Naturwissenschaften (2003) 90:136–140 DOI 10.1007/s00114-003-0403-x
Ulrike Brehm · Wolfgang E. Krumbein · Katarzyna A. Palin´ska
Microbial spheres: a novel cyanobacterial–diatom symbiosis
Received: 4 July 2002 / Accepted: 13 January 2003 / Published online: 12 February 2003 Springer-Verlag 2003
Abstract Cyanobacteria, algae and bacteria are widespread inhabitants of North Sea microbial mats. Our studies of these populations showed uncommon modes of living and extraordinary structures, which have not been described before. The structures are spherical objects covering a community of cyanobacteria, diatoms and bacteria. The cultivation of these communities in the laboratory and intensive observations of their exceptional movement has led to some spectacular findings. The sphere formations go through different phases with variation in the dominance of different microorganisms. The role of the bacteria is the most important in the first phase, and can be increased by the addition of signal substances. Spheres surrounded by envelopes of unknown composition and permeability appear, with numerous bacteria and sporadic diatoms inside. Then the cyanobacteria penetrate the spheres and arrange themselves at the surface. The communities proliferate over some weeks and are finally released. Laboratory expositions of the microbial communities to different parameters pinpoint the limits of sphere formation. The metabolic products of the sphere communities are concentrated in the spheres and lead to a different kind of compound compared with the surrounding environment. In this way, the microbial communities strongly influence the structure of the sediments. Uncommon circular structures, which develop into spheres between 0.08 and 3 mm in size were found in subcultures of non-axenic filamentous cyanobacteria enrichments from North Sea microbial mats. These filamentous cyanobacteria (Phormidium sp.) together with associated benthic diatoms of the genus Navicula and associated heterotrophic bacteria were held as reproducible synergistic cultures. Phormidium sp. filaments tightly intertwined with each other, formed the U. Brehm ()) · W. E. Krumbein · K. