity of phototrophic bacteria in microbial mats from Arctic hot springs (Greenland)

Guus Roeselers, Tracy B. Norris, Richard W. Castenholz, Søren Rysgaard, Ronnie N. Glud, Michael Kühl, and Gerard Muyzer.

Environmental Microbiology (2007) 9:26–938 Chapter 3

ABSTRACT

We investigated the genotypic diversity of oxygenic and anoxygenic phototrophic microorganisms in microbial mat samples collected from three hot spring localities on the east coast of Greenland. These hot springs harbour unique Arctic microbial ecosystems that have never been studied in detail before. Specific oligonucleotide primers for cyanobacteria, purple sulfur bacteria, green sulfur bacteria and Choroflexus/Roseiflexus-like green non- sulfur bacteria were used for the selective amplification of 16S rRNA gene fragments. Amplification products were separated by denaturing gradient gel electrophoresis (DGGE) and sequenced. In addition, several cyanobacteria were isolated from the mat samples, and classified morphologically and by 16S rRNA-based methods. The cyanobacterial 16S rRNA sequences obtained from DGGE represented a diverse, polyphyletic collection of cyanobacteria. The microbial mat communities were dominated by heterocystous and non- heterocystous filamentous cyanobacteria. Our results indicate that the cyanobacterial community composition in the samples were different for each sampling site. Different layers of the same heterogeneous mat often contained distinct and different communities of cyanobacteria. We observed a relationship between the cyanobacterial community composition and the in situ temperatures of different mat parts. The Greenland mats exhibited a low diversity of anoxygenic phototrophs as compared with other hot spring mats which is possibly related to the photochemical conditions within the mats resulting from the Arctic light regime.

INTRODUCTION

Microbial mats are layered structures composed of physiologically different groups of microorganisms (van Gemerden, 1993). They are present in a variety of environments where grazing is limited, such as hot springs, shallow coastal lagoons, hypersaline ponds and permanently ice-covered lakes (Stal and Caumette, 1994). The top layer of microbial mats is typically dominated by oxygenic phototrophs, such as cyanobacteria and diatoms, with underlying or intermixed layers of anoxygenic phototrophs, i.e. green and purple sulfur bacteria (GSB and PSB) and green non-sulfur bacteria (Nübel et al., 2002; Martinez-Alonso et al., 2005). Cyanobacteria are the most important primary producers in these ecosystems (Stal, 1995). Their photosynthetic activity fuels a variety of mineralization processes and conversions catalysed by heterotrophic and lithotrophic bacteria in the mat community (Ward et al., 1998). Geothermal springs with homeothermic source water temperatures above 0°C can be found all over Greenland, especially around the island of Disko on Western Greenland, where the occurrence of several thousands of such springs have been estimated (Heide- Jørgensen and Kristensen, 1998). However, the hottest springs, with source water temperatures of 55–62°C are found on Greenland's east coast at a number of locations north and south of Scoresbysund. In these springs, temperatures are high enough to allow significant growth of thermophilic microorganisms and most springs are characterized by the presence of thick gelatinous microbial mats (Kühl et al., 2004). These Arctic hot springs are all located close to 70°N latitude, and the phototrophic organisms are thus subject to extreme seasonal variations in the supply of photosynthetic active radiation (PAR); ranging from continuous daylight during the Arctic summer, to almost complete darkness during the winter when the sun does not rise above the horizon for several months (Cockell and Rothschild, 1997; 1999). Within the Arctic environment, hot spring microbial mats form isolated and disconnected habitats, which raises the question whether endemic taxa or

34 Microbial mats from Greenland species exist in such microbial communities (Castenholz, 1996). Some of the East Greenland springs have been known since the early explorers visited Greenland, and the presence of coloured slimy coatings in the hot springs was mentioned in several early expedition journals (e.g. Nordenskjöld, 1907). Presence of cyanobacteria was also mentioned in the few botanical and geological surveys of the springs that have been reported (Halliday et al., 1974). However, the microbiology of the East Greenland hot springs was never explored in detail.

Figure 1. Pictures showing some of the field sites at Kap Tobin (A, B), Rømerfjord (C), and Nørrefjord (D, E), where thick gelatinous microbial mats covered the bottom of the hot springs.

35 Chapter 3

Table 1. Source water characteristics at the three sampling sites

Spring Source Temperature (oC) Salinity (‰) pH Oxygen ((M) Kap Tobin 60-62 13-16 7.0-7.5 5-8 Nørrefjord 53-57 9-13 8.2-8.6 6-8 Rømer Fjord 56-61 5-7 9.1-9.5 0

In July 2003 an expedition was carried out to perform the first microbiological field studies in hot springs at three localities (i.e. Nørrefjord, Rømerfjord and Kap Tobin) on the East Greenland coast (Kühl et al., 2004). In these localities, over 70 springs with source water temperatures of > 40°C were found. All springs, irrespective of pronounced differences in water chemistry and temperature, contained characteristic orange, green and brownish coloured gelatinous biofilms covering the sides and the bottom of the springs. Microscopic observation of these biofilms revealed that they were mainly composed of a dense network of filamentous cyanobacteria embedded in a slimy matrix of exopolymers. The biogeochemistry of these microbial mats will be presented elsewhere (M. Kühl, S. Rysgaard and R. N. Glud, unpublished data). In this study we provide an inventory of the oxygenic and anoxygenic photoautotrophic microorganisms in the microbial mat samples collected during the expedition. The microbial diversity of these unique Arctic ecosystems was studied by genetic fingerprinting analysis of samples and pure cultures of cyanobacteria isolated from various mat samples. This approach enabled identification of the dominant cyanobacteria, PSB, GSB and green non-sulfur bacteria belonging to the Chloroflexaceae in the microbial mat communities.

MATERIALS & METHODS

Environmental parameters of the springs The mat samples used in this study were obtained from East Greenland hot springs at Kap Tobin (70°24.9'N; 21°57.04'W) and Nørrefjord (71°08.3'N; 22°2.1'W) on Jameson Land north of Scoresbysund, and at Rømerfjord (69°43.7'N; 23°41.9'W), on Blosseville's Coast south of Scoresbysund. The source water in the springs was characterized by temperature measurements with an electronic thermometer (Omnitherm, Germany), salinity determination with a calibrated refractometer (Atago, USA), pH determination with a portable pH-meter (Radiometer, Denmark) and oxygen determination by Winkler titration (Strickland and Parsons, 1972).

Sample collection Samples were taken from a number of springs at each locality harbouring microbial mats growing at different temperatures and exhibiting different coloration and structure (Tables 1 and 2). Most samples were cut out with a sharp knife and carefully transferred into 25 mm diameter Petri-dishes that were sealed in situ with air- and water-proof tape (3M, USA). Additional samples were taken with small Perspex cores and cut 50 ml syringes that were sealed in situ. Samples were kept cool (< 5–10°C) during transport back to Denmark, where they were frozen and kept at -80°C. Samples for molecular analysis were shipped on dry ice to the laboratory in Delft. Fresh samples for isolation attempts were kept in darkness at 4–6°C.

36 Microbial mats from Greenland

Isolation of cyanobacteria The cultures were isolated from most of the samples by various methods previously described in Castenholz (1988). Small pieces of inoculum were placed on BG-11 and D medium agar plates, incubated at 45°C at approximately 20 Wm-2 of coolwhite fluorescent illumination, and trichomes were allowed to migrate out (or grow out) from the source. Single trichomes (or filaments) were then picked off (including the small agar piece supporting the trichome) with a watchmaker's forceps and inoculated into liquid medium. In addition, especially for unicellular species, small pieces of inoculum were broken up and vigorously dispersed with a syringe using a large needle (17 gauge) or cannula. This was then processed through a dilution to extinction series and the final enrichment at 45° and 50°C was streak diluted on D medium agar plates to produce clonal colonies. Some of these colonies were then resuspended and replated to obtain new colonies, which were then transferred to liquid medium (BG-11 and D). Unfortunately, the original material from the springs had been stored for several months at 4–6°C, rather than at room temperature, which is recommended for hot spring collections (Castenholz, 1988). Because of this, some of the collected material may not have resulted in viable cells. Therefore, the cultures isolated probably represent an incomplete picture of the species present. Photomicrographs were taken with a Zeiss Axioplan microscope (Carl Zeiss, Jena, Germany), using a 100x plan-neofluar objective with Nomarski differential interference contrast (DIC) optics. The cultures are kept in the Culture Collection of Microorganisms from Extreme Environments (CCMEE) at the University of Oregon (http://cultures.uoregon.edu) both as growing batch cultures at 45° and at -80°C.

DNA extraction Genomic DNA was extracted from the mat samples by applying approximately 500 mg frozen biomass to the UltraClean Soil DNA Isolation Kit (Mo Bio Laboratories, Carlsbad, CA, USA) according to the manufacturer's protocol. Cell lysis was confirmed by phase- contrast microscopy. DNA dilutions were stored at -20°C. Genomic DNA extraction from culture isolates was performed using a modification of the method of More et al. (1994). Culture aliquots were pelleted and approximately 100 l of pelleted cells was transferred to 2-ml screw-cap microcentrifuge tubes. Next, 0.75 g of 0.1 mm diameter zirconia/silica beads (BioSpec Products, Bartlesville, OK) were added to each tube along with 600 l of 120 mM sodium phosphate (pH 8.0) and 400 l of lysis buffer (10% (wt/vol) sodium dodecyl sulfate, 0.5 M Tris-HCl (pH 8.0) and 0.1 M NaCl). Cells were lysed by shaking for 3 min at high speed on a Mini-Beadbeater (BioSpec Products, Bartlesville, OK) and then centrifuged for 3 min at 13 000 g. Supernatant (700 l) was collected and DNA was precipitated on ice using 2:5 (v/v) of 7.5 M ammonium acetate and then centrifuged again. The supernatant was collected and the DNA was isopropanol precipitated. Finally, the pellet was washed with ice-cold 70% (v/v) ethanol; air-dried and resuspended in 100 l of 10 mM Tris (pH 8.0).

Polymerase chain reaction amplification of rRNA gene fragments Extreme care was taken to prevent any DNA contamination of solutions and plastic disposables used for PCR. All heat sterilized plastic tubes were exposed to UV light for 30 min before use. Only DNA and RNA-free water (Sigma-Aldrich, St. Louis, MO, USA) was used to prepare PCR reagent stock solutions and reaction mixtures. To amplify the 16S rRNA encoding gene fragments of cyanobacteria, The DNA dilutions were used as template DNA in 100 l of PCR reactions using the universal primer 359F-GC and an equimolar mixture of the reverse primers 781R(a) and 781R(b), and PCR conditions as described by

37 Chapter 3

Nübel et al. (1997). Primer pairs 341F + GC (Schäfer and Muyzer, 2001)/GSB822R (Overmann et al., 1999), and Chr986F (Overmann et al., 1999)/1392R + GC (Schäfer and Muyzer, 2001) were specific for anoxygenic phototrophic proteobacteria belonging to genera of Chromatiaceae (i.e. the PSB) and for the members of the Chlorobi (GSB) respectively. The PCR amplification was carried out as described by Overmann et al. (1999). The primers CCR-344-F and CCR-1338-R + GC were used to amplify 16S rRNA gene encoding sequences affiliated to filamentous phototrophic bacteria within the kingdom of 'green non- sulfur bacteria' or 'Chloroflexus relatives' (Nübel et al., 2001). This PCR was carried out with a denaturation step of 2 min at 94°C, followed by 30 cycles of denaturation for 45 s at 94°C, annealing for 45 s at 60°C, and extension for 4 s at 72°C, followed by a final extension step of 7 min at 72°C. All amplification reactions were performed in a T1 Thermocycler (Biometra, Westburg, the Netherlands). The primers 8F and 1492R (Amann et al., 1995) were used to amplify 16S-rRNA gene sequences from the cyanobacterial isolates, resulting in a product of approximately 1485 bp. The PCR amplification cycle was as follows: initial denaturation for 2 min at 95°C, then 34 cycles of 1 min of denaturation at 95°C, 1 min annealing at 45°C, and 1 min extension at 72°C followed by a final extension of 7 min at 72°C.

Denaturing gradient gel electrophoresis Denaturing gradient gel electrophoresis (DGGE) was performed as described by Schäfer and Muyzer (2001). Briefly, one mm thick 6% acrylamide gels with a urea- formamide (UF) gradient of 20–70% were used for cyanobacterial 16S rRNA gene fragments. Gradients of 20–80% were used for 16S rRNA gene fragments of GSB and PSB. Gradients of 30–80% UF were used for Chloroflexus relatives. Gels were run for 16 h at 100 V and at a constant temperature of 60°C. The gels with Chloroflexus amplicons were run for 20 h at 72 V and at 60°C. Gels were stained in an ethidium bromide solution and analysed and photographed using the GelDoc UV Transilluminator (Bio-Rad, Hercules, CA, USA). The dominant bands were excised from the DGGE gels with a sterile surgical scalpel. Each gel slice was placed in 15 l of sterile water for 24 h at 4°C. Subsequently, the solution was used as template DNA for re-amplification as described above. The PCR products were again subjected to DGGE analysis to confirm the purity and their position relative to the bands from which they were originally excised. The PCR products were purified using the QIAquick Gel Extraction Kit (QIAGEN, Hilden, Germany). The purified PCR products were sequenced by a commercial company (BaseClear, Leiden, the Netherlands), using the appropriate specific forward primers without GC clamp.

Comparative sequence analysis Partial 16S rRNA gene sequences were first compared with the sequences stored in the GenBank nucleotide database using the blast algorithm (Altschul et al., 1990) in order to obtain a first identification of the mat community members. Subsequently, the sequences were imported into the ARB SSU rRNA database (available at http://www.arb-home.de) (Ludwig et al., 2004) and aligned based on the secondary structure of the small subunit (SSU) rRNA. The dissimilarity values were used to calculate distance matrices. Distance matrix trees were generated by the Neighbour-Joining (NJ) method with the Felsenstein correction as implemented in the paup 4.0B software (Sinauer, Sunderland, MA, USA). The NJ calculation was subjected to bootstrap analysis (1000 replicates). The partial 16S rRNA gene sequences were deposited in the GenBank database. The sequences were deposited under Accession numbers: DQ430942 to DQ431006.

38 Microbial mats from Greenland

RESULTS

Mat structure and spring water chemistry The Nørrefjord and Rømerfjord localities harboured numerous hot springs with source water temperatures > 50°C, while only one spring, with several sources, was found at Kap Tobin. The water chemistry differed among the three localities (Table 1). The Kap Tobin spring reached the highest source temperature of ~62°C, was almost pH neutral (pH 7–7.5) and had a relatively high salinity of ~15‰. A slight smell of sulfide was detected close to the source, where yellow-whitish material indicated sulfur precipitation. The springs at Nørrefjord were alkaline (pH 8.2–8.6) and had a slightly lower and more variable salinity of 9–13‰. The springs at Rømerfjord were highly alkaline (pH 9–9.5) with a low salinity of 5–7‰. Whitish material and streamers indicating sulfur precipitation covered the sediment closest to the source in several springs at Rømerfjord. The source water had a low oxygen content (< 10 M) at Kap Tobin and Nørrefjord, while the source water in hot springs at Rømerfjord was anoxic. Despite the differences in water chemistry, all springs harboured massive growth of several cm thick microbial mats with a very gelatinous and translucent structure on top of the sediment and gravel in the springs (Fig. 1). Around fast flowing sources, green streamers were also frequently observed. Microscopic investigation of the mats showed the presence of a dense network of cyanobacteria embedded in copious amounts of exopolymers. Most mats had a thick orange-greenish surface layer on top of a darker green layer (Fig. 1E, Table 2). In a few mats at Nørrefjord, a reddish layer was sometimes observed below the surface layers. Some springs at Nørrefjord also harboured floating mats with a dark brownish surface colour of the air-exposed parts.

39 Chapter 3

Table 2. Temperature, morphology and color of the collected mat samples. K1-K12 were collected from Kap Tobin, R0-R8 were collected from Rømerfjord and, N1-N10 were collected from Nørrefjord.

Sample T (oC) Mat morphology and color K1 60-61 Dark green mat and streamers K2 45-50 Gelatinous mat, top layer K3 45-50 Gelatinous mat, underlying layer K4 45-50 Gelatinous mat, orange top layer K5 45-50 Gelatinous mat, green under layer K6 50 Cushion forming mat K7 <40 Greenish-brown mat K8 61,5 Sample from source K9 50 Thick greenish gelatinous mat K10 40-45 Orange cushion mat K11 40-42 Brown mat K12 ~35 Green-brown mat R0 56-58 White streamer material with sediment close to source R1 42 Orange reddish mat, reddish top layer R2 42 Orange reddish mat, green underneath top layer R3 50-53 Rigid gelatinous orange mat R4 45-48 Surface layer of thick gelatinous greenish mat R5 20 Dense dark green mat (receives splashes of hot water from source) R6 50 Orange gelatinous mat R7 40 Orange-green gelatinous mat R8 ~58 Black sediment below and between white streamers N1 ~50 Green streamers near source N2 49 Orange mat, top layer N3 49 Orange mat, green layer N4 49 Orange mat, red layer N5 45 Green gelatinous mat N6 42 Cohesive brown surface layer on top of a pale green yellowish layer N7 46 Orange mat N8 46 Orange gelatinous mat N9 ~20 Brown mat, surface layer N10 46 Orange mat

Diversity of oxygenic phototrophs Polymerase chain reaction denaturing gradient gel electrophoresis (PCR-DGGE) analysis with primers specific for cyanobacteria and chloroplasts revealed differences in community composition among the three locations as well as among the individual samples of each location. Most of the DGGE profiles from the Kap Tobin mat samples showed three or four dominant bands, while the DGGE profiles from Nørrefjord and, Rømerfjord samples showed only one or two dominant bands (Fig. 2). All DNA sequences except for three (i.e. band R0-1, R8-9 and band R8-10, Fig. 2B) were related to cyanobacteria. A phylogenetic tree was constructed with the sequences derived from the DNA fragments (Fig. 3). The Nørrefjord mat samples were characterized by the presence of sequences related to filamentous cyanobacteria of the genera Chlorogloeopsis, Fischerella and, Leptolyngbya (Figs 2A and 3). The mat samples from Kap Tobin were characterized by the presence of sequences related to Leptolyngbya and Scytonema species (Figs 2C and 3). Rømerfjord was characterized by the presence of Phormidium species (Figs 2B and 3).

40 Microbial mats from Greenland

The sequences derived from band R0-1, band R8-8, and band R8-9 are affiliated to uncultured members of the genus Nitrospira, suggesting partial non-specificity of the cyanobacterial primers. These sequences were not incorporated in the phylogenetic tree. Out of the 10 samples collected from Nørrefjord, samples N2, N5, N8 and N10 showed one dominant band at identical position on the DGGE gel (Fig. 2A). The identical sequences derived from these bands were affiliated to Leptolyngbya species (Fig. 3). All 12 samples obtained from the Kap Tobin spring, except for sample K7, K11 and K12, showed a dominant band at identical positions on the DGGE gel. The sequences derived from these bands were 99% similar to a slightly higher positioned band that was present in K3, K4, K5, K6 and K9. All bands clustered closely together with the Leptolyngbya affiliated sequences found in the Nørrefjord springs (Fig. 3) Nevertheless, within this cluster the Nørrefjord sequences exhibited only 95% similarity with the Kap Tobin sequences. Samples N6 and N9 from Nørrefjord and sample K12 from Kap Tobin all showed a dominant band positioned high on the denaturing gradient gel (Fig. 2A and C) with high sequence similarities (> 99%). All of them exhibited at least 8% dissimilarity with the closest related Anabaena species from the database. The identical sequences from band R1-2 and band R2-4 (Fig. 2B) showed 15% dissimilarity to their closest relative in the database, an uncultured Antarctic cyanobacterium derived from the ice cover of Lake Bonney (Antarctica) (Gordon et al., 2000). Photomicrographs of cyanobacterial isolates GRN-1, GRN-7, GRN-8 and, GRN-11 show, respectively, Chlorogloeopsis-like trichomes, Leptolyngbya-like filaments, Mastigocladus (Fischerella) laminosus-like filaments and, typical Synechococcus-like unicells (Fig. 4). The 16S rRNA sequences obtained from these pure cultures of cyanobacteria, which were isolated from the mat samples, clustered together with DGGE derived sequences (Fig. 3). The sequences of isolates GRN-1 to GRN-5 as well as the DGGE band sequences N1-1 and K1-1 were all more than 99% similar to the sequence of Chlorogloeopsis sp. PCC 7518, which was originally isolated from an Icelandic hot spring (Fig. 3). Sequences of isolates GRN-6 and GRN-7 clustered together with the sequences K1-2, K4-3, K4-4, K6-6, and, K8-10, which are all affiliated to Leptolyngbya spp

41 Chapter 3

Figure 2. Denaturing gradient gel electrophoresis patterns of 16S rRNA gene fragments obtained after enzymatic amplification using primers specific for oxygenic phototrophs and genomic DNA from various microbial mat samples. A. Mat samples from Nørrefjord hot spring (samples N1 to N10). B. Mat samples from Rømerfjord hot spring (samples R0 to R8). C. Mats samples from Kap Tobin hot spring (samples K1 to K12). Numbers at the left of each lane correspond to bands that were excised, PCR-amplified and sequenced.

42 Microbial mats from Greenland

Figure 3. Evolutionary tree showing the phylogenetic affiliations of the cyanobacterial 16S rRNA gene sequences derived from the DGGE gels shown in Fig. 2. The sequences obtained in this study are printed bold. Escherichia coli (AJ567606) was used as an out-group, but was pruned from the tree. Scale bar indicates 10% sequence divergence. Bootstrap values (1000 replicates) that were > 50 are placed at the nodes of the branches.

43 Chapter 3

Figure 4. Nomarski DIC photomicrographs of cyanobacterial isolates. All cyanobacteria are shown at the same magnification and the scale bar in all panels indicates 10 M. A) 'High temperature form' Chlorogloeopsis trichomes isolated from mat sample Kap Tobin 9B (57–61°C). B) Leptolyngbya-like filaments (~2.0–2.2 M wide) isolated from mat sample Kap Tobin 16 (40°C). C) Mastigocladus (Fischerella) laminosus-like filaments isolated from mat sample Nørrefjord 11 (green layer, 45°C). It should be noted that the Pasteur Culture Collection (PCC) classifies all Mastigocladus as Fischerella. D) Typical Synechococcus-like unicells, about 1.5 M wide, isolated from mat sample Nørrefjord 11 (green layer 45°C).

Diversity of anoxygenic phototrophs Polymerase chain reaction with primers specific for GSB only yielded significant amounts of PCR products from samples N3, N4, N8, R5, K1 and K9. Subsequent DGGE analysis showed a dominant band at the same position for each sample (Fig. 5). A total of six bands were excised, re-amplified and sequenced. The obtained sequences had an identical position in the phylogenic tree and are related to an uncultured Chlorobium species (Fig. 6A). Amplification with primers specific for PSB resulted in significant PCR products from eight of the 31 samples. All Chromatium affiliated sequences were derived from samples from the Kap Tobin mats, except one that came from a Rømerfjord sample (Fig. 6B). Purple sulfur bacteria were absent or below detection levels in the other samples. The DGGE with the PSB 16S rRNA gene fragments was characterized by the presence of a single band for each sample with the exception of K11, which showed three bands (Fig. 5). The phylogenetic tree (Fig. 6B) with the sequences obtained from the excised DGGE bands showed that bands K7-4, K10-6, K11-7 and K12-10 were similar to Isochromatium buderi (Imhoff et al., 1998) and K119 was similar to Thiocapsa roseopersicina, both are known microbial mat inhabitants. The other bands were related to uncultured Chromatiaceae.

44 Microbial mats from Greenland

Figure 5. Denaturing gradient gel electrophoresis patterns of 16S rRNA gene fragments obtained after enzymatic amplification using primers specific for green sulfur bacteria (GSB) and purple sulfur bacteria (PSB) and DNA samples from microbial mats of Rømerfjord hot spring (R5), Nørrefjord hot spring (N3, N4, N8 and N10), and Kap Tobin hot spring (K1, K6, K7 K9, K10, K11 and K12). Numbers at the left of each lane correspond to bands that were excised, PCR-amplified and sequenced.

Amplification with primers specific for photosynthetic green non-sulfur bacteria within the Chloroflexaceae resulted in a significant amount of PCR product in 15 out of 31 samples. However, only four samples resulted in a clear DGGE pattern from which DNA sequences could be derived successfully (Figs 6C and 7). The DGGE profile of Rømerfjord sample R5 contained two bands. One DGGE band (Band R5-6) was closely related to Roseiflexus castenholzii, a red filamentous green non-sulfur bacterium that was first isolated from microbial mats in alkaline hot springs in Japan (Hanada et al., 2002). The other band from R5 clustered together with band 1 from Nørrefjord sample N3 and an uncultured Chloroflexus-like bacterium. Bands 2 and 3 from Nørrefjord sample N3 clustered together with the freshwater planktonic species Chloronema giganteum (Dubinina and Gorlenko, 1975) and the hot spring inhabiting Chloroflexus aggregans and Chloroflexus aurantiacus (Hanada et al., 1995).

45 Chapter 3

Figure 6. Evolutionary trees showing the phylogenetic affiliations of the green sulfur bacteria (A) purple sulfur bacteria (B) and Chloroflexaceae types (C). The 16S rRNA gene sequences used to generate the trees were obtained from DNA fragments excised from the denaturing gradient gel shown in Fig. 5 and 7. Escherichia coli (AJ567606) was used as an out-group, but pruned from the trees. Scale bar indicates 10% sequence divergence. Bootstrap values (1000 replicates) that were > 50 are placed at the nodes of the branches.

46 Microbial mats from Greenland

Figure 7. Denaturing gradient gel electrophoresis patterns of 16S rRNA gene fragments obtained after enzymatic amplification using primers specific for Chloroflexus spp. and phylogenetic relatives and DNA samples from microbial mats of Rømerfjord hot springs (R5 and R8), and Nørrefjord hot springs (N3 and N5). Numbers at the left of each lane correspond to bands that could be excised, PCR-amplified and sequenced.

DISCUSSION

Our survey of phototrophic genotypes revealed that cyanobacteria were the predominant oxygenic phototrophs. The differences in the cyanobacterial community structure among the three locations were more prevalent than the differences among individual samples from one location. Most of the samples from the Kap Tobin mats showed several dominant bands, while the samples from Nørrefjord and, Rømerfjord showed only one or two dominant bands suggesting a higher number of dominant cyanobacterial strains in the Kap Tobin mats. The analysed mat samples were morphologically very diverse and were sampled from different in situ temperatures (Table 2), and our inventory of the microbial phototrophs did not reveal a clear-cut correspondence between the genotypic diversity and the morphological characteristics of the mats. However, the phylogenetic analysis showed that the 16S rRNA gene sequences affiliated with Chlorogloeopsis PCC7518, Chlorogloeopsis fritschii and Fischerella muscicola were all derived from microbial mats growing at  45°C, like the clusters of sequences affiliated with Leptolyngbya species found in the Kap Tobin and Nørrefjord mats. Chlorogloeopsis PCC7518 [(formerly referred to as 'HTF' Mastigocladus (Castenholz, 1969)] and Leptolyngbya species have been described to inhabit high temperature environments (Castenholz, 1996; 2001). However, the morphotypes C. fritschii and F. muscicola are not known as thermophiles. The sequences affiliated with Anabaenopsis sp. and Scytonema sp. originated from samples with temperatures below 45°C (Table 1, Fig. 3).

47 Chapter 3

The cyanobacterial 16S rRNA gene sequences obtained from the DGGE gels and from the isolates were related to a diverse, polyphyletic group of cyanobacteria. The presence of long branching sequences in the cyanobacterial phylogenetic tree could indicate that some of the phylotypes obtained in this study are restricted to this specific environment. The Arctic light regime clearly distinguishes the Greenland mats from microbial mats found in geothermal springs under more temperate climates, although Icelandic hot springs at 66°N latitude are only on a ~4 degrees lower latitude than those near Scoresbysund. Sperling (1975) found that hot spring mats persisted during the winter in some Icelandic hot springs. The mats also persisted during winter in Kap Tobin and Nørrefjord and had the same morphology and colour as the summer mats (H. C. Scoresby Hammeken, pers. comm.). We speculate that some cyanobacteria inhabiting the Greenland springs may have special ecophysiological adaptations to the high Arctic light regime and the resulting photochemical effects. For example, it has been shown that phylogenetically distinct Antarctic oscillatorians show a high level of adaptive flexibility in pigmentation in response to changes in PAR supply (Quesada and Vincent, 1997). Recently, it was shown that unicellular cyanobacteria growing at 60–65°C in some geothermal springs employ a complex network of metabolic switching during a diel cycle (Steunou et al., 2006), during which the mat undergoes a change from highly supersaturated oxygen conditions in daylight to almost complete anoxia during night-time. Expression of genes coding for enzymes involved in nitrogen fixation, and the N2-fixing activity were closely coupled to the diminishing light and oxygen levels thereby minimizing inhibition of nitrogenase by oxygen in the mat. Such a mechanism would not work in the Greenland hot springs, which are constantly supersaturated with oxygen throughout the Artic summer months (Kühl et al., 2004). This may explain the abundance of sequences affiliated with heterocystous cyanobacteria like Chlorogloeopsis, Fischerella and Scytonema species in our samples. The fact that microbial mats can become highly saturated with oxygen during daytime would probably only allow diazotrophy by heterocystous cyanobacteria, because non-heterocystous types usually require a dark period for this anaerobic process (Stal, 2000). We speculate that the persistence of oxic conditions in the mats for several months during summer may select for heterocystous cyanobacteria. The temporal and spatial distribution of diazotrophic activity within the mats could be an interesting subject for further investigation. Recent molecular ecology studies support the hypothesis of Antarctic cyanobacterial endemism (Priscu et al., 1998; Vincent, 2000; Taton et al., 2003). There is, however, no consensus concerning the extent of endemism in Arctic cyanobacteria. The fact that many of the identified cyanobacterial sequences exhibit high dissimilarity with the sequences available in the databases may be an indication of the existence of endemic or at least specially adapted cyanobacteria in these environments. On the one hand, the dominant, mat building primary producers in these ecosystems might have unique adaptations to the Arctic light regime, and the resulting shift between long periods of continuous oxic or anoxic conditions in the mats. On the other hand, the island-like characteristics of the hot springs and the resulting geographical isolation of their inhabitants could be considered as one of the possible factors of speciation (Papke et al., 2003). Thermophilic cyanobacteria of the genus Synechococcus are absent in Iceland and Alaska (Castenholz, 1996) while in the western United States thermophilic Synechococcus species of this morphotype are found in chemically diverse hot springs (Ferris et al., 1996a). Because all thermophilic forms of Synechococcus appear to be absent in Iceland, the presence of members of this genus (i.e. isolate GRN-11) (Figs 3 and 4D) in the Greenland springs is quite surprising, and may reflect the fact that these springs are found in a

48 Microbial mats from Greenland geologically much older setting than the springs on Iceland (the oldest crust of Iceland is only 20 million years old, while Greenland is c. 3.7 billion years old). In contrast, the thermophilic cyanobacterium Chlorogloeopsis PCC7518, found in hot springs world-wide (Castenholz, 1996), also occurs in the Greenland hot springs. Several of the other cyanobacterial isolates and genotypes we found in the Greenland hot springs are representative of genera and species that occur commonly in geographically close hot springs on Iceland. A spreading of microbes from Iceland to East Greenland and vice versa may be facilitated, e.g. by migrating birds, which are known to use hot springs on both Iceland and the East Greenland coast as resting places; a similar dispersal mechanism has been proposed for the unique flora found surrounding the East Greenland hot springs (Halliday et al., 1974). Evidently, this type of dispersal works only for thermophilic microbes that can survive exposure to much lower temperatures and desiccation, and the fact that thermophilic Synechococcus have not spread from Greenland to Iceland could thus indicate that these thermophiles have a much lower tolerance to conditions involved in dispersal, such as freezing or desiccation. Although, it has been shown that widely varying chemical environments do not a priori restrict Synechococcus colonization (Papke et al., 2003), possible unidentified abiotic effects on the distribution of this genotype cannot be ruled out. No sequences were found that are affiliated with algal chloroplasts indicating that eukaryotic algae are very low in number or absent in the hot spring mats. This was also evident from microscopic investigations of many mat samples. The DGGEs and the derived sequence affiliations revealed a lower diversity among the GSB and PSB compared with other mat ecosystems (Ferris et al., 1996b; Elshahed et al., 2003). Green sulfur bacteria are a taxonomically very distinct group of anoxyphototrophic bacteria. They require strictly anoxic conditions, and they need only about one-fourth of the light intensity of the PSB to attain comparable growth rates. The low diversity of GSB could again be related to the continuous oxic state of the mats throughout the Arctic summer (Kühl et al., 2004). Although pigment synthesis in phototrophic PSB is inhibited by continuous exposure to oxygen most species of the Chromatiaceae are relatively tolerant to periodic exposure. The DGGE profile of sample K11 showed one band (Fig. 5, Band K11-9) affiliated to T. roseopersicina, which is ranked among the most oxygen tolerant members of the Chromatiaceae. Nevertheless, the diversity of PSB was also low. Green and purple sulfur bacteria could not be detected in any of the Rømerfjord samples except for sample R5 with a relatively low in situ temperature of 20°C. The source water of Rømerfjord has a higher average pH (9.1–9.5) than the source water from Kap Tobin and Nørrefjord. The majority of the GSB and PSB show optimal growth rates ranging from pH 6.5 to 8, but some halophilic PSB are true alkaliphiles with optimal growth at pH 10 (Sorokin and Kuenen, 2005). The only PSB sequence derived from sample R5 (Figs 5 and 7B) shows distant affiliation to a Halothiobacillus strain (a non-photosynthetic bacterium), which is not known as akaliphilic. It must be noted, however, that only two-thirds of the presently known 16S rRNA gene sequences of Chromatiaceae will be amplified with primer Chr986f (Overmann et al., 1999). Hence, we cannot exclude the possibility that the diversity of Chromatiaceae is higher than encountered by our approach. Mat communities dominated by Chloroflexus spp. are well represented in geothermal springs where elevated sulfide concentrations inhibit cyanobacteria and promote anoxygenic photoautotrophy (Giovannoni et al., 1987). We could only obtain Chloroflexaceae DGGE profiles for a few mat samples and from some DGGE bands we were unable to derive DNA sequences. We conclude that Chloroflexaceae are not very abundant in the studied mat systems especially compared with similar hot spring mats, e.g.

49 Chapter 3 in Iceland (Skirnisdottir et al., 2000). Samples from Nørrefjord contained a few sequences that cluster loosely with C. aurantiacus and C. aggregans (Hanada et al. (1995). Roseiflexus spp., another group of green non-sulfur bacteria, appears to predominate in non-sulfidic hot springs (van der Meer et al., 2005). Only one sequence affiliated with R. castenholzii was derived from Rømerfjord sample R5. In conclusion, our survey of the phototrophic inhabitants of the hot springs of Greenland shows the presence of phylogenetically diverse populations of filamentous and unicellular cyanobacteria and a few representatives of anoxygenic phototrophs. Cyanobacteria appear to be the dominant mat inhabitants and their diversity includes cyanobacteria with a cosmopolitan distribution in hot springs. We detected a thermophilic Synechococcus sp., which was not found in Alaskan or in Icelandic hot springs. The physiology and geographical distribution of thermophilic cyanobacteria should be the subject of future investigations. This will enhance our understanding of the role and the dispersal of cyanobacteria as the dominant primary-producing organisms in geothermal environments and their ability to survive in high latitudes. In addition, the described hot springs offer a unique opportunity to study the effect of the Arctic light regime on microbial ecosystems without the persistent low temperatures normally associated with polar environments.

ACKNOWLEDGEMENTS

This work was supported by grants from the European Union (G.M. and M.K.) (Contract: QLK3-CT-2002-01938), and from the Carlsberg Foundation (M.K.) and the Danish Natural Science Research Council (M.K. and S.R.). Work carried out in Oregon was aided by a NASA Astrobiology Cooperative Grant (R.W.C.). We gratefully thank Yunyun Zou, Esengül Yildrim, Hans Christian Scoresby Hammeken and Hans Pedersen for technical assistance in the laboratory and the field. Mike Madigan is thanked for generously supplying us with cultures of Chloroflexus and Roseiflexus sp. for use in the genetic analysis.

REFERENCES

Altschul, S.F., Gish, W., Miller, W., Myers, E.W., and Lipman, D.J. (1990) Basic local alignment search tool. J Mol Biol 215: 403–410. Amann, R.I., Ludwig, W., and Schleifer, K.H. (1995) Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol Rev 59: 143–169. Castenholz, R.W. (1969) The thermophilic cyanophytes of Iceland and the upper temperature limit. J Phycol 5: 360–368. Castenholz, R.W. (1988) Culturing methods for cyanobacteria. In Methods in Enzymology. Glazer, A.N., and Packer, L. (eds). : Academic Press, pp. 68–93. Castenholz, R.W. (1996) Endemism and biodiversity of thermophilic cyanobacteria. Nova Hedwigia 112: 33–47. Castenholz, R.W. (2001) Phylum BX. Cyanobacteria. Oxygenic photosynthetic bacteria. In Bergey's Manual of Systematic Bacteriology. Boone, D.R., Castenholz, R.W., and Garrity, G.M. (eds). : Springer,pp. 474–487. Cockell, C.S., and Rothschild, L.J. (1997) Photosynthetic rhythmicity in an antarctic microbial mat and some considerations on polar circadian rhythms. Antarct J US A 32: 156–157. Cockell, C.S., and Rothschild, L.J. (1999) The effects of UV radiation A and B on diurnal variation in photosynthesis in three taxonomically and ecologically diverse microbial mats. Photochem Photobiol 69: 203–210. Dubinina, G.A., and Gorlenko, V.M. (1975) New filamentous photosynthesizing green bacteria with gas vacuoles. Mikrobiologiia 44: 511–517.

50 Microbial mats from Greenland

Elshahed, M.S., Senko, J.M., Najar, F.Z., Kenton, S.M., Roe, B.A., Dewers, T.A., et al. (2003) Bacterial diversity and sulfur cycling in a mesophilic sulfide-rich spring. Appl Environ Microbiol 69: 5609–5621. Ferris, M.J., Ruff-Roberts, A.L., Kopczynski, E.D., Bateson, M.M., and Ward, D.M. (1996a) Enrichment culture and microscopy conceal diverse thermophilic Synechococcus populations in a single hot spring microbial mat habitat. Appl Environ Microbiol 62: 1045–1050. Ferris, M.J., Muyzer, G., and Ward, D.M. (1996b) Denaturing gradient gel electrophoresis profiles of 16S rRNA-defined populations inhabiting a hot spring microbial mat community. Appl Environ Microbiol 62: 340–346. van Gemerden, H. (1993) Microbial mats, a joint venture. Mar Geol 113: 3–25. Giovannoni, S.J., Revsbech, N.P., Ward, D.M., and Castenholz, R.W. (1987) Obligately phototrophic Chloroflexus: primary production in anaerobic hot spring microbial mats. Arch Microbiol 147: 80–87. Gordon, D.A., Priscu, J., and Giovannoni, S. (2000) Origin and phylogeny of microbes living in permanent antarctic lake ice. Microb Ecol 39: 197–202. Halliday, G., Kliim-Nielsen, L., and Smart, I.H.M. (1974) Studies on the flora of the North Blosseville Kyst and on the hot springs of Greenland. Meddel Grønland 199: 1–47. Hanada, S., Hiraishi, A., Shimada, K., and Matsuura, K. (1995) Chloroflexus aggregans sp. nov., a filamentous phototrophic bacterium which forms dense cell aggregates by active gliding movement. Int J Syst Bacteriol 45: 676–681. Hanada, S., Takaichi, S., Matsuura, K., and Nakamura, K. (2002) Roseiflexus castenholzii gen. nov., sp. nov., a thermophilic, filamentous, photosynthetic bacterium that lacks chlorosomes. Int J Syst Bacteriol 52: 187–193. Heide-Jørgensen, H.S., and Kristensen, R.M. (1998) Puilassoq, the warmest homothermal spring of Disco Island. Ber Polarforschung 330: 32–43. Imhoff, J.F., Suling, J., and Petri, R. (1998) Phylogenetic relationships among the Chromatiaceae, their taxonomic reclassification and description of the new genera Allochromatium, Halochromatium, Isochromatium, Marichromatium, Thiococcus, Thiohalocapsa and Thermochromatium. Int J Syst Bacteriol 48: 1129–1143. Kühl, M., Glud, R.N., Rysgaard, S., and Pedersen, H. (2004) Microbiological studies in hot springs at Daneborg and Scoresbysund. In Zackenberg Ecological Research Operations, 9th Annual Report. Rasch, M., and Caning, K. (eds). : Danish Polar Center, pp. 72–74. Ludwig, W., Strunk, O., Westram, R., Richter, L., Meier, H., Kumar, Y., et al. (2004) ARB: a software environment for sequence data. Nucleic Acids Res 32: 1363–1371. Martinez-Alonso, M., Van Bleijswijk, J., Gaju, N., and Muyzer, G. (2005) Diversity of anoxygenic phototrophic sulfur bacteria in the microbial mats of the Ebro Delta: a combined morphological and molecular approach. FEMS Microbiol Ecol 52: 339–350 van der Meer, M.T., Schouten, S., Bateson, M.M., Nübel, U., Wieland, A., Kühl, M., et al. (2005) Diel variations in carbon metabolism by green nonsulfur-like bacteria in alkaline siliceous hot spring microbial mats from Yellowstone National Park. Appl Environ Microbiol 71: 3978–3986. More, M.I., Herrick, J.B., Silva, M.C., Ghiorse, W.C., and Madsen, E.L. (1994) Quantitative cell lysis of indigenous microorganisms and rapid extraction of microbial DNA from sediment. Appl Environ Microbiol 60: 1572–1580. Nordenskjold, O. (1907) On the geology and physical geography of East Greenland. Meddelelser orn Grønland 28: 151–284. Nübel, U., Garcia-Pichel, F., and Muyzer, G. (1997) PCR primers to amplify 16S rRNA genes from cyanobacteria. Appl Environ Microbiol 63: 3327–3332. Nübel, U., Bateson, M.M., Madigan, M.T., Kühl, M., and Ward, D.M. (2001) Diversity and distribution in hypersaline microbial mats of bacteria related to Chloroflexus spp. Appl Environ Microbiol 67: 4365–4371. Nübel, U., Bateson, M.M., Vandieken, V., Wieland, A., Kühl, M., and Ward, D.M. (2002) Microscopic examination of distribution and phenotypic properties of phylogenetically diverse Chloroflexaceae- related bacteria in hot spring microbial mats. Appl Environ Microbiol 68: 4593–4603. Overmann, J., Coolen, M.J., and Tuschak, C. (1999) Specific detection of different phylogenetic groups of chemocline bacteria based on PCR and denaturing gradient gel electrophoresis of 16S rRNA gene fragments. Arch Microbiol 172: 83–94. Papke, R.T., Ramsing, N.B., Bateson, M.M., and Ward, D.M. (2003) Geographical isolation in hot spring cyanobacteria. Environ Microbiol 5: 650–659.

51 Chapter 3

Priscu, J.C., Fritsen, C.H., Adams, E.E., Giovannoni, S.J., Paerl, H.W., McKay, C.P., et al. (1998) Perennial Antarctic lake ice: an oasis for life in a polar desert. Science 280: 2095–2098. Quesada, A., and Vincent, W.F. (1997) Strategies of adaptation by Antarctic cyanobacteria to ultraviolet radiation. Eur J Phycol 32: 335–342. Schäfer, H., and Muyzer, G. (2001) Denaturing gradient gel electrophoresis in marine microbial ecology. In Methods in Microbiology, Marine Microbiology. Paul, J.H. (ed.). : Academic Press, pp. 425–468. Skirnisdottir, S., Hreggvidsson, G.O., Hjorleifsdottir, S., Marteinsson, V.T., Petursdottir, S.K., Holst, O., and Kristjansson, J.K. (2000) Influence of sulfide and temperature on species composition and community structure of hot spring microbial mats. Appl Environ Microbiol 66: 2835–2841. Sorokin, D.Y., and Kuenen, J.G. (2005) Haloalkaliphilic sulfur-oxidizing bacteria in soda lakes. FEMS Microbiol Rev 29: 685–702. Sperling, A. (1975) Algal ecology of southern icelandic hot springs in winter. Ecology 56: 183–190. Stal, L.J. (1995) Physiological ecology of cyanobacteria in microbial mats and other communities. New Phytol 131: 1–32. Stal, L.J. (2000) Microbial mats and Stromatolites. In The Ecology of Cyanobacteria. Their Diversity in time and space. Whitton, B.A., and Potts, M. (eds). : Kluwer Academic Publishers, pp. 61–120. Stal, L.J., and Caumette, P. (1994) Microbial Mats: Structure, Development, and Environmental Significance. Berlin, Germany: Springer. Steunou, A.S., Bhaya, D., Bateson, M.M., Melendrez, M.C., Ward, D.M., Brecht, E., et al. (2006) In situ analysis of nitrogen fixation and metabolic switching in unicellular thermophilic cyanobacteria inhabiting hot spring microbial mats. Proc Natl Acad Sci USA 103: 2398–2403. Strickland, J.D.H., and Parsons, T.R. (eds) (1972) A practical handbook of seawater analysis.: fisheries research board of canada. Taton, A., Grubisic, S., Brambilla, E., De Wit, R., and Wilmotte, A. (2003) Cyanobacterial diversity in natural and artificial microbial mats of Lake Fryxell (McMurdo Dry Valleys, Antarctica): a morphological and molecular approach. Appl Environ Microbiol 69: 5157–5169. Vincent, W.F. (2000) Evolutionary origins of Antarctic microbiota: invasion, selection and endemism. Antarct Sci 12: 374–385. Ward, D.M., Ferris, M.J., Nold, S.C., and Bateson, M.M. (1998) A natural view of microbial biodiversity within hot spring cyanobacterial mat communities. Microbiol Mol Biol Rev 62: 1353–1370.

52 4 Heterotrophic pioneers facilitate phototrophic biofilm development

G. Roeselers, M.C.M. van Loosdrecht, and G. Muyzer

Microbial Ecology (2007) DOI: 10.1007/s00248-007-9238-x Chapter 4

ABSTRACT

Phototrophic biofilms are matrix-enclosed microbial communities, mainly driven by light energy. In this study, the successional changes in community composition of freshwater phototrophic biofilms growing on polycarbonate slides under different light intensities were investigated. The sequential changes in community composition during different developmental stages were examined by denaturing gradient gel electrophoresis (DGGE) of polymerase chain reaction (PCR)-amplified16S rRNA gene fragments in conjugation with sequencing and phylogenetic analysis. Biofilm development was monitored with subsurface light sensors The development of these biofilms was clearly light dependent. It was shown that under high light conditions the initial colonizers of the substratum predominantly consisted of green algae, whereas at low light intensities, heterotrophic bacteria were the initial colonizers. Cluster analysis of DGGE banding patterns revealed a clear correlation in the community structure with the developmental phases of the biofilms. At all light intensities, filamentous cyanobacteria affiliated to Microcoleus vaginatus became dominant as the biofilms matured. It was shown that the initial colonization phase of the phototrophic biofilms is shorter on polycarbonate surfaces precolonized by heterotrophic bacteria.

INTRODUCTION

Phototrophic biofilms are formed on surfaces in a range of terrestrial and aquatic environments (e.g., Chan et al., 2003; Jarvi et al., 2002; Ortega-Morales et al., 2000). The major energy source in these biofilms is photosynthesis. Aerobic diatoms, green algae and cyanobacteria use light energy and reduce carbon dioxide, providing organic substrates and oxygen. The photosynthetic activity fuels processes in the total biofilm community, including the heterotrophic fraction (Canfield and Des Marais 1993, Pearl et al., 2000). The microorganisms produce extracellular polymeric substances (EPS) that hold the biofilm together (8de Brouwer et al., 2005; Richert et al., 2005). There is a growing interest in the application of phototrophic biofilms, for instance in wastewater treatment in constructed wetlands, bioremediation (Schumacher et al., 2003; Walsby 2005), aquaculture (Bender and Phillips, 2004), and in the development of antifouling agents (Bhadury and Wright 2004; Callow and Callow 2002; Patil and Anil 2005). To be able to optimize biofilm development for specific applications it is key to understand the structure and functioning of phototrophic biofilms. The laboratory-based cultivation of phototrophic biofilms provides a valuable alternative for natural systems by allowing experimental manipulation of the entire microbial ecosystem. In this study, we addressed the formation of oxygenic phototrophic biofilms in the context of primary ecological succession. Successional community changes during the colonization of new habitats or after environmental disturbances have been described for communities of planktonic and benthic microalgae (Helbling et al., 2005; Schäfer et al., 2001), and more recently in bacterial communities (Ferris et al., 1997; Bhadury and Wright 2004; Martiny et al., 2003). Primary succesional development of an ecosystem is always initiated by the settlement of pioneer organisms in a previously uncolonized environment. In the present work, we have studied the early events of surface colonization to identify pioneer organisms that initiate phototrophic biofilm succession We cultivated biofilms in a specially developed flow-lane incubator with controlled external light, temperature, and flow velocity. In addition, cultivation experiments were

54 Heterotrophic pioneers

carried out in small flow-cell-type incubators. The growth of the biofilms was monitored continuously. DGGE of PCR-amplified SSU rRNA gene fragments in conjunction with Unweighted Pair Group Method with Arithmetic mean (UPGMA) clustering analysis, DNA sequencing and phylogenetic analysis were used to monitor changes in the community composition of the biofilms that developed under different light conditions.

MATERIALS & METHODS

Biofilm incubator The incubator used in this study contained four separate flow channels through which a volume of 4L medium circulated over a surface covered with 47 polycarbonate slides (76 x 25 x 1 mm). Polycarbonate slides were used as a substratum for biofilm adhesion. Each light chamber contained an adjustable light source and the circulation speed of the culture medium could be regulated precisely. The medium was refreshed twice a week. Biofilm growth was monitored and recorded with three light sensors that were positioned directly under selected polycarbonate slides. Each light sensor contained three independent photodiodes. Light absorbance was used as an indicator for biomass accumulation. The incubator design was described in more detail by Zippel & Neu (2005). In addition, aluminum flow-cell type incubators with a channel dimension of 4 x 26 x 108 mm, covered with polycarbonate lids were used. The aluminum body of the flow cells contained a water channel for temperature control. The transparent polycarbonate lids formed the substratum for biofilm growth.

Growth conditions and inoculum The mineral medium used was a modification of the BG11 as described by Stanier et al. (1971). Ammonium ferric citrate green was replaced by FeCl3. The vitamins cyanocobalamin (40 g/L), thiamine HCL (40 g/L), and biotin (40 g/L) were added.

NaSiO3 9H20 (57 mg/mL) was added to allow the growth of freshwater diatoms. Phototrophic biofilm samples obtained from a sedimentation tank of the wastewater treatment plant (WWTP) at Fiumicino Airport (Rome, Italy) were used as inoculum of biofilm development at the polycarbonate slides within the flow lane incubator. The community composition of these environmental phototrophic biofilms was described in detail by Albertano et al. (1999). In order to reduce predation pressure, inocula were frozen for two days at –20oC to kill protozoa, metazoa, and nematodes. The biofilms were grown at 20oC and at a medium flow rate of 100 L/h. The biofilms were grown under a photon flux density (PFD) of 15, 30, 60 and 120 Amol photons m–2 s–1. A diurnal cycle of 16h light and 8h dark was applied. The growth and development of the biofilms was monitored for 35 days. The influence of heterotrophic bacteria on the development of phototrophic biofilms was studied in separate biofilm cultivation experiments. Two identical flow-cells were inoculated with an environmental phototrophic biofilm sample collected from a helophyte pond system from the municipal WWTP at St. Maartensdijk (The Netherlands) and incubated for seven days in complete darkness. Flow-cells with modified BG11 medium were maintained at a constant temperature of 20oC, an irradiance of 60 Amol photons m–2 s–1, and a flow rate of 0.5 l/h. After seven days the transparent polycarbonate top of one flow cell (flow cell II) was replaced by a new sterile top. Both flow-cells were again inoculated and biofilm development on the polycarbonate top lid was documented over a period of 20 days.

55 Chapter 4

Image analysis of biofilm scans A CanoScan LiDE 60 flatbed scanner (Cannon Inc., Tokyo, Japan) was used to capture images of the polycarbonate incubated in the flow cells (I and II). Every day, the lids were removed from the flow-cells and gently placed on the glass surface of the flatbed scanner. Slides were covered with a white plastic tray to exclude external light. The resulting scans were saved as an 8-bit tagged image file format (TIFF). Image analysis was performed using ImageJ 1.36b [National Institutes of Health (NIH), Bethesda, MD, USA]. Lids were carefully placed back onto the flow-cells for further incubation. Mean and the standard deviation of the grey values of each image were calculated using the integrated measurement tool. It must be noted that the tone value of an 8-bit pixel ranges from 0 to 255, where 0 is the blackest black, and 255 is the whitest white.

Sample collection Biofilms grown under four different light conditions were sampled when they reached the initial colonization, exponential and mature phase of their development. These phases were defined as the times when the subsurface light sensors measured respectively 10, 50, and 80% light absorbance. One polycarbonate slide was removed from the incubator lanes at the exit side of the flow lane at each sampling event. The samples were immediately stored at -20oC for further analysis. Duplicate samples were examined by bright-field microscopy (x100 to x400 magnification) with a Zeiss Axioplan microscope (Carl Zeiss, Jena, Germany)

DNA extraction Biofilm biomass was scraped from the slides with a sterile razorblade. Genomic DNA was extracted by applying approximately 300 mg biomass to the UltraClean Soil DNA Isolation Kit (Mo Bio Laboratories, Carlsbad, CA, USA) according to the manufacturer’s protocol. Complete cell lysis was verified afterwards by using phase-contrast microscopy. The quantity and quality of the extracted DNA were analyzed by spectrophotometry using the NanoDrop ND-1000 (NanoDrop Technologies, Wilmington, DE, USA) and by agarose gel electrophoresis. DNA dilutions were stored at –20oC.

PCR amplification of rRNA gene fragments Extreme care was taken to prevent any DNA contamination of solutions and plastic disposables used for PCR. All heat sterilized plastic tubes were exposed to UV light for 30 minutes before use. Only DNA and RNA free water (W4502, Sigma-Aldrich, St. Louis, MO, USA) was used to prepare PCR reagent stock solutions and PCR reaction mixtures. To amplify the bacterial 16S rRNA encoding gene fragments, the DNA dilutions were used as template DNA in 50 l PCR reactions using the primers 341F-GC and 907R, and PCR conditions as described by Schäfer et al. (2001). This PCR was carried out with a denaturation step of 5 minutes at 94oC, followed by 35 cycles of denaturation of 1 min at 94oC, annealing of 1 min at 60oC, and extension of 1 min at 72oC, followed by a final extension step of 15 minutes at 72oC. All amplification reactions were performed in a T1 Thermocycler (Biometra, Westburg, the Netherlands).

Denaturing gradient gel electrophoresis of PCR products DGGE was performed as described by Schäfer & Muyzer (2001). Briefly, one mm thick 6% acrylamide gels with a urea-formamide gradient of 20-80% were used for bacterial 16S rRNA gene fragments. Gels were run in Tris-acetate-EDTA buffer for 16 hrs at 100 V and at a constant temperature of 60o C. Gels were stained in an ethidium bromide solution

56 Heterotrophic pioneers

and analyzed and photographed using the GelDoc UV Transilluminator (Bio-Rad, Hercules, CA, USA). The dominant bands were excised from the DGGE gels with a sterile surgical scalpel. Each small gel slice was placed in 15 Cl of sterile water for 24 h at 4o C. Subsequently, 2 Cl of the solution was used as template DNA for re-amplification as described above. The PCR products were again subjected to DGGE analysis to confirm the purity and their position relative to the bands from which they were originally excised. The PCR products were purified using the QIAquick Gel Extraction Kit (QIAGEN, Hilden, Germany). Purified PCR products were sequenced by a commercial company (BaseClear, Leiden, the Netherlands). DGGE profiles were compared visually on the basis of the presence and relative density of bands. In addition, the presence and absence of DGGE bands in different samples were scored in binary matrices. This binary matrix was translated into a distance matrix using the Jaccard coefficient. A dendrogram was then constructed by the UPGMA clustering method (Griffiths eta l., 2000). Jaccard coefficient and UPGMA calculations were done with the software package Primer 6 (PRIMER-E Ltd, Plymouth, UK).

Comparative sequence analysis Partial 16S rRNA gene sequences with a length between 400 and 500 bp were first compared to the sequences stored in the Genbank nucleotide database using the BLAST algorithm (Altschul et al., 1990) in order to obtain a first identification of the biofilm community members. Subsequently, the sequences were imported into the ARB SSU rRNA database (available at http://www.arb-home.de) (Ludwig et al., 2004). The dissimilarity values were used to calculate distance matrixes. Distance matrix trees were generated by the Neighbour-Joining (NJ) method with the Felsenstein correction as implemented in the PAUP 4.0B software (Sinauer, Sunderland, MA, USA). The NJ calculation was subjected to bootstrap analysis (1000 replicates). The DNA sequences obtained in this study were deposited in the GenBank Nucleotide database and assigned accession numbers DQ388949 to DQ388967.

RESULTS

Biofilm growth was monitored by the decrease of subsurface light. The duration of the initial colonization phase after inoculation was ~8 days in the biofilms grown at a photon flux density (PFD) of 60 and 120 mol photons m–2 s–1 (Fig. 1). At 30 Cmol photons m–2 s–1, the initial phase lasted until ~22 days after inoculation. The biofilm growing at 15 Cmol photons m–2 s–1 did not show an exponential growth phase. Therefore, the second and third PFD 15 samples were collected respectively at day 25 and day 35. The fluctuations in light absorbance after higher light biofilms reached their mature phase, were caused by detachment and subsequent recolonization of parts of the biofilm, also known as sloughing (van Loosdrecht et al., 1995).

57 Chapter 4

Figure 1. Growth curves of phototrophic biofilms growing at four incident light intensities (120, 60, 30, and 15 #mol photons m–2 s–1). Biofilm development is indicated as the increasing light absorbance derived from the decrease of subsurface light.

The average dry weight per slide at the end of the growth experiment was approximately 0.15 g in the PFD 60 and PFD 120 lanes, about 0.06 g in the PFD 30 lane and 0.008 g in the PFD 15 lane. This shows that the sensors are suitable for monitoring biomass increase, but they are not suitable as indicators for the actual biomass. The DGGE band patterns describing bacterial community diversity during the initial, exponential, and mature phase of biofilm development under the four different light intensities were all different. The initial phase DGGE patterns from the high light intensity biofilms (PFD 60 and 120) shared two dominant bands (Fig.2, band 1 and 4). Band 1 was affiliated to the chloroplast of Scenedesmus obliquus and band 4 was affiliated to an unidentified member of the Bacteroidetes (formerly known as the Cytophaga-Flavobacteria- Bacteroides (CFB) group). The presence of Scenedesmus-like unicellular algae in the initial phase at high light intensity was confirmed by microscopic observations. Bands at the same position in the gel and with the same phylogenetic affiliation as band 1 were also present during the exponential and mature phase at high light intensities. During the exponential phase, thick dominant bands appeared at similar positions in the DGGE profiles of the PFD 60, and PFD120 biofilms. These bands represented DNA sequences affiliated to the filamentous cyanobacterium Microcoleus vaginatus, a known inhabitant of desert soil crusts (Garcia- Pichel et al., 2001) (Fig. 2, bands 8, 12, 13, and 17 and Fig. 3). However, the bands corresponding to Scenedesmus obliquus were also present during the mature phase. Light microscopic observations confirmed that during the exponential and mature phase, the top layer of the biofilm consisted predominantly of Microcoleus filaments, which covered the Scenedesmus-like cells.

58 Heterotrophic pioneers

Figure 2. DGGE patterns of 16S rRNA gene fragments obtained after enzymatic amplification using general bacterial primers and genomic DNA samples from phototrophic biofilms growing at four incident light intensities (120, 60, 30, and 15 .mol photons m–2 s–1). Samples were collected during the initial, exponential and mature phases of biofilm development. Numbers at the left of each lane correspond to bands that were excised, PCR amplified, and sequenced.

The initial colonization phase of the lower light intensity biofilms (15 and 30 5mol) was characterized by the presence of one dominant band affiliated to the -proteobacteria. The exponential phase DGGE pattern of the PFD 30 biofilm showed three dominant bands corresponding to Scenedesmus obliquus, Microcoleus vaginatus, and Synechocystis sp. The pattern from the mature phase PFD 30 samples showed two dominant bands corresponding to sequences affiliated to Microcoleus vaginatus, and Synechocystis sp. (Fig. 2, bands 16 and 17). The Scenedesmus obliquus band had almost completely disappeared. After 35 days the PFD 15 biofilms showed bands affiliated to unicellular Synechocystis- like cyanobacteria. (Fig. 2, band 19 and Fig. 3). The presence of unicellular cyanobacteria was confirmed by light microscopy. The UPGMA dendogram (Fig. 4) constructed from the DGGE profiles shows the highest similarity between samples from similar developmental phases. It is remarkable that the dendogram shows one big cluster with the initial phase DGGE patterns and one cluster with the exponential and mature phase DGGE patterns. However, DGGE patterns from the three PFD 15 biofilm samples were all in one cluster.

59 Chapter 4

Figure 3. Evolutionary tree showing the phylogenetic affiliations as revealed by comparative analysis of the general bacterial 16S rRNA gene sequences derived from the DGGE bands shown in Figure 2. The sequences obtained in this study are printed bold. Escherichia coli was used as an out-group, but was pruned from the tree. Bootstrap values based on 1000 replicates are indicated for branches supported by >50% of trees. Scale bar indicates 10% estimated sequence divergence.

Figure 4. UPGMA dendogram showing the combined clustering analyses of digitized DGGE profiles (Fig. 2). The analysis is based on the presence or absence of bands at certain positions in each lane of the gel. *The biofilm cultivated at 15 6mol photons m–2 s–1 did not reach 50 and 80% light absorbance.

60 Heterotrophic pioneers

In order to get a better understanding of the role of heterotrophs in the phototrophic biofilm, a separate cultivation experiment was conducted using small flow-cells. Analysis of surface images obtained with the scanner showed that the more biomass developed on the flow-cell lid, the darker the image would appear. Although the actual biomass was not determined we assumed that changing average grey values could be used as an indirect indication of biofilm growth. A similar biofilm analysis method was applied by Milferstedt et al. (2006). In this study it was shown that average biofilm biomass is highly correlated with the grey levels of biofilm images obtained with a scanner. Biofilm growth on polycarbonate surface pre-incubated in the dark was compared with that on pristine polycarbonate surface. Mean grey values calculated for each slide scan showed that the duration of the initial colonization phase on the slides that were pre- incubated in the dark was shorter than on new sterile slides (Fig. 5 and 6A). Replicate experiments showed the same trend, although the rate at which biofilms developed was variable (Fig. 6B). Microscopic examination confirmed that after 7 days of preincubation in the dark, the polycarbonate lids were colonized by heterotrophic bacteria.

Figure 5. Biofilm development on polycarbonate incubated in two flow cells. Both flow cells were pre-incubated for 7 days with inoculated BG11 medium in complete darkness. At day 8, the polycarbonate lid of flow cell II was changed. Subsequently, both flow-cells were subjected to 60 1mol photons m–2 s–1 irradiation for 19 days. Column I (flow-cell I) and II (flow cell II) show images of the lids that were scanned at different time points. The mean and standard deviation (SD) of the pixel grey values in the scanned area are indicated under each image. It must be noted that the grey value of an 8-bit pixel ranges from 0 to 255.

61 Chapter 4

Figure 6. Biofilm development on polycarbonate incubated in two flow cells. A) Inverse of the mean grey values of biofilm images (Figure 5) captured over a period of 19 days: () Flow cell I. ( ) Flow cell II. B) Replicate experiments showing biofilm development (inverse grey values) on pre-colonized polycarbonate (Flow cell I, white) and pristine polycarbonate (Flow cell II, black) after 1, 10 and 19 days of incubation. Replicates are indicated as 1, 2, and 3.

DISCUSSION

Our results show that the cultivated biofilms were inhabited by a phylogenetically diverse array of microorganisms. We found sequences affiliated to eukaryote chloroplasts, cyanobacteria, -proteobacteria, and the Bacteroidetes group. The community composition of the cultivated phototrophic biofilms clearly changed over time. The daily increase in light absorbance during the exponential phase was similar for the light intensities of 30, 60 and 120 ,mol photons m–2 s–1. We only observed that the duration of the initial colonization phase increased when light intensities decreased. We think that cyanobacteria are responsible for the exponential growth of the biofilm, because we observed dominant DGGE bands representing filamentous cyanobacterial species in the

62 Heterotrophic pioneers

profiles of the exponential and mature phase biofilms. The UPGMA analysis of the DGGE patterns shows one cluster with the initial phase biofilms and one cluster with the exponential and mature phase biofilms. This indicates that these biofilm communities experienced similar changes in biodiversity during their development. The fact that the DGGE patterns from the three developmental phases of the low light biofilm (PFD 15) clustered together in the UPGMA dendogram suggests that the low light biofilm did not really mature during the experimental time. The phylogenetic analysis of DGGE bands showed that at high light intensities unicellular algae affiliated to Scenedesmus sp. attached first to the polycarbonate substratum. It seems that as soon as the surface was covered with some biomass Microcoleus vaginatus strains were able to overgrow the biofilm resulting in an exponential increase in biofilm thickness. At lower light intensities, the polycarbonate surface was initially colonized by -proteobacteria. Previous studies showed that heterotrophic bacteria play an important role as early colonizers during biofilm development on submerged glass slides in a marine environment (Chan et al., 2003). The exponential growth of the PFD 30 biofilm was also caused by the rapid growth of M. vaginatus-like cyanobacteria. However, the exponential and mature stages of the PFD 15 and PFD 30 biofilms also contained unicellular Synechocystis-like cyanobacteria. It could be hypothesized that filamentous cyanobacteria outcompete unicellular cyanobacteria for light by their ability to migrate towards optimal light conditions. At lower light intensities this advantage could be reduced because of a lower light-affinity coefficient (Staal et al., 2002; Walsby 2005). It is possible that the succession observed in the large flow lane incubator mainly occurred because conditions in the biofilm changed, largely because of microbial activity and growth itself. Early colonizers might change the environmental conditions in the biofilm, e.g. by altering the substratum surface (Li et al., 2004; Martiny et al., 2003), and hence facilitating the attachment of filamentous cyanobacteria. Another explanation for the dominance of cyanobacteria in the exponential and mature phase biofilms could be the different critical light intensities of algae and cyanobacteria (Weising and Huisman 1994). The initially fast growing Scenedesmus strains may experience increasing light limitation resulting from self-shading. Although cyanobacteria cannot reach the maximum growth rates of green algae, they show higher growth rates at low light intensities. Competition experiments in light-limited chemostats with cultures of Aphanizomenon, Microcystis and Scenedesmus sp. by Huisman et al. (1999) showed that the cyanobacteria outcompete Scenedesmus for light, independent of their initial abundance. This critical light intensity effect has not been described before for benthic communities. An important difference between continuous suspended cultures and cultivated biofilms is that the dilution rate of the reactor imposes no artificial loss factor on biofilm communities. Our separate cultivation experiments with the flow cells seem to confirm the importance of heterotrophic pioneers (Chan et al., 2003). Analysis of the polycarbonate surface scans suggests that colonization by heterotrophic bacteria enhances the establishment of a phototrophic biofilm. However, it is also possible that the observed difference in biofilm growth is a result of surface conditioning by the association of organic compounds to the polycarbonate. Because the mineral medium used in the incubator contained no organic carbon, it seems that heterotrophic pioneers utilized organic compounds that were present in the inoculum. The slow growth of heterotrophs at relatively low carbon conditions could explain the long initial phase at low light intensities. It is possible that at higher light intensities

63 Chapter 4

photototrophic pioneers, like Scenedesmus sp. grow faster and provide the system with sufficient carbon. In conclusion, our results show that the development of phototrophic biofilms under controlled conditions involves a series of successional community changes. The influence of light on the biofilm growth rate and the community composition appeared more prevalent during the initial phase of biofilm development. Phototrophic biofilms developed faster on polycarbonate surfaces that were precolonized by heterotrophic bacteria. These data illustrate that although a mature biofilm community can be dominated by a few species, there is also a minor set of organisms, e.g. heterotrophic bacteria, that may be crucial for the initial establishment and the consequent development of that biofilm community. These results may have implications for the cultivation of phototrophic biofilms in wastewater treatment applications and for control of biofouling dominated by phototrophic biofilms. Future experiments should focus on the mechanisms by which heterotrophic precolonization enhances phototrophic biofilm development.

ACKOWLEDGEMENTS

This research was supported by the European Union (PHOBIA project, contract QLK3-CT- 2002-01938). We thank Marc Staal (NIOO-KNAW Centre for Estuarine and Marine Ecology, the Netherlands) for the design of the flow-cell incubator. We thank our co-workers in the Environmental Biotechnology group for their careful reading and commenting on the manuscript.

REFERENCES

Albertano, P., Congestri R, Shubert LE (1999) Cyanobacterial biofilms in sewage treatment plants along the Thyrrenian coast (Mediterranean Sea), Italy. Arch Hydrobiol Suppl Algol Stud 94: 13-24 Altschul, S.F., Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment search tool. J Mol Biol 215: 403-410 Bender, J., Phillips P (2004) Microbial mats for multiple applications in aquaculture and bioremediation. Bioresour Technol 94: 229-238 Bhadury, P., Wright PC (2004) Exploitation of marine algae: biogenic compounds for potential antifouling applications. Planta 219: 561-578 Callow, M.E., Callow JE (2002) Marine biofouling: a sticky problem. Biologist (London) 49: 10-14 Canfield, D.E., Des Marais DJ (1993) Biogeochemical cycles of carbon, sulfur, and free oxygen in a microbial mat. Geochim Cosmochim Acta 57: 3971-3984 Chan, B.K., Chan WK, Walker G (2003) Patterns of biofilm succession on a sheltered rocky shore in Hong Kong. Biofouling 19: 371-380 de Brouwer, J.F., Wolfstein K, Ruddy GK, Jones TE, Stal LJ (2005) Biogenic stabilization of intertidal sediments: the importance of extracellular polymeric substances produced by benthic diatoms. Microb Ecol 49:501-512 Ferris, M.J., Nold SC, Revsbech NP, Ward DM (1997) Population structure and physiological changes within a hot spring microbial mat community following disturbance. Appl Environ Microbiol 63: 1367- 1374 Garcia-Pichel, F., Lopez-Cortes A, Nübel U (2001) Phylogenetic and morphological diversity of cyanobacteria in soil desert crusts from the Colorado plateau. Appl Environ Microbiol 67: 1902-1910 Griffiths, R.I., Whiteley AS, O'Donnell AG, Bailey MJ (2000) Rapid method for coextraction of DNA and RNA from natural environments for analysis of ribosomal DNA- and rRNA-based microbial community composition. Appl Environ Microbiol 66: 5488-5491 Helbling, E.W., Barbieri ES, Marcoval MA, Goncalves RJ, Villafane VE (2005) Impact of solar ultraviolet radiation on marine phytoplankton of Patagonia, Argentina. Photochem Photobiol 81: 807-818

64 Heterotrophic pioneers

Huisman, J., Jonker RR, Zonneveld C, Weissing FJ (1999) Competition for light between phytoplankton species: experimental tests of mechanistic theory. Ecology 80: 211-222 Jackson, C.R. (2003) Changes in community properties during microbial succession. Oikos 101: 444- 448 Jarvi, H.P., Neal C, Warwick A, White J, Neal M, Wickham HD, Hill LK, Andrews MC (2002) Phosphorus uptake into algal biofilms in a lowland chalk river. Sci Total Environ 282: 353-373 Li, J., Helmerhorst EJ, Leone CW, Troxler RF, Yaskell T, Haffajee AD, Socransky SS, Oppenheim FG (2004) Identification of early microbial colonizers in human dental biofilm. J Appl Microbiol 97: 1311- 1318 Ludwig, W., Strunk O, Westram R, Richter L, Meier H, Kumar, Y, 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 Acids Res 32: 1363-1371 Martiny, A.C., Jorgensen TM, Albrechtsen HJ, Arvin E, Molin S (2003) Long-term succession of structure and diversity of a biofilm formed in a model drinking water distribution system. Appl Environ Microbiol 69: 6899-6907 Milferstedt, K., Pons MN, Morgenroth E (2006) Optical method for long-term and large-scale monitoring of spatial biofilm development. Biotechnol Bioeng 94:773-782 Ortega-Morales, O., Guezennec J, Hernandez-Duque G, Gaylarde CC, Gaylarde PM (2000) Phototrophic biofilms on ancient Mayan buildings in Yucatan, Mexico. Curr Microbiol 40: 81-85 Paerl, H.W., Pinckney JL, Steppe TF (2000) Cyanobacterial-bacterial mat consortia: examining the functional unit of microbial survival and growth in extreme environments. Environ Microbiol 2: 11-26 Patil, J.S., Anil AC (2005) Biofilm diatom community structure: influence of temporal and substratum variability. Biofouling 21: 189-206 Richert, L., Golubic, S, Guedes, R L, Ratiskol J, Payri C, Guezennec J (2005) Characterization of exopolysaccharides produced by cyanobacteria isolated from Polynesian microbial mats. Curr Microbiol 51:379-384 Schäfer, H., Bernard L, Courties C, Lebaron P, Servais P, Pukall R, Stackebrandt E, Troussellier M, Guindulain T, Vives-Rego J, Muyzer G (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., Muyzer G (2001) Denaturing gradient gel electrophoresis in marine microbial ecology. In: Paul JH (ed.) Methods in Microbiology, Marine Microbiology. Academic Press, New York (Methods in Microbiology, vol 30, pp 425-468) Schumacher, G., Blume T, Sekoulov I (2003) Bacteria reduction and nutrient removal in small wastewater treatment plants by an algal biofilm. Water Sci Technol 47: 195-202 Staal, M., te Lintel Hekkert S, Herman P, Stal LJ (2002) Comparison of models describing light dependence of N2 fixation in heterocystous cyanobacteria. Appl Environ Microbiol 68: 4679-4683 Stanier, R.Y., Kunisawa R, Mandel M, Cohen-Bazire G (1971) Purification and properties of unicellular blue-green algae (order Chroococcales). Bacteriol Rev 35: 171-205 van Loosdrecht, M.C.M., Eikelboom D, Gjaltema A, Mulder A, Tijhuis L, Heijnen JJ (1995) Biofilm structures. Water Sci Technol 32: 35-43 Vymazal, J., Sladedek V, Stach J (2001) Biota participating in wastewater treatment in a horizontal flow constructed wetland. Water Sci Technol 44: 211-214 Walsby, A.E. (2005) Stratification by cyanobacteria in lakes: a dynamic buoyancy model indicates size limitations met by Planktothrix rubescens filaments. New Phytol 168: 365-376 Weissing, F.J., Huisman J (1994) Growth and competition in a light gradient. J Theor Biol 168: 323-336 Zippel, B., Neu TR (2005) Growth and structure of phototrophic biofilms under controlled light conditions. Water Sci Technol 52: 203-209

65

5 On the reproducibility of microcosm experiments – different community composition in parallel phototrophic biofilm microcosms

Guus Roeselers, Barbara Zippel, Marc Staal, Mark C. M. van Loosdrecht, and Gerard Muyzer.

FEMS Microbiology Ecology (2006) 58:169-178 Chapter 5

ABSTRACT

Phototrophic biofilms were cultivated simultaneously using the same inoculum in three identical flow-lane microcosms located in different laboratories. The growth rates of the biofilms were similar in the different microcosms, but denaturing gradient gel electrophoresis (DGGE) analysis of both 16S and 18S rRNA gene fragments showed that the communities developed differently in terms of species richness and community composition. One microcosm was dominated by Microcoleus and Phormidium species, the second microcosm was dominated by Synechocystis and Phormidium species, and the third microcosm was dominated by Microcoleus- and Planktothrix-affiliated species. No clear effect of light intensity on the cyanobacterial community composition was observed. In addition, DGGE profiles obtained from the cultivated biofilms showed a low resemblance with the profiles derived from the inoculum. These findings demonstrate that validation of reproducibility is essential for the use of microcosm systems in microbial ecology studies.

INTRODUCTION

Microcosms are constructed, simplified ecosystems that are used to mimic natural ecosystems under controlled conditions. They provide an experimental area for ecologists to study natural processes. Hence, microcosm studies can be very useful to study the effects of disturbance or to determine the role of key species. These simplified systems can still contain a high diversity of species and therefore require self-organizing processes to reach and maintain system stability (Odum, 1989; Kangas and Adey, 1996). Reproducibility is indisputably essential to validate the use of microcosm systems in microbial ecology studies. Hence, it is surprising that reproducibility assessments are scarcely documented in the literature (e.g. Heydorn et al., 2000). In the present study, cultivated phototrophic biofilms were chosen as a model to assess reproducibility. Structure, growth dynamics and physiology of heterotrophic biofilms have been extensively studied. But until recently phototrophic biofilms have received little attention for this aspect. Phototrophic biofilms occur on contact surfaces in a range of terrestrial and aquatic environments. They can best be described as surface-attached microbial communities driven by light as energy source. Diatoms, green algae and cyanobacteria are the major primary producers that generate energy and reduce carbon dioxide, providing organic substrate and oxygen. Their oxygenic photosynthetic activity fuels metabolic processes and conversions in the entire biofilm community, including the heterotrophic fraction (Paerl et al., 2000). The microorganisms produce extracellular polymeric substances (EPS) that hold the biofilm together (Wimpenny et al., 2000; Cogan and Keener, 2004). There is a growing interest in the application of phototrophic biofilms, e.g. bioremediation (Schumacher et al., 2003), aquaculture (Bender and Phillips, 2004) and biohydrogen production (Prince and Kheshgi, 2005). The study of artificial phototrophic biofilms may also increase our understanding of the development of more complex phototrophic biofilms, such as microbial mats and stromatolites (Des Marais, 1990). In order to enhance our understanding of the complex phototrophic biofilm physiology, the individual community members should not be studied separately. Since biofilm communities in nature are often difficult to investigate and experimental conditions are ambiguous, a number of different laboratory-based experimental biofilm model systems have been developed (Palmer, 1999; Heydorn et al.,

68 On the reproducibility of microcosm experiments

2000; Jackson et al., 2001). In most cases these systems were used to study single-species biofilms or mixed biofilms with a predefined community composition. In the current study identical open flow-lane incubator systems (i.e. microcosms) located at three different laboratories in Europe were inoculated with the same environmental biofilm sample at the same time and operated under exactly the same conditions. Biomass samples were collected from the three different incubators at the moment that the biofilms reached their mature stage. In addition, replicate biofilm growth experiments with the same inoculum were carried out in small flow cell incubators located in one laboratory. Bacterial and eukaryotic community compositions were compared using denaturing gradient gel electrophoresis (DGGE) (Schäfer and Muyzer, 2001) of PCR-amplified small subunit rRNA (SS rRNA) gene fragments in conjunction with DNA sequencing and phylogenetic analysis.

MATERIALS AND METHODS

Biofilm incubator. The flow-lane incubator systems ('large incubators') used in this study contained four separate flow channels through which a volume of 4 L medium circulated over a surface covered with 47 polycarbonate slides (76 _ 25 _ 1 mm). The polycarbonate slides were used as a substratum for biofilm adhesion. Each light chamber contained an adjustable light source and the circulation speed of the culture medium could be regulated precisely. The medium was refreshed twice a week. Biofilm growth was monitored and recorded with three light sensors that were positioned directly under selected polycarbonate slides. Each light sensor contained three independent photodiodes. Decrease of subsurface light below the substratum was used as an indicator for biomass accumulation. The incubator design was described in more detail by Zippel and Neu (2005). Identically designed and produced biofilm incubators used in this study were located at the Department of Biotechnology of Delft University of Technology in the Netherlands, at the Department of River Ecology of the UFZ Centre for Environmental Research in Magdeburg (Germany), and at the Netherlands Institute of Ecology in Yerseke (the Netherlands). Two small flow cell-type incubators with a glass top were used for reproducibility experiments in the laboratory at Delft University of Technology. Each flow cell contained one polycarbonate slide. The aluminium body of the flow cells contained a water channel for temperature control.

Growth conditions and inoculum. The mineral medium used was a modification of BG11 as described by Stanier et al., (1971). Ammonium ferric citrate green was replaced by FeCl3. The vitamins cyanocobalamin (40 g L-1), thiamine HCL (40 g L-1) and biotin (40 g L-1) were added. NaSiO3·9H2O (57 mg mL-1) was added to allow the growth of freshwater diatoms (Guillard and Hargraves, 1993). Phototrophic biofilm material was collected from an overflow weir of the sedimentation basin of the wastewater treatment plant (WWTP) at Fiumicino Airport (Rome) (Albertano et al., 1999). The biomass was homogenized and aliquoted into 50 mL tubes. In order to reduce predation pressure, the biomass was frozen at -20°C to kill protozoa, metazoa and nematodes. The aliquots were shipped on dry ice to the three laboratories. Care was taken that all tubes remained frozen for the same period. The biofilm biomass was

69 Chapter 5 used to inoculate the polycarbonate slides within the four flow lanes in each laboratory on the same day. Biofilms were grown at 30°C and at a medium flow rate of 100 l h-1. The biofilms were grown under irradiances of 60 and 120 mol photons m-2 s-1. A diel light cycle of 16 h light : 8 h dark was applied. The growth and development of the biofilms was monitored for 34 days. Biofilm cultivation experiments with the flow cell incubators were carried out at a constant temperature of 30°C, an irradiance of 100 mol photons m-2 s-1 and a medium flow rate of 0.5 l h-1.

Sample collection. Biofilms grown under different light conditions in the three large incubators were sampled when they reached the mature stage of their development. This stage was defined as the moment when the average light intensity measured by the three submerged light sensors was <10% of the applied light. One polycarbonate slide was removed from the incubator lanes at the exit side of the flow lane at each sampling event. After 10 days the polycarbonate slides were removed from two flow cell incubators which were operated in parallel to each other. Another two slides were removed from two flow cell incubators that were operated for 20 days. Slides were immediately frozen and stored at -20°C. To eliminate potential variability introduced by analysis at different laboratories, samples collected from incubators in Magdeburg and Yerseke were shipped on dry ice to the Department of Biotechnology at Delft University of Technology for further analysis.

DNA extraction. Biofilm biomass was scraped from the polycarbonate slides with a sterile razor blade. Genomic DNA was extracted by applying c. 300 mg biomass to the UltraClean Soil DNA Isolation Kit (Mo Bio Laboratories, Carlsbad, CA) according to the manufacturer's protocol. Complete cell lysis was verified afterwards using phase-contrast microscopy. The quantity and quality of the extracted DNA was analyzed by spectrophotometry using the NanoDrop ND-1000 (NanoDrop Technologies, Delaware) and by agarose gel electrophoresis. DNA dilutions were stored at -20°C.

PCR amplification of rRNA gene fragments. Extreme care was taken to prevent any DNA contamination of solutions and plastic disposables used for PCR. All heat-sterilized plastic tubes were exposed to UV light for 30 min before use. Only DNA- and RNA-free water (W4502, Sigma-Aldrich, St Louis, MO) was used to prepare PCR reagent stock solutions and PCR reaction mixtures. To amplify the bacterial 16S rRNA-encoding gene fragments, the DNA dilutions were used as template DNA in 50 l PCR reactions using the primers 359F-GC and 907R, and PCR conditions as described by Schäfer et al. (2001). This PCR was carried out with a denaturation step of 5 min at 94°C, followed by 35 cycles of denaturation of 1 min at 94°C, annealing of 1 min at 60°C, and extension of 1 min at 72°C, followed by a final extension step of 10 min at 72°C. To amplify the 16S rRNA-encoding gene fragments of cyanobacteria we used the universal primer 359F-GC and an equimolar mixture of the reverse primers 781R(a) and 781R(b), and PCR conditions as described by Nübel et al. (1997). To amplify eukaryotic 18S rRNA encoding-gene fragments we used the EukA-f and Euk516-r+GC primers (Diez et al., 2001). This PCR was carried out with a denaturation step of 15 min at 94°C, followed by 33 cycles of denaturation of 1 min at 94°C, annealing of 1 min

70 On the reproducibility of microcosm experiments at 55°C, and extension of 3 min at 72°C, followed by a final extension step of 10 min at 72°C. All amplification reactions were performed in a T1 Thermocycler (Biometra, Westburg, the Netherlands).

DGGE of PCR products. DGGE was performed as described by Schäfer and Muyzer (2001). Briefly, 1 mm thick 6% acrylamide gels with a urea-formamide (UF) gradient of 20–80% were used for bacterial 16S rRNA gene fragments. Gradients of 20–60% were used for 18S rRNA gene fragments. An acrylamide gel without UF was cast on top of the gradient gel to obtain good loading slots. From each PCR reaction 20 L product, containing c. 1 g DNA, was mixed with 6 l of 10 x gel loading solution and loaded onto the gel. Gels were run in 1 x TAE (Tris- acetate-EDTA buffer, 50 M stock solution: 242 g of Tris base, 57.1 ml of glacial acetic acid, 100 mL of 0.5 M EDTA pH 8.0) for 16 h at 100 V and at a constant temperature of 60°C. Gels were stained in an ethidium bromide solution and analyzed and photographed using the GelDoc UV Transilluminator (Bio-Rad, Hercules, CA). The dominant bands were excised from the DGGE gels with a sterile surgical scalpel. Each small gel slice was placed in 15 l of sterile water for 24 h at 4°C. Subsequently, 2 l of the solution was used as template DNA for re-amplification as described above. The PCR products were again subjected to DGGE analysis to confirm their purity and position relative to the bands from which they were originally excised. The PCR products were purified using the QIAquick PCR Purification Kit (QIAGEN, Hilden, Germany). The purified PCR products were sequenced on an ABI 3730 sequencer (Applied Biosystems, Foster City, CA) by a commercial company (BaseClear, Leiden, the Netherlands). The sequencing reactions were carried out with the appropriate specific forward primers without GC clamp. DGGE profiles were compared visually on the basis of the presence and relative density of bands. In addition, the presence and absence of DGGE bands in different samples were scored in binary matrices. The binary matrices derived from the three DGGE gels were combined into one matrix. This binary matrix was translated into a distance matrix using the Jaccard coefficient. A dendrogram was then constructed by the UPGMA (Unweighted Pair Group Method with Arithmetic mean) clustering method (Griffiths et al., 2000). Jaccard coefficient and UPGMA calculations were carried out with the software package Primer 6 (PRIMER-E Ltd, Plymouth, U.K.).

Comparative sequence analysis. Partial 16S and 18S rRNA gene sequences with lengths of between 400 and 500 bp were first compared to the sequences stored in the Genbank nucleotide database using the blast algorithm (Altschul et al., 1990) in order to obtain a tentative identification of the biofilm community members. Subsequently, the sequences (including closest blast hits) were imported into the ARB SS rRNA database (available at http://www.arb-home.de) (Ludwig et al., 2004) and aligned based on the secondary structure of the SS rRNA. The dissimilarity values were used to calculate distance matrices. Distance matrix trees were generated by the Neighbour-Joining (NJ) method with the Felsenstein correction as implemented in the PAUP 4.0B software (Sinauer, Sunderland, MA). The NJ calculation was subjected to bootstrap analysis (1000 replicates). DNA sequences were deposited in the GenBank Nucleotide database and assigned accession numbers DQ366036–DQ366083.

71 Chapter 5

RESULTS

Biofilm growth rate. The biofilm growth rate was monitored by the decrease of subsurface light. The lag-time after inoculation was ~4 days, and visible growth of the phototrophic biofilms started between 4 and 8 days after inoculation. It was found that the growth rates were in each incubator higher at 120 photons mol m-2 s-1. Phototrophic biofilms growing at 120 mol photons m-2 s-1 reached their mature stage c. 5 days earlier than the biofilms incubated at lower light intensities. The biofilm growing at 120 mol photons m-2 s-1 in Magdeburg grew fastest and reached its mature stage within 12 days of inoculation (Fig. 1).

Figure 1. Growth curves of phototrophic biofilms growing in three incubators at two incident light intensities (60 and 120 mol photons m-2 s-1). Biofilm development is indicated as the increasing light absorbance derived from the decrease of subsurface light.

DGGE and phylogenetic analysis of Bacteria. The bacterial 16S rRNA gene-DGGE profiles from the cultivation experiments in the two flow-cell incubators were nearly identical. The profiles from the 10-day-old biofilms from both flow cells consisted of one dominant band and two faint bands. Both profiles from the 20-day-old biofilm showed one dominant band and several very faint bands (Fig. 2). The DGGE profiles obtained from the inoculum and the cultivation experiments with the large incubator setups located in three different laboratories (Delft, Magdeburg and Yerseke) were also compared. The bacterial 16S rRNA gene sequence DGGE profiles (Fig. 3A) revealed little similarity between the inoculum and the cultivated biofilms. The inoculum profile contained only three dominant bands. The top band in this profile (Fig. 3A, band 1) was affiliated to chloroplasts of a Scenedesmus-like alga. The second and the third band (Fig. 3A, bands 2 and 3) were affiliated to Phormidium-like cyanobacteria.

72 On the reproducibility of microcosm experiments

Figure 2. DGGE patterns of 16S rRNA gene fragments obtained after enzymatic amplification using general bacterial primers and genomic DNA samples from 10- and 20-day-old phototrophic biofilms cultivated in two flow cell incubators (A and B).

The DGGE profiles from the Delft biofilms contained five dominant bands more than the inoculum profile, suggesting a higher biodiversity (Fig. 3A). Both profiles from the Delft incubator were very similar although the 120 mol m-2 s-1 profile contained two bands (Fig. 3A, bands 5 and 8) that were absent in the 60 mol m-2 s-1 profile. Band 5 was affiliated to Cytophaga-like bacteria and band 8 was affiliated to a deep-branching unidentified bacterium (Fig. 4). The bacterial 16S rRNA gene-DGGE profiles from the Yerseke biofilms were highly similar for both light intensities. The high light profile contained one band that was absent in the low light profile. Two other dominant bands, which were present in both profiles, were affiliated to Cyanobacterium stanieri species (Fig. 3A; band 11) and Erythrobacter longus species, aerobic bacteria that contain bacteriochlorophyll a (Fig. 3a; band 12, and Fig. 4). The top band in both profiles from the Magdeburg biofilms had an identical position to the top bands in both Delft profiles. These bands (Fig. 3A; band 4, band 9 and band 13) showed affiliation to the chloroplasts of Scenedesmus (Fig. 4). The high light profile from Magdeburg contained one dominant band that was absent in the low light profile (Fig. 3A; band 14) and one dominant band that was only faintly visible in the low light profile (Fig. 3A; band 15). Furthermore, one dominant band that was present in the low light profile was absent in the high light profile (Fig. 3A; band 18).

73 Chapter 5

74 On the reproducibility of microcosm experiments

Figure 4. Evolutionary tree showing the phylogenetic affiliations as revealed by comparative analysis of the bacterial 16S rRNA gene sequences derived from the DGGE bands shown in Figure 3A. The sequences obtained in this study are printed bold. Escherichia coli (AJ567606) was used as an out-group, but was pruned from the tree. Accession numbers of sequences are noted behind the taxon names. Scale bar indicates 10% estimated sequence divergence. Numbers on the branches are bootstrap values; only values higher than 50% are given.

75 Chapter 5

The UPGMA dendogram that was constructed from the DGGE profiles obtained from the three large incubators shows the highest similarity between samples derived from the same lab. It is remarkable that the DGGE profiles and the UPGMA dendogram (Fig. 5) show that the community composition and biodiversity of biofilms obtained from one incubator but cultured at different light intensities were more similar than the biofilms grown at the same light conditions but in different laboratories.

Figure 5. UPGMA dendrogram showing the combined clustering analyses of the digitized DGGE profiles (Figures 3A–C) using the unweighted pairwise grouping method with mathematical averages (Jaccard coefficient of similarity). The analysis is based on the presence or absence of bands at certain positions in each lane of each gel.

DGGE and phylogenetic analysis of cyanobacteria and chloroplasts. The cyanobacterial 16S rRNA gene DGGE profile (Fig. 3B) from the Delft biofilms showed high similarities for both growth irradiances. Several bands are present in both profiles. Band number 5 (Fig. 3B), visible in the low light profile, is also faintly visible in the high light profile. Bands 1 and 6 are present under both light conditions, and are both affiliated to Synechocystis sp. (Fig. 6). A band with the same position is present in the inoculum, but its sequence was not obtained. Band couple 2 and 7 as well as couple 3 and 8 are also present under both light conditions and show the same phylogenetic affiliation (Fig. 6). Band number 4 is only present at the high light profile. The high and low light profiles from Yerseke are almost identical. A dominant band at the top of both profiles (Fig. 3B, band 9) is affiliated to the same Scenedesmus chloroplast as band 5 from the Delft low light profile. Another band present at both light conditions in Yerseke shares its position with bands 3 and 8 in the Delft profiles. These bands correspond with sequences affiliated to Phormidium tenue (Figs 3B and 7). Identically positioned bands 11 and 12 are both affiliated to Microcoleus vaginatus. The Magdeburg biofilm profiles show a very thick dominant band at the high light intensity that is very faintly visible in the low light intensity profile (Fig. 3B; band 13), which is affiliated to Planktothrix sp. The thick dominant band that is present in the low light profile

76 On the reproducibility of microcosm experiments

(Fig. 3B; band 14), which was only faintly visible in the high light profile, is affiliated to a Microcoleus vaginatus strain (Fig. 6).

Figure 6. Evolutionary tree showing the phylogenetic affiliations as revealed by comparative analysis of the cyanobacterial 16S rRNA gene sequences derived from the DGGE bands shown in Figure 3B. The sequences obtained in this study are printed bold. Escherichia coli (AJ567606) was used as an out-group sequence, but was pruned from the tree. Accession numbers of sequences are noted behind the taxon names. Scale bar indicates 10% estimated sequence divergence. Numbers on the branches are bootstrap values; only values higher than 50% are given.

DGGE and phylogenetic analysis of eukaryotes. The 18S rRNA gene DGGE profiles (Fig. 3C) show a band that was dominantly present in the biofilms cultivated at both light intensities in the three different labs. This band is faintly visible in the inoculum. The sequences derived from these bands (Fig. 3C; bands 3, 5, 8, 11, 14 and 15) are identical and are closely affiliated to the unicellular alga Chlorella fusca (Fig. 7). The profile derived from the inoculum contains only two dominant bands. Band 1 is affiliated to Scenedesmus communis and band 2 is affiliated to Coelastrella multistriata (Fig. 3C; bands 1 and 2). The profiles from both light intensities from Delft are similar. At both intensities, the biofilms from Delft and Yerseke show bands that are affiliated to the rotifer Brachionus

77 Chapter 5 plicatilis (Fig. 3C; bands 4, 6, and 9, and Fig. 7). A Spaeromonas-like appears to be present only in the high light biofilms (120 mol photons m-2 s-1) from Yerseke and Magdeburg (Fig. 3C; band 7 and band 13, and Fig. 7).

Figure 7. Evolutionary tree showing the phylogenetic affiliations as revealed by comparative analysis of the eukaryotic SS rRNA gene sequences derived from the DGGE bands shown in Figure 3C. The sequences obtained in this study are printed bold. Trypanosoma cruzi (AF245380) was used as an out-group sequence, but was pruned from the tree. Accession numbers of sequences are noted behind the taxon names. Scale bar indicates 10% estimated sequence divergence. Numbers on the branches are bootstrap values; only values higher than 50% are given.

78 On the reproducibility of microcosm experiments

DISCUSSION

In summary, the cultivated biofilms were inhabited by a phylogenetically diverse array of prokaryotes including unicellular and filamentous cyanobacteria, as well as bacteria belonging to the Bacteroidetes (formerly known as Cytophaga–Bacteroides–Flavobacteria group), the Alphaproteobacteria and the Betaproteobacteria. The biofilms also included eukaryotes such as green algae, fungi and protozoa. This shows that phototrophic biofilms, although depending on the primary production of oxygenic phototrophs, can develop as a small ecosystem with many different functional groups of organisms, perhaps reflecting the presence of a variety of ecological niches due to spatial heterogeneity. The community composition of the large microcosm biofilms was in all cases very different from the initial inoculum. This suggests that the conditions within the incubator do not resemble the environmental conditions at the overflow weir of the sedimentation basin from which the inoculum was collected. In addition, the treatment of the inoculum may have had a large effect on the community development in the incubators. Although we see that the growth characteristics and the community composition of the phototrophic biofilms in the large incubators are influenced by the light intensities it is remarkable that the effect of cultivation in different incubators in different laboratories is more prevalent than the effect of light conditions. In general, all DGGE profiles (Figs 3a–c) derived from the same incubator setup shared dominant bands despite their cultivation under different light regimes, while the profiles from the same light regimes were profoundly different for each incubator setup. This is confirmed by the UPGMA dendrogram (Fig. 5) describing the relatedness of the three DGGE profiles. The dendrogram shows a clear separation between the different incubators and no similarity within the corresponding light intensities. These results indicate that, despite efforts to operate the incubators in each laboratory under identical conditions, the species composition of the mature biofilm communities was highly variable. These differences suggest a poor reproducibility in species composition of microcosm experiments with benthic microbial communities, such as phototrophic biofilms. The literature describes experiments with heterotrophic biofilms which exhibited a high degree of reproducibility (Heydorn et al., 2000; Jackson et al., 2001; Lewandowski et al., 2004). However, these studies focused mainly on the structural development of heterotrophic biofilms with a defined number of species. Although the structural development was not compared in detail in this study, we observed that the diversity in community composition is not reflected in the growth curves of the biofilms (Fig. 2). It has been reported that DGGE profiles of replicate pelagic marine mesocosms showed a high resemblance (Schäfer et al., 2001; Lindstrom et al., 2004). Pelagic environments are traditionally conceptualized as chemically and physically more homogeneous, and hence biologically more homogenous, than benthic habitats such as biofilms. The spatial heterogeneity creates a manifold of microenvironments within multi- species biofilms, and it can be anticipated that this will stimulate a high species diversity and structural complexity. This will result in poor reproducibility of model ecosystem experiments. The duplicate cultivation experiments with the flow cell incubators showed high DGGE profile similarity. Although the flow cells are small in size and are completely closed systems, which were sterilized before inoculation (unlike the large microcosms), these results indicate that reproducible cultivation of phototrophic biofilms is possible. It could be argued that small and unidentified differences in the operating conditions of the large incubators induced the divergence in species composition. It must be

79 Chapter 5 noted that extreme care was taken to keep all conditions identical. Since the large incubators are open systems, another explanation could be that the starting community was not very resistant to invaders. Succession is usually defined as an ordered unidirectional process of species replacement leading to a stable climax community. However, it has also been proposed that regular disturbances could prevent communities from reaching a stable climax stage (Massol-Deya et al., 1997). Our biofilm communities diverged to different compositions from identical starting communities and conditions. Therefore, it seems reasonable to hypothesize that a rather unstable community was cultivated (Grimm et al., 1992) or that, for instance, fluctuating nutrient concentrations disturbed the succession towards a stable reproducible climax community. At the moment of sampling, we identified the biofilms as mature. However, the end of exponential growth does not necessarily mean that a stable climax community has established. Therefore, we cannot exclude that the biofilms were still in a transient state, developing slowly towards a final convergence. There is often an implicit, and untested, assumption that when a model ecosystem becomes more complex in terms of species diversity and environmental parameters, then it becomes more difficult to maintain conditions identical and stable between replicate experiments (Kangas and Adey, 1996; Wynn and Paradise, 2001). It has been postulated that chaotic dynamics and other nonlinear phenomena can play a role in community ecology (Allen et al., 1993; Vandermeer et al., 2002). Although empirical evidence of chaos, or complex behaviour, in ecosystems is scarce (Clodong and Blasius, 2004), it is possible that the observed variation results from intrinsic complex and even chaotic behaviour of microbial communities (Becks et al., 2005). Ecological studies have shown that there are trade-offs of microcosm size with predictability and experimental reproducibility (Kangas and Adey, 1996). These scale effects could be relevant for the observed differences in reproducibility between the large incubators and the small flow cell incubators. Our findings demonstrate that secure experimental validation of reproducibility is essential for the use of microcosm systems in microbial ecology studies, especially for conclusions concerning differences in community composition and biodiversity in benthic systems. Future experiments should focus on the relationship between the observed differences in community composition and other biofilm parameters such as chlorophyll-a content per gram dry mass, oxygen profiles, and EPS fractions. In addition, the temporal and spatial aspects of phototrophic biofilm community compositions in response to environmental disturbances remain an interesting subject for further investigations.

ACKNOWLEDGEMENTS

This research was supported by the European Union (PHOBIA project, contract QLK3-CT-2002-01938). We thank Patrizia Albertano and coworkers (University of Rome 'Tor Vergata', Italy) for providing the inoculum. We thank Jan Rijstenbil (NIOO-KNAW, Netherlands Institute of Ecology) for coordination of the PHOBIA project. This is publication no. 3824 of NIOO-KNAW.

80 On the reproducibility of microcosm experiments

REFERENCES

Albertano, P., Congestri, R., and Shubert, L.E. (1999) Cyanobacterial biofilms in sewage treatment plants along the Thyrrenian coast (Mediterranean Sea), Italy. Arch Hydrobiol: Algol Stud 94 (Suppl): 13–24. Allen, J.C., Schaffer, W.M., and Rosko, D. (1993) Chaos reduces species extinction by amplifying local population noise. Nature 364: 229–232. Altschul, S.F., Gish, W., Miller, W., Myers, E.W., and Lipman, D.J. (1990) Basic local alignment search tool. J Mol Biol 215: 403–410. Becks, L., Hilker, F.M., Malchow, H., Jurgens, K., and Arndt, H. (2005) Experimental demonstration of chaos in a microbial food web. Nature 435: 1226–1229. Bender, J., and Phillips, P. (2004) Microbial mats for multiple applications in aquaculture and bioremediation. Bioresour Technol 94: 229–238. Clodong, S., and Blasius, B. (2004) Chaos in a periodically forced chemostat with algal mortality. Proc Biol Sci 271: 1617–1624. Cogan, N.G., and Keener, J.P. (2004) The role of the biofilm matrix in structural development. Math Med Biol 21: 147–166. Des Marais, D.J. (1990) Microbial mats and the early evolution of life. Trends Ecol Evol 5: 140–144. Diez, B., Pedros-Alio, C., Marsh, T.L., and Massana, R. (2001) Application of denaturing gradient gel electrophoresis (DGGE) to study the diversity of marine picoeukaryotic assemblages and comparison of DGGE with other molecular techniques. Appl Environ Microbiol 67: 2942–2951. Griffiths, R.I., Whiteley, A.S., O'Donnell, A.G., and Bailey, M.J. (2000) Rapid method for coextraction of DNA and RNA from natural environments for analysis of ribosomal DNA- and rRNA-based microbial community composition. Appl Environ Microbiol 66: 5488–5491. Grimm, V., Schmidt, E., and Wissel, C. (1992) On the application of stability concepts in ecology. Ecol Model 63: 143–161. Guillard, R.R.L., and Hargraves, P.E. (1993) Stichochrysis immobilis is a diatom, not a chrysophyte. Phycologia 32: 234–236. Heydorn, A., Ersboll, B.K., Hentzer, M., Parsek, M.R., Givskov, M., and Molin, S. (2000) Experimental reproducibility in flow-chamber biofilms. Microbiology 146: 2409–2415. Jackson, G., Beyenal, H., Rees, W.M., and Lewandowski, Z. (2001) Growing reproducible biofilms with respect to structure and viable cell counts. J Microbiol Methods 47: 1–10. Kangas, P., and Adey, W. (1996) Mesocosms and ecological engineering. Ecol Eng 6: 1–5. Lewandowski, Z., Beyenal, H., and Stookey, D. (2004) Reproducibility of biofilm processes and the meaning of steady state in biofilm reactors. Water Sci Technol 49: 359–364. Lindstrom, E.S., Vrede, K., and Leskinen, E. (2004) Response of a member of the Verrucomicrobia, among the dominating bacteria in a hypolimnion, to increased phosphorus availability. J Plankton Res 26: 241–246. Ludwig, W., Strunk, O., Westram, R., et al. (2004) ARB: a software environment for sequence data. Nucleic Acids Res 32: 1363–1371. Massol-Deya, A., Weller, R., Rios-Hernandez, L., Zhou, J.Z., Hickey, R.F., and Tiedje, J.M. (1997) Succession and convergence of biofilm communities in fixed-film reactors treating aromatic hydrocarbons in groundwater. Appl Environ Microbiol 63: 270–276. Nübel, U., Garcia-Pichel, F., and Muyzer, G. (1997) PCR primers to amplify 16S rRNA genes from cyanobacteria. Appl Environ Microbiol 63: 3327–3332. Odum, H.T. (1989) Ecological engineering and self-organization. Ecological Engineering: An Introduction to Ecotechnology (Mitsch WJ and Jørgensen SE, eds), pp. 79–101. John Wiley and Sons. Paerl, H.W., Pinckney, J.L., and Steppe, T.F. (2000) Cyanobacterial-bacterial mat consortia: examining the functional unit of microbial survival and growth in extreme environments. Environ Microbiol 2: 11–26. Palmer, R.J. Jr (1999) Microscopy flowcells: perfusion chambers for real-time study of biofilms. Methods Enzymol 310: 160–166. Prince, R.C., and Kheshgi, H.S. (2005) The photobiological production of hydrogen: potential efficiency and effectiveness as a renewable fuel. Crit Rev Microbiol 31: 19–31.

81 Chapter 5

Schäfer, H., and Muyzer, G. (2001) Denaturing gradient gel electrophoresis in marine microbial ecology. Methods in Microbiology, Marine Microbiology, Vol. 30 (Paul JH, ed), pp. 425–468. Academic Press. 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. Schumacher, G., Blume, T., and Sekoulov, I. (2003) Bacteria reduction and nutrient removal in small wastewater treatment plants by an algal biofilm. Water Sci Technol 47: 195–202. Stanier, R.Y., Kunisawa, R,, Mandel, M., and Cohen-Bazire, G. (1971) Purification and properties of unicellular blue-green algae (order Chroococcales). Bacteriol Rev 35: 171–205. Vandermeer, J., Evans, M.A., Foster, P., Hook, T., Reiskind, M., and Wund, M. (2002) Increased competition may promote species coexistence. Proc Natl Acad Sci USA 99: 8731–8736. Wimpenny, J., Manz, W., and Szewzyk, U. (2000) Heterogeneity in biofilms. FEMS Microbiol Rev 24: 661–671. Wynn, G., and Paradise, C.J. (2001) Effects of microcosm scaling and food resources on growth and survival of larval Culex pipiens. BMC Ecol 1: 3. Zippel, B., and Neu, T.R. (2005) Growth and structure of phototrophic biofilms under controlled light conditions. Water Sci Technol 52: 203–209.

82 6 Development of a PCR for the detection and identification of cyanobacterial nifD genes

Guus Roeselers, Lucas J. Stal, Mark C.M. van Loosdrecht, and Gerard Muyzer

Journal of Microbiological Methods (in press) Chapter 6

ABSTRACT

In this study we have designed degenerate primers after comparative analysis of nifD gene sequences from public databases, and developed a PCR protocol for the amplification of nifD sequences from cyanobacteria. The primers were tested on a variety of nitrogenase-containing and nitrogenase-lacking bacteria. By using this protocol, we amplified nifD sequences from DNA that was isolated from three phototrophic microbial communities. Denaturing gradient gel electrophoresis (DGGE) and clone library analysis of the nifD amplicons showed the presence of distinct groups of diazotrophic cyanobacteria in each of the investigated microbial communities. Phylogenetic trees constructed from the sequences of nifD gene fragments are congruent with those based on ribosomal RNA gene sequences.

INTRODUCTION

Cyanobacteria are oxygenic photoautotrophs that occur in virtually any illuminated environment on Earth and they are globally important primary producers. Cyanobacteria evolved early in Earth’s history and they played a crucial role in the evolution of life through the oxygenation of the atmosphere. Primary production by the cyanobacteria is supported by their capacity of utilizing and assimilating a plethora of different sources of nitrogen such as ammonium, nitrate, nitrite, urea and amino acids, particularly arginine and glutamine (Flores and Herrero, 1994, 2005). Many cyanobacteria can also fix and assimilate atmospheric nitrogen (N2), a property that is confined to a few specialized Bacteria and Archaea that possess the N2-fixing enzyme nitrogenase. Cyanobacteria are the only oxygenic phototrophs that combine the fixation of CO2 and N2. Hence, these organisms do not only play a role in the global carbon cycle but they make also important contributions to the nitrogen cycle. Nitrogenase is a complex consisting of two enzymes. Dinitrogenase is a tetramer with a molybdenum-iron (MoFe) co-factor composed of two identical subunits encoded by nifD (Lammers and Haselkorn, 1984) and the other two identical subunits are encoded by nifK (Mazur and Chui, 1982). Dinitrogenase reductase contains an iron cofactor and is composed of two identical subunits encoded by nifH (Mevarech et al., 1980). There are two alternative nitrogenase systems known that differ in their cofactors. The vanadium nitrogenase contains vanadium in stead of molybdenum as co-factor and is encoded by vnfH/vnfDGK and the iron-only nitrogenase encoded by anfHDGK (Bishop and Joerger, 1990). Alternative nitrogenases are rare among cyanobacteria and only the vanadium nitrogenase has been found (Boison et al., 2006). NifH is one of the oldest existing functional genes known in the history of gene evolution (Raymond et al., 2004) and its sequence is highly conserved. Bacterial phylogeny based on sequence divergences of nifH are generally in agreement with the phylogeny inferred from 16S rRNA gene sequences (Ueda et al., 1995; Zehr et al., 1995; Falcon et al., 2002). Currently, the database for nifH is one of the largest non-ribosomal gene datasets and includes a large number of uncultivated organisms (Zehr et al., 2003). NifD gene sequences could provide additional information for evaluating phylotypes of diazotrophic microorganisms. Compared to nifH there are relatively few nifD sequences available which is a handicap for the phylogenetic analysis. However, nifD genes, when used as phylogenetic markers, promise to provide more resolution among closely-related diazotrophic microorganisms and could better distinguish nif gene family

84 Detection and identification of cyanobacterial nifD genes members as well as the alternative nitrogenases such as the vanadium-containing enzyme (Zehr et al., 2003; Henson et al., 2004a). Here, we demonstrate that nifD gene sequences can be used to detect and identify diazotrophic cyanobacteria in natural communities. PCR products generated using primers homologous to conserved regions in the cyanobacterial nifD genes were subjected to DGGE and clone library analysis in order to determine the genetic diversity of diazotrophic cyanobacteria in environmental samples.

MATERIALS AND METHODS

Pure cultures and environmental samples Samples were collected from a phototrophic biofilm (sample LC4), grown under nitrogen limiting conditions in a special designed incubator (Roeselers et al., 2006). Microbial mat samples were collected from a geothermal hot spring at Kap Tobin on the east coast of Greenland (sample K12; (Kühl et al., 2004; Roeselers et al., 2007) and from the beach on the North Sea island of Schiermonnikoog in the Netherlands (sample GB; (Stal et al., 1985). Various axenic strains of cyanobacteria, Bacteria, Archaea and Eukarya were obtained from culture collections (Table 1). Strains were grown using specific media and growth conditions as recommended by the culture collections from which they were obtained.

DNA extraction Genomic DNA was extracted from pure cultures, microbial mat and, biofilm samples by applying approximately 300 mg biomass to the UltraClean Soil DNA Isolation Kit TM (Mo Bio Laboratories, Carlsbad, CA, USA) according to the manufacturer’s protocol. Complete cell lysis was verified afterwards by using phase-contrast microscopy. The quantity and quality of the extracted DNA was analyzed by spectrophotometry using the NanoDrop ND-1000 TM (NanoDrop Technologies, Wilmington, DE, USA) and by agarose gel electrophoresis. DNA dilutions were stored at –20oC.

Primer design. Amplification primers were designed on the basis of multiple comparisons of available nifD sequences in GenBank. The DNA sequences as well as their protein translations were aligned using ClustalX 1.83 (Chenna et al., 2003). The alignments were used to identify conserved regions specific for cyanobacterial nifD sequences. Degenerate primers flanking a conserved region of the nifD gene were designed using Primer Premier 5 (PremierBiosoft, Palo Alto, CA, USA) The primers nifD552-F (5’ TCCGKGGKGTDTCTCAGTC 3') and nifD861-R (5’ CGRCWGATRTAGTTCAT 3’) target a fragment of approximately 310 bp, from position 552 to 861 in nifD sequence of Anabaena cylindrica PCC 7122 (AF442506; (Henson et al., 2002). Primer specificity with reference to published nifD sequences were checked with the Blastn algorithm for short, nearly exact matches. PCR amplification was tested with various pure cultures of cyanobacteria and other bacteria (Table 1). The reverse primer was equipped with a 40-base-pair GC-rich sequence (Schäfer and Muyzer, 2001) for DGGE analysis. Gradient PCR was used to determine optimum annealing temperatures.

85 Chapter 6

PCR amplification All PCR reactions were conducted in 50 Gl aliquots containing 10-50 ng template DNA, 25 pM of each primer, 10 pM MgCl2, 25 Gl Taq PCR Master Mix (QIAGEN, Hilden, Germany) and water. The PCR reactions were carried out with a denaturation step of 5 minutes at 95oC, followed by 35 cycles of 1 min denaturation at 95oC, 1 min annealing at 52oC, and 1 min extension at 72oC, followed by a final extension step of 15 min at 72oC. All amplification reactions were performed in a T1 Thermocycler (Biometra, Westburg, the Netherlands). Extreme care was taken to prevent any DNA contamination of solutions and plastic disposables used for PCR. All heat sterilized plastic tubes were exposed to UV light for 30 min before use. Only DNA and RNA free water (W4502, Sigma-Aldrich, St. Louis, MO, USA) was used to prepare PCR reagent stock solutions, PCR reaction mixtures, and negative controls.

Perpendicular DGGE analysis of a nifD fragment The melting behavior was experimentally established by perpendicular DGGE analysis (Myers et al., 1987; Muyzer et al., 1993). Figure 1A shows the melting curve of the nifD PCR product from Nodularia spumigena CCY 9414. From this perpendicular DGGE analysis, an optimal gradient of denaturants was defined, i.e., 25 to 55%, to obtain the best resolution in the separation of different nifD sequences in parallel denaturing gradient gels.

Determination of the optimal electrophoresis time. To determine the length of electrophoresis time, a so-called ‘‘time travel’’ experiment was performed (Muyzer et al., 1993). NifD fragments with a GC clamp of Lyngbya aestuarii CCY 9616 (EF186051) and Nodularia spumigena CCY 9414 (EF186055) were loaded at 30-min intervals for up to 6 h onto a 6% polyacrylamide gel containing a 25 to 55% denaturing gradient (Fig. 1B).

Parallel DGGE analysis From each PCR reaction, 20 Gl containing approximately 1 mg DNA was mixed with 6 Gl of 10x gel loading solution and loaded onto a 6% polyacrylamide gel containing a 25 to 55% denaturing gradient. Gels were run in 1x TAE (Tris-acetate-EDTA buffer, 50X stock solution: 242g of Tris base, 57.1ml of glacial acetic acid, 100ml of 0.5M EDTA pH 8.0) for 5 h at 100 V, at 60°C. Gels were stained with ethidium bromide and analyzed and photographed using the GelDoc UV Transilluminator (Bio-Rad, Hercules, CA, USA). Dominant bands were excised from the DGGE gel with the environmental samples (Fig. 2). Each small gel slice was placed in 15 Gl of sterile water for 24 h at 4°C. Subsequently, 2 Gl of the solution was used as template DNA for re-amplification using the PCR protocol as described above. The PCR products were again subjected to DGGE analysis to confirm the purity and their position relative to the bands from which they were originally excised. PCR products were purified using the QIAquick Gel Extraction Kit (QIAGEN, Hilden, Germany). Purified PCR products were sequenced by a commercial company (BaseClear, Leiden, the Netherlands). The sequencing reactions were carried out with the nifD552-F primer.

Construction of clone libraries Clone libraries representing the diversity of nifD containing cyanobacteria in microbial mat samples from Greenland (sample K12) and Schiermonnikoog (GB), and in the cultivated phototrophic biofilm (LC4) were constructed as follows. NifD PCR products were cloned using the TOPO TA Cloning Kit (Invitrogen, Carlsbad, California, USA). The

86 Detection and identification of cyanobacterial nifD genes manufacturer's instructions were followed for ligation and the transformation of Escherichia coli TOP10 cells (Invitrogen) by heat-shock treatment at 42°C. 24 h of incubation at 37°C on Luria-Bertani agar plates containing 50 Ig ml–1 ampicillin and 40 mg ml–1 X-Gal (5-bromo- 4-chloro-3-indolyl-ß-D-galactopyranoside) was used to select transformants. At least 30 transformant colonies from both microbial mat samples and the biofilm sample were selected for direct colony PCR with the universal primers M13 Forward -20 (5’ GTAAAACGACGGCCAG 3’) and M13 Reverse (5’ CAGGAAACAGCTATGAC 3’) targeting vector regions flanking the nifD insert. Subsequently, the PCR products were sequenced with the M13-F (-20) primer by a commercial company (Macrogen, Seoul, Korea)

Phylogenetic analysis NifD sequences obtained in this study and sequences obtained from GenBank were aligned using ClustalX 1.83 (Chenna et al., 2003). A neighbor-joining (NJ) tree was constructed in PAUP* 4.0b10 (Sinauer Associates, Sunderland, MA, USA), using the Jukes- Cantor model. The NJ calculation was subjected to bootstrap analysis (1000 replicates). Trees were rooted with outgroup analysis using the nifD sequence of Azotobacter vinelandii (X06886). Sequences determined in this study have been deposited in the GenBank database under accession numbers EF186025 to EF186056.

RESULTS

PCR of nifD fragments from pure cultures and environmental samples Primers nifD552-F and nifD861-R were successfully used to specifically amplify partial nifD sequences from genomic DNA that was extracted from 15 axenic cultures of diazotrophic cyanobacteria belonging to the different subsections as difined by Rippka et al., (1979). In order to determine their specificity, the primers were also tested against various other nitrogenase-containing and nitrogenase-lacking microorganisms. A weak product was obtained with genomic DNA extracted from Methanosarcina barkeri and Vibrio diazotrophicus. However, these products had molecular weights between 800 and 1000 bp, instead of the expected 310 bp. Sequence analyses of these non-specific products showed no similarity to any nifD sequence. The results of this specificity experiment are summarized in Table 1. NifD amplicons were obtained for all three environmental samples: LC4 (a cultivated phototrophic biofilm), K12 (a microbial mat from a thermal spring at the east coast of Greenland), and GB (a microbial mat from a sandy beach on the Dutch Wadden Sea island of Schiermonnikoog).

DGGE analysis of environmental samples Perpendicular DGGE analysis and a so-called time travel experiment revealed the optimal denaturing gradient of 25 to 55%, and an optimal electrophoresis time of 5 h at 100 V (Fig.1). These conditions were used for DGGE analyses of nifD fragments amplified from the environmental samples (LC4, K12, and GB). Four bands were excised from the cultivated biofilm sample while 2 bands were obtained for each of the microbial mats (Fig. 2). NifD fragments from the excised bands were re-amplified and subsequently sequenced. Comparative sequence analysis (Fig. 3) showed that all bands represented cyanobacterial nifD sequences, except for band 8 (sample GB) that showed 86% similarity to a diazotrophic b-proteobacterium belonging to the genus Herbaspirillum (You et al., 2005). The DGGE pattern obtained from sample K12 showed two separate bands but the sequences that were

87 Chapter 6 distantly affiliated to nifD of Chlorogloeopsis sp., were identical indicating a DGGE artifact (Janse et al., 2004). The cultivated phototropic biofilm (LC4) showed two bands affiliated to nifD of Cylindrospermum species (bands 3 and 5), one band affiliated to Calothrix sp. and one band showing 95% similarity to Anabaena sp. CA. In addition to the Herbaspirillum like nifD sequence, sample GB revealed another band that clustered with nifD of Anabaena sp. CA (Fig. 3).

Table. Specificity of the cyanobacterial nifD PCR

a b c Organism Source N2 fixing PCR with nifD primers Cyanobacteria Anabaena variabilis, ATCC 29413 CCY + + Anabaena planctonica, CCY 0369 CCY + + Calothrix desertica, SAG 35.79 SAG + + Calothrix sp., ATCC 27905 CCY + + Chlorogloeopsis fritschii, ATCC 27193 CCY + + Cyanothece sp., CCY 0110 CCY + + Gloeothece sp., PCC 6909 CCY + + Fischerella muscicola, SAG 1427-1 CCY + + Lyngbya aestuarii, CCY 9616 CCY + + Mastigocladus laminosus, SAG 4.84 SAG + + Myxosarcina sp., CCY 0025 CCY + + Nodularia spumigena, CCY 9414 CCY + + Nostoc insulare, SAG 54.79 SAG + + Nostoc muscorum, PCC 7120 CCY + + Oscillatoria limnetica, CCY 9509 CCY + + Planktothrix agardhi, SAG 6.89 SAG - - Synechococcus sp. Miami BG043511 CCY + + Synechocystis sp., PCC 6803 ATCC - - Other bacteria Acetobacter diazotrophicus, DSM 56 DSMZ + - Azotobacter vinelandii, DSM 85 DSMZ + - Bacillus azotofixans, DSM 1735 DSMZ + - Burkholderia tropica, DSM 15359 DSMZ + - Clostridium pasteurianum, DSM 525 DSMZ + - Escherichia coli, ATCC 11775 ATCC - - Frankia sp., DSM 44263 DSMZ + - Helicobacter pilori ATCC 26695 ATCC - - Pseudomonas fluorescens, NCCB 2037 NCCB - - Rhizobium leguminosarum, DSM 30132 DSMZ + - Vibrio diazotrophicus, DSM 2604 DSMZ + -/+ Archaea Methanosarcina barkeri, DSM 3647 DSMZ + -/+ Eukarya Scenedesmus obliquus SAG 276-12 SAG - - aCulture collections: ATCC; American Type Culture Collection, Rockville, Maryland. CCY; Culture Collection Yerseke, NIOO Centre for Estuarine and Marine Ecology, Yerseke, The Netherlands. DSMZ; Deutsche Sammlung von Microorganismen und Zellkulturen, Braunsweigh, Germany. NCCB: The Netherlands Culture Collection of Bacteria, Utrecht University, Utrecht, The Netherlands. SAG: Sammlung von Algenkulturen Georg-August-Universität, Göttingen, Germany. bOrganisms capable of fixing atmospheric nitrogen. cAmplification products yielded by applying primers nifD1371F and nifD1683R. [- ) no product, +) product of 312 kb, -/+) non-specific product of high molecular weight].

88 Detection and identification of cyanobacterial nifD genes

Figure 1. (A) Perpendicular DGGE separation pattern of PCR-amplified nifD gene fragments from Nodularia spumigena CCY 9414. From this perpendicular DGGE analysis, we defined an optimal gradient of 25 to 55% denaturants. (B) Time-travel DGGE with nifD fragments of Lyngbya aestuari CCY 9616 (band 1) and Nodularia spumigena CCY 9414 (band 2). Between 180 and 210 min of electrophoresis at 100 V and 60oC, the DNA molecules denatured and slowed down drastically.

89 Chapter 6

Figure 2. DGGE patterns of cyanobacterial nifD gene fragments obtained after PCR amplification of DNA samples from a microbial mat sample from Greenland (K12), a cultivated phototrophic biofilm (LC4), and a microbial mat sample from Schiermonnikoog (GB). Numbers at the left of each lane correspond to bands that were excised, PCR-amplified and sequenced.

NifD clone libraries. Sequence analysis of 30 randomly selected nifD clones obtained from the phototrophic biofilm and the microbial mat samples revealed a diverse assemblage of nifD gene sequences (Tabel 2). Each library was clearly dominated by one strain. The K12 library contained 16 clones related to Fischerella sp., which were similar to the sequence, derived from DGGE band 1 and 2. The LC4 library contained 21 clones related to Calothrix sp. PCC 7507, identical to DGGE band 4, and the GB library contained 21 clones related to Cylindrospermum stagnale, similar to DGGE band 7 (Fig. 2). All other sequences obtained from excised DGGE bands were also present in the clone libraries. Most sequences were affiliated to cyanobacterial nifD but some clones obtained from samples K12, LC4 and GB showed a low sequence similarity to proteobacterial nifD sequences (Fig. 3 and Table 2, clones K12-B10, LC4-H1, GB-E1, and GB-C11).

90 Detection and identification of cyanobacterial nifD genes

Table 2. Phylogenetic summary based on clone library analysis of partial cyanobacterial nifD sequences.

Representative Accession No. of Closest relative in Genbank Identity Putative taxon clone no. clones (Blastn) (%) K12-A4 EF186037 16 Fischerella sp. UTEX 1903 85 Stigonematales K12-B2 EF186038 1 Nostoc sp. PCC 7423 95 Nostocales K12-B9 EF186039 5 Synechococcus RF-1 90 Chroococcales K12-B10 EF186040 6 Bradyrhizobium sp. Lpsp.1b 89 Proteobacteria K12-C4 EF186042 2 Synechococcus RF-1 86 Chroococcales LC4-G1 EF186041 21 Calothrix sp. PCC 7101 88 Nostocales LC4-H1 EF186043 3 Roseomonas gilardii 91 Proteobacteria Anabaena variabilis ATCC LC4-H2 EF186044 4 95 Nostocales 29413 LC4-H5 EF186045 2 Nostoc sp. PCC 7120 95 Nostocales GB-C11 EF186032 3 Herbaspirillum sp. B501 84 Proteobacteria GB-D5 EF186033 3 Calothrix sp. PCC 7507 85 Nostocales GB-E1 EF186034 1 Herbaspirillum sp. B501 82 Proteobacteria Cylindrospermum stagnale GB-E5 EF186035 21 95 Nostocales PCC 7417 Cylindrospermum stagnale GB-E10 EF186036 2 96 Nostocales PCC 7417

Phylogenetic analysis of partial nifD sequences The topology of the cyanobacterial nifD tree is largely consistent with the tree based on 1500 bp long nifD sequences shown by Henson et al. (Henson et al., 2004b). Some sequences cluster closely together with high bootstrap values, whereas others are relatively deep branching, showing variation in the evolutionary distance among cyanobacterial groups. NifD sequences obtained from heterocystous cyanobacteria appeared in a monophyletic cluster with non-heterocystous cyanobacteria as sister groups. The heterocystous genera Anabaena, Nostoc, Chlorogloeopsis, Scytonema and Fischerella form a close cluster, which was also shown by phylogenetic analysis of 16S rRNA gene (Wilmotte, 1994) and nifH sequences (Zehr et al., 1997) The sequence from clone LC4-H1, with 91% sequence similarity to a Roseomonas gilardii strain (Table 2) seems positioned within the cyanobacterial tree, clustering together with clone K12-B2, which showed 95% sequence similarity to Nostoc sp. PCC 7423.

91 Chapter 6

Figure 3. Neighbor-joining tree showing the phylogenetic affiliations of cyanobacterial nifD sequences. Sequences obtained from the present study are in bold. The nifD sequence of Azotobacter vinelandii (X06886) was used as an out group. Scale bar indicates 5% sequence divergence. Bootstrap values (50%) for 1000 replicates are placed at the nodes of the branches.

92 Detection and identification of cyanobacterial nifD genes

DISCUSSION

In this study we describe the development of a PCR specific for nifD to determine the diversity of diazotrophic cyanobacteria in various environments. PCR primers specifically targeting cyanobacterial 16S rRNA gene sequences (Nübel et al., 1997) have proven to be useful for the study of cyanobacterial diversity in general. However, it is usually not possible to specifically detect diazotrophic cyanobacteria with a marker based on 16S rRNA genes, although an exception to this rule may be found among unicellular N2-fixing cyanobacteria (Mazard et al., 2004). A number of probes and primers have been described that successfully target genes encoding subunits of nitrogenase (Ueda et al., 1995; Zehr et al., 1995) but most nitrogenase diversity studies are based on the phylogenetic analysis of nifH. Although cyanobacterial nifH genes cluster closely together (Zehr et al., 1997) it has not been possible to design PCR primers that specifically target cyanobacterial nifH genes. The purpose of this study was to develop a molecular tool for the detection of diazotrophic cyanobacteria in environmental samples. Comparative sequence analysis of cyanobacterial nifD present in Genbank revealed conserved regions in the sequences that allowed the development of PCR primers specific for cyanobacterial nifD genes. The DGGE and clone library analyses of samples K12, LC4 and GB demonstrated that this PCR approach was successful for the determination and identification of diazotrophic cyanobacteria in complex microbial communities. Notwithstanding the large number of degenerate bases in the primers we designed, rigorous testing showed a clear specificity for cyanobacteria even if DGGE and clone library analysis revealed the amplification of a few proteobacterial sequences. We think that the primers described here could be particularly useful for the study of cyanobacterial mats and biofilms. Microbial mats may contain a variety of different cyanobacterial taxa, including heterocystous, filamentous non-heterocystous, and unicellular cyanobacteria, and exhibit often high rates of N2 fixation (Paerl et al., 1996). Although anoxygenic phototrophic or chemotrophic bacteria present in these microbial mats might also contribute to N2 fixation, experiments with the inhibitor of photosystem II, DCMU, argue against this (Bebout et al., 1993). Also Steunou et al. (2006) demonstrated that nitrogenase activity in hot spring microbial mats was attributed to the hemophilic unicellular Synechococcus inhabiting this environment. Our nifD PCR approach in combination with reverse transcriptase and quantitative PCR could be used to specifically identify and quantify spatial and temporal variation in cyanobacterial nitrogenase transcription in microbial communities. The primers designed for this study are based on a limited number of sequences. It can therefore not be excluded that newly sequenced cyanobacterial nifD genes do not contain the sequences that are targeted by the primers described here. It is also possible that more nifD sequences not belonging to cyanobacteria are amplified by these primers. However, based on the information we have obtained during this study we feel confident that we have developed a robust method. Still, the phylogenetic analysis of nifD is hampered by the limited number of sequences available to date. This is illustrated by the observation that the closest relative in Genbank of clone LC4-H1 was a Roseomonas strain while phylogenetic analysis revealed similarity with cyanobacterial nifD sequences. However, we anticipate that the databases of nifD sequences will expand rapidly, allowing increasingly robust phylogenetic analyses. This study has already increased the number of cyanobacterial nifD sequences in Genbank substantially.

93 Chapter 6

In conclusion, our approach will prove to be useful to investigate the phylogenetic affiliation, activity and, ecological significance of diazotrophic cyanobacteria in complex communities. The analysis of nifD gene sequences will improve the robustness of phylogenetic reconstructions based on 16S rRNA and other functional genes sequences.

ACKNOWLEDGEMENTS

This research was supported by the European Union (PHOBIA project, contract QLK3-CT- 2002-01938). We thank Dr. Michael Kühl for providing the microbial mat sample from Kap Tobin.

REFERENCES

Bebout, B.M., Fitzpatrick, M.W., and Paerl, H.W. (1993) Identification of the sources of energy for nitrogen fixation and physiological characterization of nitrogen-fixing members of a marine microbial mat community. Appl Environ Microbiol 59: 1495-1503. Bishop, P.E., and Joerger, R.D. (1990) Genetics and molecular-biology of alternative nitrogen-fixation systems. 1040-2519 41: 109-125. Boison, G., Steingen, C., Stal, L.J., and Bothe, H. (2006) The rice field cyanobacteria Anabaena azotica and Anabaena sp. CH1 express vanadium-dependent nitrogenase. Arch Microbiol 186: 367- 376. Chenna, R., Sugawara, H., Koike, T., Lopez, R., Gibson, T.J., Higgins, D.G., and Thompson, J.D. (2003) Multiple sequence alignment with the Clustal series of programs. Nucleic Acids Res 31: 3497- 3500. Falcon, L.I., Cipriano, F., Chistoserdov, A.Y., and Carpenter, E.J. (2002) Diversity of diazotrophic unicellular cyanobacteria in the tropical North Atlantic Ocean. Appl Environ Microbiol 68: 5760-5764. Flores, E., and Herrero, A. (1994) Assimilatory nitrogen metabolism and its regulation. In The Molecular Biology of Cyanobacteria. Bryant, D.A. (ed): Kluwer Scientific Publications, Dordrecht, The Netherlands, pp. 487–517. Flores, E., and Herrero, A. (2005) Nitrogen assimilation and nitrogen control in cyanobacteria. Biochem Soc Trans 33: 164-167. Henson, B.J., Watson, L.E., and Barnum, S.R. (2002) Molecular differentiation of the heterocystous cyanobacteria, Nostoc and Anabaena, based on complete nifD sequences. Curr Microbiol 45: 161-164. Henson, B.J., Watson, L.E., and Barnum, S.R. (2004a) The evolutionary history of nitrogen fixation, as assessed by nifD. J Mol Evol 58: 390-399. Henson, B.J., Hesselbrock, S.M., Watson, L.E., and Barnum, S.R. (2004b) Molecular phylogeny of the heterocystous cyanobacteria (subsections IV and V) based on nifD. Int J Syst Evol Microbiol 54: 493- 497. Janse, I., Bok, J., and Zwart, G. (2004) A simple remedy against artifactual double bands in denaturing gradient gel electrophoresis. J Microbiol Methods 57: 279-281. Kühl, M., Glud, R.N., Rysgaard, S., and Pedersen, H. (2004) Microbiological studies in hot springs at Daneborg and Scoresbysund. In Zackenberg Ecological Research Operations, 9th Annual Report. Rasch, M., and Caning, K. (eds). Copenhagen: Danish Polar Center, pp. 72-74. Lammers, P.J., and Haselkorn, R. (1984) Sequence of the nifD geneogy coding for the alpha subunit of dinitrogenase from the cyanobacterium Anabaena. Cell 44: 905-911. Mazard, S.L., Fuller, N.J., Orcutt, K.M., Bridle, O., and Scanlan, D.J. (2004) PCR analysis of the distribution of unicellular cyanobacterial diazotrophs in the Arabian Sea. Appl Environ Microbiol 70: 7355-7364. Mazur, B.J., and Chui, C.F. (1982) Sequence of the gene coding for the beta-subunit of dinitrogenase from the blue-green alga Anabaena. Proc Natl Acad Sci U S A 79: 6782-6786. Mevarech, M., Rice, D., and Haselkorn, R. (1980) Nucleotide sequence of a cyanobacterial nifH gene coding for nitrogenase reductase. Proc Natl Acad Sci U S A 77: 6476-6480.

94 Detection and identification of cyanobacterial nifD genes

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. Myers, R.M., Maniatis, T., and Lerman, L.S. (1987) Detection and localization of single base changes by denaturing gradient gel electrophoresis. Methods Enzymol 155: 501-527. Nübel, U., Garcia-Pichel, F., and Muyzer, G. (1997) PCR primers to amplify 16S rRNA genes from cyanobacteria. Appl Environ Microbiol 63: 3327-3332. Paerl, H.W., Fitzpatrick, M., and Bebout, B.M. (1996) Seasonal nitrogen fixation dynamics in a marine microbial mat: Potential roles of cyanobacteria and microheterotrophs. Limnol Oceanogr 41: 419-427. Raymond, J., Siefert, J.L., Staples, C.R., and Blankenship, R.E. (2004) The natural history of nitrogen fixation. Mol Biol Evol 21: 541-554. Rippka, R., Deruelles, J., Waterbury, J.B., Herdman, M., and Stanier, R.Y. (1979) Generic assignments, strain histories and properties of pure cultures of cyanobacteria. Journal of General Microbiology 111: 1-61. Roeselers, G., Zippel, B., Staal, M., van Loosdrecht, M.C.M., and Muyzer, G. (2006) On the reproducibility of microcosm experiments - different community composition in parallel phototrophic biofilm microcosms. FEMS Microbi Ecol 58: 169-178. Roeselers, G., Norris, T., Castenholz, R., Rysgaard, S., Glud, R., Kühl, M., and Muyzer, G. (2007) Diversity of phototrophic bacteria in microbial mats from Arctic hot springs (Greenland). Environ Microbiol 9: 26-38. Schäfer, H., and Muyzer, G. (2001) Denaturing gradient gel electrophoresis in marine microbial ecology. In Methods in Microbiology, Marine Microbiology. Paul, J.H. (ed). New York: Academic Press, pp. 425-468. Stal, L.J., van Gemerden, H., and Krumbein, W.E. (1985) Structure and development of a benthic marine microbial mat. FEMS Microbiol Lett 31: 111–125. Steunou, A.S., Bhaya, D., Bateson, M.M., Melendrez, M.C., Ward, D.M., Brecht, E. et al. (2006) In situ analysis of nitrogen fixation and metabolic switching in unicellular thermophilic cyanobacteria inhabiting hot spring microbial mats. Proc Natl Acad Sci U S A 103: 2398-2403. Ueda, T., Suga, Y., Yahiro, N., and Matsuguchi, T. (1995) Remarkable N2-fixing bacterial diversity detected in rice roots by molecular evolutionary analysis of nifH gene sequences. J Bacteriol 177: 1414- 1417. Wilmotte, A. (1994) Molecular evolution and taxonomy of the cyanobacteria. In The Molecular Biology of the Cyanobacteria. Bryant, D.A. (ed). Dordrecht, Netherlands: Kluwer, pp. 1-25. You, M., Nishiguchi, T., Saito, A., Isawa, T., Mitsui, H., and Minamisawa, K. (2005) Expression of the nifH gene of a Herbaspirillum endophyte in wild rice species: daily rhythm during the light-dark cycle. Appl Environ Microbiol 71: 8183-8190. Zehr, J.P., Mellon, M.T., and Hiorns, W.D. (1997) Phylogeny of cyanobacterial nifH genes: evolutionary implications and potential applications to natural assemblages. Microbiology 143 ( Pt 4): 1443-1450. Zehr, J.P., Jenkins, B.D., Short, S.M., and Steward, G.F. (2003) Nitrogenase gene diversity and microbial community structure: a cross-system comparison. Environ Microbiol 5: 539-554. Zehr, J.P., Mellon, M., Braun, S., Litaker, W., Steppe, T., and Paerl, H.W. (1995) Diversity of heterotrophic nitrogen fixation genes in a marine cyanobacterial mat. Appl Environ Microbiol 61: 2527- 2532.

95

7 Diversity and expression of cyanobacterial hupS genes in pure cultures and in a nitrogen limited phototrophic biofilm

Guus Roeselers, Eline H. Huisjes, Mark C. M. van Loosdrecht, and Gerard Muyzer

Manuscript in preparation Chapter 7

ABSTRACT

We developed PCR primers for conserved regions within the cyanobacterial small subunit uptake hydrogenase (hupS) gene family. These primers were used to PCR amplify partial hupS sequences from 15 cyanobacterial strains. HupS clone libraries were constructed from PCR-amplified gDNA and reverse-transcribed mRNA extracted from phototrophic biofilms cultivated under nitrate limiting conditions. Partial hupS gene sequences derived from cyanobacteria, some of which were not previously known to contain hup genes, were used for phylogenetic analysis. Phylogenetic trees constructed with hupS genes are congruent with those based on 16S rRNA genes, indicating that hupS sequences can be used to identify cyanobacteria expressing hup. Sequences from heterocystous and non-heterocystous cyanobacteria form two separate clusters. Analysis of clone library data showed a discrepancy between the presence and the activity of cyanobacterial hupS genes in phototrophic biofilms. Results showed that the hupS gene can be used to characterize the diversity of natural populations of diazotrophic cyanobacteria, and to characterize gene expression patterns of individual species or strains.

INTRODUCTION

Diazotrophic cyanobacteria can reduce atmospheric dinitrogen gas (N2) into ammonia (NH3), a form in which the nitrogen is available for biological reactions. Nitrogen fixation requires the activity of nitrogenase (nif), an oxygen sensitive multiprotein complex.

Most nitrogenases are catalysts for hydrogen production as they liberate H2 during the reduction of nitrogen to ammonia (Mancinelli et al., 1996; Schutz et al., 2004; Tamagnini et al., 2002). In cyanobacteria, a NiFe-uptake hydrogenase recycles the produced H2 by the nitrogenase machinery. The recycling of hydrogen has been suggested to have three physiological functions. I) It generates ATP via the oxyhydrogen (Knallgas) reaction, II) it protects the sensitive nif complex from oxidative inactivation, and III) it supplies reducing equivalents (electrons) for N2 reduction and other cell functions (Bothe et al., 1977). The uptake hydrogenase complex consists of two subunits: a large subunit encoded by hupL and a small subunit encoded by hupS. Uptake hydrogenases are located in the cytoplasmic face of the cell membrane or thylakoid membrane (Vignalis et al., 2004), where they utilize hydrogen produced by the nitrogenase. Cyanobacterial species or genetically modified strains with a reduced uptake hydrogenase activity could possibly be used for photobiological H2 production (Happe et al., 2000; Kumazawa and Mitsui 1985; Tsyngankov et al., 1999). Molecular approaches provide a way to identify microorganisms that have the potential to produce hydrogen and that are expressing the characteristic hup machinery. Previously it was shown that there is a high degree of sequence similarity among cyanobacterial hup genes (Tamagnini et al., 2005). In this study, we developed PCR primers targeting conserved regions within the cyanobacterial hupS gene family. We analyzed hupS diversity and transcription in cultivated phototrophic biofilms by the direct retrieval and analysis of mRNA that was reverse transcribed, amplified with hupS specific primers, and cloned. The objective of this study was to design PCR primers to amplify cyanobacterial hupS gene fragments, in order to develop a molecular approach for characterizing cyanobacterial populations and hupS gene expression in environmental samples.

98 Cyanobacterial uptake hydrogenases

Insight in the natural molecular variation of hup genes in cyanobacteria could be used to identify a genetic background and a set of growth conditions that would be amenable to the development of photobiological H2 production (Hansel and Lindblad 1998).

MATERIALS AND METHODS

Phototrophic biofilm incubator Samples were collected from a phototrophic biofilm, cultivated under nitrogen limiting conditions in a specially designed incubator (Roeselers et al., 2007). The flow-lane incubator system contained four separate flow channels (LC1, LC2, LC3 and LC4) through which a volume of 4 l mineral medium circulated over a surface covered with 47 polycarbonate slides (76 x 25 x 1 mm). Polycarbonate slides were used as a substratum for biofilm adhesion. Each light chamber contained an adjustable light source and the circulation speed of the culture medium could be regulated precisely. Biofilm growth was monitored and recorded with three light sensors, positioned directly under selected polycarbonate slides. Each light sensor contained three independent photodiodes. Light absorbance was used as an indicator for biomass accumulation. The incubator design was described in more detail by Zippel and Neu (2005).

Growth conditions The mineral medium used was a modification of BG11 (Stanier et al., 1971). The medium contained no nitrogen source. Ammonium ferric citrate green was replaced by -1 -1 - FeCl3. The vitamins cyanocobalamin (40 Ag l ), thiamine HCL (40 Ag l ) and biotin (40 Ag l 1 ) were added. NaNO3 (0.5 mM) was added to the medium supplied to flow channels LC1 and LC 3. The medium was refreshed twice a week. Inoculation biomass was collected from PVC poles that were placed vertically in a helophyte pond system at the municipal wastewater treatment facility in St. Maartensdijk (Tholen, The Netherlands). In order to reduce predation pressure, the inoculum was frozen at -20 oC to kill protozoa, metazoa and nematodes. The inoculum was thawed, sieved through a 3.00 Am mesh, diluted in BG11 medium and added to the four flow channels. Biofilms were cultivated at 25 oC and at a medium flow rate of 50 l h -1. The biofilms were grown at a light intensity of 60 Amol photons m -2 s -1. A light cycle of 16 h light and 8 h dark was applied. Biofilms samples were collected for DNA and RNA extraction after 31 days of cultivation. Axenic cultures of Anabaena variabilis strain SAG 1403-4B were grown in Erlenmeyer flasks with either modified BG11 medium without combined nitrogen or -2 -1 supplemented with 1 mM NaNO3, under a light intensity of 120 Amol m s at 25°C. Various other axenic strains of cyanobacteria, Bacteria, Archaea and Eukarya were obtained from culture collections (Table 1). Strains were grown using specific media and growth conditions as recommended by the culture collections from which they were obtained.

Primer design Amplification primers were designed on the basis of comparisons of available hupS sequences in GenBank at the National Center for Biotechnology Information (NCBI). The DNA sequences as well as their protein translations were aligned using ClustalX 1.83 (Chenna et al., 2003). The alignments were used to identify regions of high conservation among the cyanobacterial sequences. Degenerate primers flanking a conserved region of the hupS gene were designed using Primer Premier 5 (PremierBiosoft, Palo Alto, CA, USA).

99 Chapter 7

The primers hupS2345-F (5’ TAA CGT MCT MTG GCT RCA A 3’) and hupS2605-R (5’ TAR CCA DTC TTT CAT BGG 3’) target a fragment of approximately 286 bp, at positions 2345 and 2630 bp, respectively, of the Anabaena variabilis ATCC 29413 hupSL sequence (Y13216) (Happe et al., 2000). Primer specificity with reference to published hupS sequences were checked with the Blastn algorithm for short, nearly exact matches. PCR amplification was tested with various pure cultures of cyanobacteria.

Biomass collection and DNA extraction Phototrophic biofilm biomass was scraped from the incubated polycarbonate slides with a sterile razor blade. Genomic DNA (gDNA) was extracted from pure cultures and biofilm samples by applying approximately 300 mg biomass to the UltraClean Soil DNA Isolation Kit (Mo Bio Laboratories, Carlsbad, CA, USA) according to the manufacturer’s protocol. The quantity and quality of the extracted DNA was analyzed by spectrophotometry using the NanoDrop ND-1000 (NanoDrop Technologies, Wilmington, DE, USA) and by agarose gel electrophoresis. Extracted DNA was stored at –20oC.

PCR amplification of hupS gene fragments HupS-PCR reactions were conducted in 50 El aliquots containing 50-100 ng template DNA, 25 pM of each primer, 5 pM MgCl2, 25 El Taq PCR Master Mix (QIAGEN, Hilden, Germany) and deionized water (W4502, Sigma-Aldrich, St. Louis, MO, USA). The PCR reactions were carried out in a T1 Thermocycler (Biometra, Westburg, the Netherlands) using the following cycling parameters: a denaturation step of 5 minutes at 94oC, followed by 35 cycles of 1 min denaturation at 94oC, 1 min annealing at 49.6oC, and 1 min extension at 72oC, followed by a final extension step of 7 min at 72oC. All amplification products were loaded together with a molecular weight marker (SmartLadder, Eurogentec, Maastricht, The Netherlands) onto 1% agarose gels.

PCR-DGGE of oxygenic phototrophic 16S rRNA genes To amplify the 16S rRNA encoding gene fragments of cyanobacteria and chloroplasts, the extracted DNA was used as template DNA in PCR reactions using the forward primer 359F-GC and an equimolar mixture of the reverse primers 781R(a) and 781R(b). PCR amplification and denaturing gradient gel electrophoresis (DGGE) were performed as described by Nübel et al. (1997).

RNA extraction RNA was extracted from approximately 100 mg wet pure culture biomass and 200 mg wet biofilm biomass using the RNeasy Mini KitTM (Qiagen, Inc., Hilden, Germany) according to the Animal Protocol as described by the manufacturer. Dnase treatment was performed by using the TURBO DNA-freeTM kit (Ambion, Foster City, CA, USA). Subsequently quantity and quality of the RNA were measured using the NanoDrop. ND- 1000. Copy DNA (cDNA) was produced by reverse transcriptase PCR (RT-PCR), performed with the iScript cDNA Synthesis Kit (Bio-Rad Laboratories, Hercules, CA, USA), according to the manufacturers protocol.

Construction of clone libraries Clone libraries representing the diversity and transcription of cyanobacterial hupS genes in the cultivated phototrophic biofilms were constructed as follows. HupS PCR products amplified from gDNA and cDNA were cloned using the TOPO TA Cloning Kit (Invitrogen, Carlsbad, California, USA). The manufacturer's instructions were followed for

100 Cyanobacterial uptake hydrogenases

ligation and the transformation of Escherichia coli TOP10 cells (Invitrogen) by heat-shock treatment at 42oC. Incubation for 20 h at 37°C on Luria-Bertani agar plates containing 50 Dg ml–1 ampicillin and 40 mg ml–1 X-Gal (5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside) was used to select transformants (blue white screening). At least 30 transformant colonies from gDNA and cDNA amplicons were selected for direct colony PCR with the universal primers M13 Forward -20 (5’ GTAAAACGACGGCCAG 3’) and M13 Reverse (5’ CAGGAAACAGCTATGAC 3’) targeting vector regions flanking the hupS insert. Subsequently, the PCR products were sequenced with the M13-F (-20) primer by a commercial company (Macrogen, Gasan-dong, South Korea).

Phylogenetic analysis HupS sequences obtained in this study and sequences obtained from GenBank were aligned using ClustalX 1.83. Phylogenetic trees were inferred and drawn with PAUP* 4.0b10 software (Sinauer Associates, Sunderland, MA, USA) using the Kimura-2 parameter model. The Neighbour Joining (NJ) calculation was subjected to bootstrap analysis (1000 replicates) using the algorithm available in PAUP*. Sequences determined in this study have been deposited in the GenBank database under accession numbers EF431930 to EF431947 and EF622220 to EF622223.

RESULTS

Biofilm cultivation Biofilm growth was monitored by the decrease of light transmission through the biofilm. The biofilms grown with 0.5 mM NaNO3 (LC1 and LC3) reached their exponential growth stage around 10 day after inoculation and their mature stage (>90 % light absorbance) around 20 days after inoculation (Fig. 1). The biofilms grown in the absence of nitrate (LC2 and LC4) showed a slow linear increase in biomass and reached a maximum light absorbance of 25% after 30 days (Fig. 1). The average dry weight per slide after 31 days of incubation was approximately 0.12 g in the presence of nitrate and about 0.02 g in absence of nitrate. The biofilm grown in presence of nitrate grew relatively homogeneously in morphology as well as in color. The biofilm grown in the absence of NaNO3 was heterogeneous in color and thickness (Fig. 2A). Microscopic examination revealed that these biofilms contained a variety of unicellular and filamentous cyanobacteria. The dark green patches consisted mainly of Anabaena-like filaments (Fig. 2B and C).

101 Chapter 7

Figure 1. Growth curves of phototrophic biofilms growing with 0.5 mM

NaNO3 (LC1= and LC3=) and without NaNO3 (LC2= and LC4=). Biofilm development is indicated as the increasing light absorbance derived from the decrease of subsurface light.

Figure 2. A) Phototrophic biofilm cultivated in the absence of combined nitrogen (LC4). Dark green spots were dominated by Anabaena-like heterocystous cyanobacteria. Scale bar indicates one cm. B) Phase-contrast micrograph (400x) showing the dense heterogeneous biofilm matrix with various filamentous and unicellular cyanobacteria. The white arrow indicates a nostocales-like filament. C) Phase-contrast micrograph (400x) showing Anabaena-like filaments with heterocysts, derived from the dark green spots in the biofilm (Fig. 2A).

PCR-DGGE analysis of 16S rRNA gene fragments The 16S rRNA gene DGGE profiles obtained from the phototrophic biofilms cultivated in flow channels LC2 and LC4 were almost identical (Fig. 3). Both profiles show one thick dominant band at the top and three small bands in the lower part. The sequence obtained from the thick top band (band 1, was affiliated to a chloroplast of the diatom Nitschia frustulum. Band 2 was affiliated to the heterocystous cyanobacterium Anabaena variabilis, band 3 showed distant affiliation to Leptolyngbya sp. PCC 9221, and band 4 was affiliated to a Microcoleus sp. (Table 2).

102 Cyanobacterial uptake hydrogenases

PCR of hupS fragments from pure cultures and biofilm samples Primers hupS2345-F and hupS2605-R were successfully used to specifically amplify partial hupS sequences from genomic DNA that was extracted from 15 axenic cultures of diazotrophic cyanobacteria. In order to determine their specificity, the primers were also tested against various other diazotrophic and non-diazotrophic microorganisms. A weak product was obtained with genomic DNA extracted from Burkholderia tropica, DSM 15359 and Paenibacillus azotoxixans DSM 1735. However, these products had molecular weights between 800 and 1000 bp, instead of the expected 286 bp. Sequence analyses of these non-specific products showed no similarity to hupS or any other known gene. Surprisingly, no hupS could be amplified from the nitrogen fixing cyanobacterium Myxosarcina sp. CCY 0025. Amplification yielded several products of high molecular weight without sequence similarity to hupS or any other known gene. The results of this specificity experiment are summarized in Table 1. No hupS sequences could be amplified from the gDNA or cDNA derived from phototrophic biofilms grown with 0.5 mM NaNO3. A clear product of the expected 286 bp was amplified from cDNA from biofilms grown without NaNO3. However, the gDNA from this sample also yielded some high molecular weight by-product.

Figure 3. DGGE profiles of 16S rRNA gene fragments obtained after PCR amplification using primers specific for oxygenic phototrophs and genomic DNA samples from phototrophic biofilms cultivated under nitrogen limiting conditions in two separate flow channels (LC2 and LC3).

103 Chapter 7

Table 1. Specificity of the cyanobacterial hupS PCR.

a b c Organism Source N2-fixing PCR with hupS primers Cyanobacteria Anabaena variabilis SAG 1403-4B SAG + + Anabaena variabilis ATCC 29413 CCY + + Calothrix desertica SAG 35.79 SAG + + Calothrix sp. ATCC 27905 CCY + + Chlorogloeopsis fritschii ATCC 27193 CCY + + Cyanothece sp. CCY 00110 CCY + + Fischerella muscicola SAG 1427-1 CCY + + Gloeothece membranacae PCC 6502 CCY + + Lyngbya aestuari CCY 9616 CCY + + Mastigocladus laminosus SAG 4.84 SAG + + Myxosarcina sp. CCY 0025 CCY + +/- Nodularia spumigena CCY 9414 CCY + + Nostoc insulare SAG 54.79 CCY + + Nostoc muscorum PCC 7120 CCY + + Planktothrix agardii SAG 6.89 SAG - - Synechococcus sp. BG043511 CCY + + Synechocystis sp. PCC 6803 ATCC - - Bacteria Azotobacter vinelandii, DSM 85 DSMZ + - Burkholderia tropica, DSM 15359 DSMZ + +/- Clostridium pasteurianum, DSM 525 DSMZ + - Escherichia coli, ATCC 11775 ATCC - - Frankia sp., DSM 44263 DSMZ + - Paenibacillus azotoxixans DSM 1735 DSMZ + +/- Rhizobium leguminosarum, DSM 30132 DSMZ + - Vibrio diazotrophicus, DSM 2604 DSMZ + - Archaea Methanosarcina barkeri, DSM 3647 DSMZ + - Eukarya Scenedesmus obliquus SAG 276-12 SAG - - aCulture collections: ATCC; American Type Culture Collection, Rockville, Maryland. CCY; Culture Collection Yerseke, NIOO Centre for Estuarine and Marine Ecology, Yerseke, The Netherlands. DSMZ; Deutsche Sammlung von Microorganismen und Zellkulturen, Braunsweigh, Germany. NCCB: The Netherlands Culture Collection of Bacteria, Utrecht University, Utrecht, The Netherlands. SAG: Sammlung von Algenkulturen Georg-August-Universität, Göttingen, Germany. bOrganisms capable of fixing atmospheric nitrogen. cAmplification products yielded by applying primers hupS2345-F and hupS2605-R. - = no product, + = product of 286 kb, -/+ = non-specific product of high molecular weight.

104 Cyanobacterial uptake hydrogenases

Transcription of hupS in Anabaena variabilis HupS mRNA was detected by RT-PCR to examine the expression of the hupS genes in N2-fixing and non-N2-fixing cultures of Anabaena variabilis SAG 1403-4B. HupS transcripts were detected in cultures cultivated in modified BG11 medium without NaNO3. Transcripts were completely missing in cells grown in BG11 medium supplemented with 1 mM NaNO3 (Fig. 4).

Figure 4. Ethidium-bromide stained agarose gel of hupS PCR amplified from gDNA and cDNA derived from A. variabilis SAG 1403-4B grown in medium with and without nitrate. PCR reactions with extracted mRNA were performed to rule out contamination with gDNA. DNA isolated from N. spumigena CCY 9414 was used as a positive control. The amplified hupS fragments are 286 bp in size, as indicated by the molecular weight (MW) marker.

HupS clone libraries Sequence analysis of 33 randomly selected hupS clones obtained from both the gDNA and cDNA amplicons revealed a diverse assemblage of hupS gene sequences (Table 2). All sequences were homologues to cyanobacterial hupS. Both libraries contained exclusively hupS sequences affiliated to Nostocaceae. However, 28 out of 33 gDNA library clones were 90% similar to Nostoc sp. PCC 7422 while this sequence represented only 10 out of 33 clones from the cDNA library. The majority of the cDNA clones (23 out of 33) were 86% similar to Anabaena variabilis ATCC 29413. The gDNA library contained also contained one clone with 88% sequence similarity to Nostoc sp. PCC 7422 and four clones with 89% similarity to Anabaena variabilis ATCC 29413 (Table 2 and Fig. 5).

105 Chapter 7

Table 2. Phylogenetic summary based on clone library analysis of cyanobacterial hupS sequences and PCR-DGGE derived 16S rRNA gene sequences.

Representative Accession No. of % of Similarity (%) to closest relative in sequence no. clones clones Genbank (Blastn) gDNA- library 33 88%; Nostoc sp. PCC 7422, hupS clone gDNA-B7 EF431930 1 3 (AB237640) 90%; Nostoc sp. PCC 7422, hupS clone gDNA-B11 EF431947 28 85 (AB237640) 89%; Anabaena variabilis ATCC clone gDNA-B12 EF431931 4 12 29413, hupS (Y13216) cDNA-library 33 86%; Anabaena variabilis ATCC clone cDNA-D5 EF431938 23 70 29413, hupS (Y13216) 90%; Nostoc sp. PCC 7422, hupS clone cDNA-G5 EF431939 10 30 (AB237640) DGGE - 98%; Nitzschia frustulum chloroplast, band 16S-1 EF622220 - - 16S rRNA (AY221721) 99%; Anabaena variabilis,16S rRNA band 16S-2 EF622221 - - (AB016520) 92%; Leptolyngbya sp. PCC 9221, 16S band 16S-3 EF622222 - - rRNA (AF317507) 96%; Microcoleus sp. UTCC 296, 16S band 16S-4 EF622223 - - rRNA (AF218377)

Phylogenetic analysis of partial hupS sequences Amplified hupS sequences and hupS sequences from Genbank were subjected to NJ-analysis. The topology of the cyanobacterial hupS NJ-tree is largely consistent with trees based on 16S rRNA genes (Waterbury and Rippka 1989). Overall, the clustering of representative cyanobacterial hupS sequences was similar in DNA and amino acid based trees (data not shown), but there is greater sequence divergence in the former. Some sequences cluster closely together with high bootstrap values, whereas others are relatively deep branching, showing variation in the evolutionary distance among cyanobacterial groups (Fig. 5). HupS sequences obtained from heterocystous cyanobacteria appeared in a monophyletic cluster with non-heterocystous cyanobacteria as a sister group.

106 Cyanobacterial uptake hydrogenases

Figure 5. Neighbor-joining tree showing the phylogenetic affiliations of cyanobacterial hupS sequences. Sequences obtained from the present study are in bold. Scale bar indicates 5% sequence divergence. Bootstrap values (50%) for 1000 replicates are placed at the nodes of the branches.

DISCUSSION

Here we describe a molecular approach to analyze the diversity of uptake hydrogenase containing cyanobacteria in environmental samples. We showed that a PCR approach based on cyanobacterial 16S rRNA genes is not suitable to specifically investigate the diversity of hupS containing cyanobacteria. The DGGE profiles showed only a limited number of bands of which only one represented a hupS containing Anabaena-like cyanobacterium. This example demonstrates the general need to employ functional genes in addition to 16S rRNA genes in microbial community analysis (Case et al., 2007; Wilmotte 1994). Although cyanobacterial hupS genes are highly conserved with >80% identity,

107 Chapter 7

previous attempts to design primers specific for cyanobacterial hupS genes failed (Schmitz et al., 1993). In this study we successfully sequenced partial hupS genes from a variety of organisms. Together with sequences available from public databases, we have designed degenerate primers and developed a PCR protocol that exclusively amplifies hupS gene sequences from cyanobacteria. Although both primers contain degenerate bases, rigorous testing showed a clear specificity for cyanobacterial hupS. We showed that these primers can be utilized to detect hupS mRNA. Analysis of A. variabilis cultures confirmed that hupS transcription does not occur under NO3 saturated conditions in which the nitrogenase genes are not expressed (Happe et al., 2000). Clone library analyses demonstrated that this approach can be used to determine and identify presence and transcription of cyanobacterial hupS in complex microbial communities without prior taxonomic knowledge of the community composition. Non-heterocystous and heterocystous cyanobacteria, including unicellular strains, form two separate clusters (Fig. 4), which is in agreement with a previous phylogenetic analysis of a limited number of complete HupSL amino acid sequences (Tamagnini et al., 2005). The unicellular diazotrophic cyanobacterium Synechococcus sp. Miami BG043511, isolated from a subtropical marine environment, was originally classified as Synechococcus sp. (Leon et al., 1986)]. It has been suggested that this strain should be classified under the genus Cyanothece (Bergman et al., 1997; Nakamura et al., 2005; Waterbury and Ripka 1989). The position of this strain in our tree based on hupS sequences strongly supports this suggestion (Fig. 4). Blast analysis of clone gDNA-B7 (EF431930) indicated 88% sequence similarity with the hupS gene Nostoc sp. PCC 7422 (AB237640). Our phylogenetic analysis indicates 92% sequence similarity between clone gDNA-B7 and N. spumigena CCY 9414 (EF431942) (Fig. 4). This example illustrates that the limited number of sequences available to date hampers phylogenetic analysis. We anticipate that the databases of hupS sequences will expand rapidly, allowing better phylogenetic analyses. Unexpectedly, analysis of gDNA and cDNA clone library data showed a discrepancy between the presence and the transcription of cyanobacterial hupS in the nitrogen limited phototrophic biofilms. The gDNA library was dominated by Nostoc sp. PCC 7422, while the cDNA library was dominated by A. varibilis ATCC 29413. Differences in the composition of the gDNA and cDNA libraries may be attributed to distinct in situ regulation of hupS expression by closely related heterocystous cyanobacteria. It has been shown that the expression of hupL by Anabaena sp. PCC 7120 requires a developmental genome rearrangement. During heterocyst differentiation, a 10.5 kb DNA element is excised from within the hupL gene by a site-specific recombinase (Carrasco et al., 1995). In contrast, molecular experiments showed that no such rearrangements occur in Nostoc sp. strain PCC 73102 (Axelsson et al., 1999, Oxelfelt et al., 1998). The amplification of hupS genes from phototrophic biofilm cDNA could reflect a basal activity of the hupSL promoter, not necessarily leading to translation of hupS. Such translational control in cyanobacteria has been described for the psbA gene of Synechococcus sp. PCC 6803 (He and Vermaas 1998). However, comparison of growth indicates that the biofilms cultivated in light chambers LC2 and LC4 were nitrogen limited. Therefore, it is expected that active diazotrophic cyanobacteria present in the biofilm expressed nitrogenase genes and subsequently expressed uptake hydrogenases. In conclusion, we demonstrated a molecular approach to assess the hupS diversity and transcription of cyanobacterial members of a population without prior taxonomic knowledge of the community composition in environmental samples. The phylogentic

108 Cyanobacterial uptake hydrogenases

analysis of hupS sequences can provide additional information for evaluating phylotypes of diazotrophic cyanobacteria. In addition this approach allows the screening for diazotrophic strains lacking the hupSL genes, or strains exhibiting variant (e.g. low) hupSL expression levels, which might be applicable in biohydrogen production (Hansel and Lindblad 1998).

ACKNOWLEDGEMENTS

This research was supported by the European Union (PHOBIA project, contract QLK3-CT-2002-01938).

REFERENCES

Axelsson, R., Oxelfelt, F., and Lindblad, P. (1999) Transcriptional regulation of Nostoc uptake hydrogenase. FEMS Microbiol Lett 170:77-81 Bergman, B., Gallon, J.R., Rai, A.N., and Stal, L.J. (1997) N-2 fixation by non-heterocystous cyanobacteria. FEMS Microbiol Rev 19:139-185 Bothe, H., Tennigkeit, J., and Eisbrenner, G. (1977) The utilization of molecular hydrogen by the blue- green alga Anabaena cylindrica. Arch Microbiol 114:43-49 Case, R.J., Boucher, Y., Dahllof, I., Holmstrom, C., Doolittle, W.F. and Kjelleberg, S. (2007) Use of 16S rRNA and rpoB genes as molecular markers for microbial ecology studies. Appl Environ Microbiol, 73:278-288 Carrasco, C.D., Buettner, J.A., and Golden, J.W. (1995) Programmed DNA rearrangement of a cyanobacterial hupL gene in heterocysts. Proc Natl Acad Sci U S A 92:791-795 Chenna, R., Sugawara, H., Koike, T., Lopez, R., Gibson, T.J., Higgins, D.G., and Thompson, J.D. (2003) Multiple sequence alignment with the Clustal series of programs. Nucleic Acids Res 31:3497- 3500 Hansel, A., and Lindblad, P. (1998) Towards optimization of cyanobacteria as biotechnologically relevant producers of molecular hydrogen, a clean and renewable energy source. Appl Microbiol Biotechnol 50:153-160 Happe, T., Schutz, K., and Bohme, H. (2000) Transcriptional and mutational analysis of the uptake hydrogenase of the filamentous cyanobacterium Anabaena variabilis ATCC 29413. J Bacteriol 182:1624-1631 He, Q., and Vermaas, W. (1998) Chlorophyll a availability affects psbA translation and D1 precursor processing in vivo in Synechocystis sp. PCC 6803. Proc Natl Acad Sci U S A 95:5830-5835 Kumazawa, S., and Mitsui, A. (1985) Comparative amperometric study of uptake hydrogenase and hydrogen photoproduction activities between heterocystous cyanobacterium Anabaena cylindrica B629 and nonheterocystous cyanobacterium Oscillatoria sp. strain Miami BG7. Appl Environ Microbiol 50:287-291 Leon, C., Kumazawa, S., and Mitsui, A. (1986) Cyclic appearance of aerobic nitrogenase activity during synchronous growth of unicellular cyanobacteria. Curr Microbiol 13:149-153 Mancinelli, R.L. (1996) The nature of nitrogen: an overview. Life Support Biosph Sci 3:17-24 Nakamura, Y., Takahashi, J., Sakurai, A., Inaba, Y., Suzuki, E., Nihei, S., Fujiwara, S., Tsuzuki, M., Miyashita, H., Ikemoto, H., Kawachi, M., Sekiguchi, H., and Kurano, N. (2005) Some cyanobacteria synthesize semi-amylopectin type alpha-polyglucans instead of glycogen. Plant Cell Physiol 46:539-545 Oxelfelt, F., Tamagnini, P., and Lindblad, P. (1998) Hydrogen uptake in Nostoc sp. strain PCC 73102. Cloning and characterization of a hupSL homologue. Arch Microbiol 169:267-274 Roeselers, G., van Loosdrecht, M. C. M., and Muyzer, G. (2007) Heterotrophic pioneers facilitate phototrophic biofilm development. Microb Ecol DOI: 10.1007/s00248-007-9238-x Schmitz, O., Kentemich, T., Zimmer, W., Hundeshagen, B., and Bothe, H. (1993) Identification of the nifJ gene coding for pyruvate: ferredoxin oxidoreductase in dinitrogen-fixing cyanobacteria. Arch Microbiol 160:62-67 Schutz, K., Happe, T., Troshina, O., Lindblad, P., Leitao, E., Oliveira, P., and Tamagnini, P. (2004) Cyanobacterial H(2) production - a comparative analysis. Planta 218:350-359

109 Chapter 7

Stanier, R.Y., Kunisawa, R., Mandel, M., and Cohen-Bazire, G. (1971) Purification and properties of unicellular blue-green algae (order Chroococcales). Bacteriol Rev 35:171-205 Tamagnini, P., Axelsson, R., Lindberg, P., Oxelfelt, F., Wunschiers, R., and Lindblad, P. (2002) Hydrogenases and hydrogen metabolism of cyanobacteria. Microbiol Mol Biol Rev 66:1-20 Tamagnini, P., Leitao, E., and Oxelfelt, F. (2005) Uptake hydrogenase in cyanobacteria: novel input from non-heterocystous strains. Biochem Soc Trans 33:67-69 Tamagnini, P., Troshina. O., Oxelfelt. F., Salema. R., and Lindblad, P. (1997) Hydrogenases in Nostoc sp. Strain PCC 73102, a strain lacking a bidirectional enzyme. Appl Environ Microbiol 63:1801-1807 Tsygankov, A.A., Borodin, V.B., Rao, K.K., and Hall, D.O. (1999) H(2) photoproduction by batch culture of Anabaena variabilis ATCC 29413 and its mutant PK84 in a photobioreactor. Biotechnol Bioeng 64:709-715 Vignais, P.M., and Colbeau, A. (2004) Molecular biology of microbial hydrogenases. Curr Issues Mol Biol 6:159-188 Waterbury, J.B., and Rippka, R. (1989) Subsection I, order Chroococcales Wettstein 1924, emend. Rippka et al. 1979., p. 1728–1746. In J. T. Staley, M. P. Bryant, N. Pfennig, and J. G. Holt (ed.), Bergey’s Manual of Systematic Bacteriology, vol. 3. Williams & Wilkins, Baltimore Wilmotte, A. (1994) Molecular evolution and taxonomy of the cyanobacteria, p. 1-25. In D. A. Bryant (ed.), The Molecular Biology of the Cyanobacteria. Kluwer, Dordrecht, Netherlands Zani, S., Mellon, M.T., Collier, J.L., and Zehr, J.P. (2000) Expression of nifH genes in natural microbial assemblages in Lake George, New York, detected by reverse transcriptase PCR. Appl Environ Microbiol 66:3119-3124 Zippel, B., and Neu, T.R. (2005) Growth and structure of phototrophic biofilms under controlled light conditions. Water Sci Technol 52:203-209

110 8 Concluding remarks

Summary

Samenvatting

About the author

List of publications

Acknowledgements Chapter 8

CONCLUDING REMARKS

Phototrophic biofilms are light-driven microbial communities attached to solid substrata. They can be found virtually anywhere in nature, and they are commercially interesting for several reasons (see Chapter 2). The research described in this thesis aimed to reveal details about the community structure of phototrophic biofilms in view of their potential applications. In order to achieve this aim the microbial community composition of natural occurring and cultivated phototrophic biofilms was determined using mainly molecular methods. Special attention was paid to the successional changes in community composition that take place during biofilm development and the relationship between biodiversity and primary productivity. In addition, we have developed molecular strategies that can be used to characterise the diversity of specific groups of cyanobacteria.

Chapter 2 provides a brief introduction in the ecology of phototrophic biofilms and discusses their actual and potential applications. These include promising applications in wastewater treatment, bioremediation, fish-feed production, soil improvement, biohydrogen production, soil improvement and the development of anti-fouling agents. This chapter shows that phototrophic biofilms are highly versatile. However, many applications are in their infancy and as yet not ready for commercial adaptation and exploitation. In Chapter 3 we describe the genotypic diversity of oxygenic and anoxygenic phototrophic microorganisms in microbial mat samples collected from three hot spring localities on the east coast of Greenland. These hot springs harbour unique Arctic microbial ecosystems that were never studied in detail before. Our results indicate that the cyanobacterial community composition in the samples were different for each sampling site. Different layers of the same heterogeneous mat often contained distinct and different communities of cyanobacteria. We observed a relationship between the cyanobacterial community composition and the in situ temperatures of different mat parts. Thermophilic cyanobacteria of the genus Synechococcus are absent in Iceland and Alaska while in the western United States thermophilic Synechococcus species of this morphotype are found in chemically diverse hot springs. Because all thermophilic forms of Synechococcus appear to be absent in Iceland, the detection of members of this genus in the Greenland springs was quite surprising, and may reflect the fact that these springs are found in a geologically much older setting than the springs on Iceland. Chapter 4 shows that the development of phototrophic biofilms under controlled conditions involves a series of successional community changes. The influence of light on the biofilm growth rate and the community composition appeared more prevalent during the initial phase of biofilm development. We also showed that phototrophic biofilms developed faster on polycarbonate surfaces that were precolonized by heterotrophic bacteria. This observation illustrates that although a mature biofilm community can be dominated by a few species, there is also a minor set of organisms, e.g. heterotrophic bacteria, that may be crucial for the initial establishment and the consequent development of that biofilm community. The variability and reproducibility of model-ecosystems was addressed in Chapter 5, which describes the simultaneous cultivation of phototrophic biofilms in three identical flow-lane microcosms located in different laboratories. The growth rates of the biofilms under different light regimes were similar in the three different microcosms, but DGGE analysis of both 16S and 18S rRNA gene fragments showed that the communities developed differently

112 Summary and concluding remarks in terms of species richness and community composition. This difference could be explained by small and unidentified differences in the operating conditions of the three incubators, but it has also been postulated that chaotic dynamics and other nonlinear phenomena can play a role in community development. Although empirical evidence of chaos, or complex behaviour, in ecosystems is scarce, it is possible that the observed variation results from intrinsic complex and even chaotic behaviour of microbial communities. Our results demonstrate that secure experimental validation of reproducibility is essential for the use of microcosm systems in microbial ecology studies, especially for conclusions concerning differences in community composition and biodiversity in benthic systems. The direct amplification and sequencing of 16S rRNA genes from the environment revolutionized microbial ecology and permanently changed the way we study microorganisms in the environment. The notion that rRNA genes could identify an organism by reconstructing its phylogeny, along with the possibility of storing sequences in databases, resulted in the rapid adoption of the 16S rRNA gene by microbiologists. This gene has now established itself as the standard in bacterial phylogeny and in microbial ecology. Lately, several functional genes were shown to be useful for the same purpose because their phylogeny is congruent or very similar to phylogenetic relationships based on 16S rRNA gene sequence analyses. These marker genes allow to describe the functional diversity of a specific environment. In addition, functional genes often provide a resolution below species level because of higher evolutionary rates of the less conserved functional molecules. In Chapter 6 we have designed degenerate primers after comparative analysis of nifD gene sequences from public databases, and developed a PCR protocol for the amplification of nifD sequences from cyanobacteria. The primers were tested on a variety of nitrogenase-containing and nitrogenase-lacking bacteria. By using this protocol, we amplified nifD sequences from DNA that was isolated from phototrophic microbial communities. We think that this approach will be useful to investigate the phylogenetic affiliation, activity and, ecological significance of diazotrophic cyanobacteria in complex microbial communities. Finally, Chapter 7 describes a PCR approach for the specific amplification of cyanobacterial uptake hydrogenases. This approach allows the screening of cyanobacteria from different habitats for the presence and transcription of hup genes. Diazotrophic strains lacking the hupSL genes, or strains exhibiting variant (e.g. low) hupSL expression levels might be applicable in biohydrogen production. The phylogenetic analysis of hupS as well as nifD sequences can provide additional information for evaluating phylotypes of diazotrophic cyanobacteria. These examples demonstrate the general need to employ functional genes in addition to 16S rRNA genes in microbial community analysis.

In conclusion, our research on the microbial ecology of photrophic biofilms contributes to better understanding of these ecosystems and their behaviour under different environmental conditions. A common theme in most chapters presented in this thesis is that we observed that the community composition of phototrophic biofilms is characterized by a high degree of temporal and spatial variability. We described the spatial heterogeneity in the cyanobacterial communities of hot spring microbial mats. We showed that succesional changes take place during biofilm development. Surprisingly we found different community compositions in incubators operated under the same conditions. As suggested in the general introduction of this thesis (Chapter 1) biofilms are as diverse as their constituent microorganisms. Hence, it could be argued that it is impossible to encompass all their

113 Chapter 8

aspects by this one definition “biofilm”. Although phototrophic biofilms appear to be a very specific biofilm manifestation, they are formed on surfaces in a wide range of terrestrial and aquatic environments. The most general conclusion that can be drawn from the work presented in this thesis is that the community composition of phototrophic biofilms is idiosyncratic to their particular successional stage and the environmental conditions. It is justified in the author’s opinion to conclude this thesis that it is probably impossible to develop a unifying model of phototrophic biofilm growth that can provide meaningful predictions about the biodiversity and phylogeny of all biofilm constituents. The best strategy to gain inside in phototrophic biofilm ecology in view of their potential applications is to study them under conditions relevant for each specific application. The continuous advances in molecular microbiology and ecology will provide improved tools for detailed studies of phototrophic biofilm ecology.

114 Summary and concluding remarks

SUMMARY

Biofilms are layered structures of microbial cells and an extracellular matrix of polymeric substances, associated with surfaces and interfaces. Biofilms trap nutrients for growth of the enclosed microbial community and help prevent detachment of cells from surfaces in flowing systems. Phototrophic biofilms can best be defined as surface attached microbial communities mainly driven by light as the energy source with a photosynthesizing component clearly present. Eukaryotic algae and cyanobacteria generate energy and reduce carbon dioxide, providing organic substrates and oxygen. The photosynthetic activity fuels processes and conversions in the total biofilm community, including the heterotrophic fraction. This thesis starts with a brief introduction in the ecology of phototrophic biofilms and discusses their actual and potential applications in wastewater treatment, bioremediation, fish-feed production, biohydrogen production, and soil improvement and their role in biofouling. The next chapter describes the diversity of phototrophic bacteria in hot spring microbial mats found on the east coast of Greenland. In this study we utilized a polyphasic approach using a combination of isolation techniques, microscopic observation of morphological features, and cultivation-independent molecular methods. We observed a relationship between the cyanobacterial community composition and the in situ temperatures of different microbial mat parts. Chapter 4 focuses on the successional changes in community composition of freshwater phototrophic biofilms growing under different light intensities. Our results suggest that surface colonization by heterotrophic pioneers facilitates the development of phototrophic biofilms. In Chapter 5 we compared the community composition of phototrophic biofilms cultivated in three microcosm systems operated under identical conditions but placed in different laboratories. Denaturing Gradient Gel Electrophoresis (DGGE) analysis of both 16S and 18S rRNA gene fragments showed that the communities developed differently in terms of species richness and community composition. Chapter 6 demonstrates that nifD gene sequences, coding for a nitrogenase subunit, can be used to detect and identify diazotrophic cyanobacteria in natural communities. PCR products generated using primers homologous to conserved regions in the cyanobacterial nifD genes were subjected to DGGE and clone library analysis in order to determine the genetic diversity of diazotrophic cyanobacteria in environmental samples. In the last chapter we describe the development of PCR primers targeting conserved regions within the cyanobacterial hupS gene family. This gene is involved in the hydrogen metabolism of diazotrophic microorganisms. We analyzed hupS diversity and transcription in cultivated phototrophic biofilms by the direct retrieval and analysis of mRNA that was reverse transcribed, amplified with hupS specific primers, and cloned. Overall, the community composition and species richness of phototrophic biofilms was shown to be highly variable. Cultivation-independent molecular methods proved very useful to study diversity and function in phototrophic biofilms.

115 Chapter 8

SAMENVATTING

Een biofilm bestaat uit microörganismen ingebed in een matrix van celproducten, aangehecht op een oppervlak. Biofilmen voorkomen dat microörganismen wegspoelen uit een milieu. Fototrofe biofilmen zijn afhankelijk van licht als energie bron. Ze bestaan voornamelijk uit fototrofe microörganismen zoals eencellige algen, diatomeeën en cyanobacteriën. De fotosynthese activiteit levert de energie voor allerlei processen en omzettingen in de biofilm, inclusief de groei van heterotrofe microörganismen. Dit proefschrift begint met een korte inleiding in de ecologie van fototrofe biofilmen en een beschrijving van mogelijke toepassingen in afvalwater zuivering, bioremediatie, visvoer productie, biologische waterstof productie, bodem verbetering en biofouling. Het volgende hoofdstuk beschrijft de diversiteit van fototrofe bacteriën in microbiële matten in heetwater bronnen op Groenland. In deze studie combineerden we isolatie technieken, microscopische analyses en cultivatie-onafhankelijke moleculaire technieken. Een duidelijk verband kon worden gelegd tussen de cyanobacteriële soorten samenstelling en de in situ temperatuur van verschillende microbiële mat delen. Hoofdstuk 4 beschrijft de successieve veranderingen in soorten samenstelling in een gecultiveerde fototrofe biofilm. Onze resultaten lijken aan te tonen dat de kolonisatie van een oppervlak door heterotrofe bacteriën de ontwikkeling van fototrofe biofilmen stimuleert. In hoofdstuk 5 vergelijken we de soorten samenstelling van fototrofe biofilmen gekweekt onder identieke omstandigheden in verschillende laboratoria. Denaturing Gradient Gel Electrophoresis (DGGE) analyse van 16S en 18S rRNA gen fragmenten toonde aan dat de soorten samenstelling en biodiversiteit in de ontwikkelende biofilmen sterk verschilde. Hoofdstuk 6 laat zien hoe nifD gen sequenties, coderend voor een onderdeel van het nitrogenase complex, gebruikt kunnen worden voor de detectie en identificatie van stikstoffixerende cyanobacteriën in het milieu. PCR product verkregen met primers specifiek voor geconserveerde delen in cyanobacteriële nifD genen werden geanalyseerd met behulp van DGGE en clone libraries. Hiermee kon de genetische diversiteit van stikstoffixerende cyanobacteriën in fototrofe biofilmen en microbiële matten bepaald worden Het laatste hoofdstuk (7) beschrijft de ontwikkeling van PCR primers specifiek voor geconserveerde delen in het cyanobacteriële hupS gen. Dit gen is betrokken bij de waterstof huishouding van stikstoffixerende microorganismen. We analyseerde de transcriptie en genetische diversiteit van hupS in gecultiveerde fototrofe biofilmen door middel van de directe extractie en analyse van hupS mRNA. In het algemeen, bleek de soorten samenstelling en biodiversiteit in fototrofe biofilmen erg variabel. Cultivatie onafhankelijke moleculaire technieken bleken zeer geschikt om de diversiteit en activiteit in fototrofe biofilmen te bestuderen.

116 Summary and concluding remarks

CURRICULUM VITAE

Guus Roeselers was born on the 4th of October 1976 in Leeuwarden, the Netherlands. He grew up in Maastricht where he finished his pre-university education at the Stedelijke Scholen Gemeenschap Maastricht. In 1997 he went to study Biology at Utrecht University. In 2002 he performed a research internship at the Molecular Microbiology group, where he studied phase variation in root colonising Pseudomonas strains, under supervision of Prof. dr. Jan Tommassen. In 2003 he moved to Madrid where he carried out an internship at the Centro Nacional de Biotecnología (CNB). There he worked under supervision of dr. Regla Bustos and Prof. dr Javier Paz-Ares on the characterisation of transcription factors involved in the ‘phosphate starvation response’ in Arabidopsis thaliana. Back in the Netherlands he performed a literature study under supervision of dr. Peter Bakker (Utrecht University, Phytopathology) on the introduction of genetically modified microorganisms in the rizosphere. In May 2003 he obtained his Msc degree and it was clear that he wanted to continue in science. From May 2003 to May 2007 he worked as a Ph.D student at the section Environmental Biotechnology at Delft University of Technology. Under supervision of Prof. dr. ir. Mark van Loosdrecht and dr. Gerard Muyzer he performed the research that is presented in this thesis. Currently he works as a postdoctoral researcher at the section Environmental Biotechnology on the potential application of microbial crystal formation in the solidification of soils. In July 2007 he was awarded a Rubicon grant from the Netherlands Organisation for Scientific Research (NWO) to work on chemoautotrophic symbioses in the group of Prof. dr. Colleen Cavanaugh at Harvard University.

117 Chapter 8

LIST OF PUBLICATIONS

Roeselers, G., Huisjes, E.H., van Loosdrecht, M.C.M., and Muyzer, G. Diversity and expression of cyanobacterial hupS genes in pure cultures and in a nitrogen limited phototrophic biofilm. Submitted for publication in FEMS Microbiology Ecology

Roeselers, G., Norris, T.B., Castenholz, R.W., Rysgaard, S., Glud, R., Kühl, M., and Muyzer, G. (2007) Diversity of phototrophic bacteria in microbial mats from Arctic hot springs (Greenland) Environmental Microbiology. 9(1): 26–38

Roeselers, G., van Loosdrecht M.C.M., and Muyzer G. (2007) Heterotrophic pioneers facilitate phototrophic biofilm development. Microbial Ecology doi:10.1007/s00248-007-9238-x

Roeselers, G., Stal, L., van Loosdrecht, M.C.M., and Muyzer, G. (2007) Development of a PCR for the detection and identification of cyanobacterial nifD genes. Journal of Microbiological Methods doi:10.1016/j.mimet.2007.06.011

Roeselers, G., van Loosdrecht, M.C.M., and Muyzer, G. (2007) Phototrophic biofilms and their potential applications. Journal of Applied Phycology in press

Roeselers, G., Zippel, B., Staal, M., van Loosdrecht, M.C.M., and Muyzer G. (2006) On the reproducibility of microcosm experiments - differential community compositions in parallel phototrophic biofilm microcosms. FEMS Microbiology Ecology. 58(2):169-178

Roeselers G., van Loosdrecht M.C.M., and Muyzer G. (2004) Biodiversity of phototrophic biofilms. In W. Verstraete (Ed.), Proceedings of the European Symposium on Environmental Biotechnology, ESEB 2004, Belgium. pp. 161-164. London: Taylor & Francis Group.

118 Summary and concluding remarks

ACKNOWLEDGEMENTS

Nu we het einde van dit verhaal naderen blijkt het schrijven van een puik dankwoord nog een lastige klus. Wat is me het meest bij gebleven aan mijn tijd in het Kluyver lab, waar ging ik bijna aan onderdoor, en wie heeft me er op de meest cruciale momenten door heen gesleept? Je ontkomt haast niet aan een lange opsomming van personen en anekdotes waarbij je onvermijdelijk essentiële dingen vergeet! Cliché of niet! Er zijn een heleboel mensen aan wie ik dank ben verschuldigd. Een aantal daarvan wil ik expliciet noemen. Allereerst mijn co-promotor, Gerard Muyzer. Ik heb onze samenwerking als uitermate plezierig ervaren. Je hebt me goed wegwijs gemaakt in het wetenschappelijke wereldje. Ook heb je gezorgd dat ik kon deelnemen aan de Microbial Diversity Course in Woods Hole, een enorm inspirerende ervaring. Mijn promotor, Mark van Loosdrecht, bedank ik voor de vrijheid die hij me heeft gegeven. Je hebt je niet bemoeid met details maar had wel een goed oog voor de grote lijnen, wat minstens zo belangrijk is. Hoe druk je het ook had, je wist altijd wel een gaatje te vinden voor een brainstorm sessie. Gijs Kuenen, jou enthousiasme, inspiratie en wijze raad, waren zeer waardevol, en hebben niet alleen richting gegeven aan dit onderzoek maar ook aan mijn verdere wetenschappelijke carrière. I want to acknowledge all participants in the PHOBIA project. Special thanks to Michael Kühl, Lukas Stal, Marc Staal, Barbara Zippel, and Roland Thar for the pleasant collaboration and discussions. I would like to thank Richard W. Castenholz, who contributed so much to the publication of the “Greenland story”. Then the environmental biotechnology group: my two dear Indian friends Shabir Dar and Raji Kumariswami with whom I shared room 0.100, I truly enjoyed our long discussions about science, politics, globalisation, religion and women’s emancipation. Dimitry Sorokin, It was a pleasance to work with you. Your knowledge about microbiology seems at times inexhaustible and you were always available for help and advise. A big ‘Thanks’ to: Mirjam Foti, Ann-Charlotte Toes, Leon van Paasen, Sander Hogewoning, Kees (Mac guru) van Sluis, Emel Sahan, Horia Banciu, Krisztina Gabor, Geert van der Kraan, Henk Jonkers, Merle de Kreuk, Wouter van der Star, Christian Picioreanu, Marlies Kampschreur, Katja Schmid, Jasper Meijer, Jiang Yang, Margarida Temudo, Weren de Vet and Robbert Kleerebezem who made my stay in Delft not only interesting but also fun. Esengül Yildirim jij hebt me de op meest chaotische momenten in het MB-lab steeds uit de brand geholpen. Udo van Dongen (kenner van obscure noise bandjes en frikandellen fijnproever) we hebben een hoop lol gehad. Ik wil de Master en Bachelor studenten bedanken met wie ik zeer plezierig heb samengewerkt: Yun Yun Zou, Natascha Helder, Fenna Kester en Eline Huisjes. Hulde voor het ondersteunend personeel: Sjaak, Jos, Astrid, Appolena, Lesly, Frieda, Hans, Marcel, Rob, Dirk, Robbert, Herman, Arno, Rob, Ronald en iedereen die ik alsnog vergeet. Verder neem ik met lichte weemoed afscheid van het oude Kluyverlab, een historische plek waar de geest van Bejierinck nog steeds rond waart. Ik ben trots op het feit dat ik als microbioloog getraind ben in “the Delft School of Microbiology”. Jef, Monique en Julia bedankt voor jullie onvoorwaardelijke steun. Als laatste bedank ik mijn allerliefste liefjes Clara en Stella zonder wie niets van dit alles mogelijk geweest zou zijn.

119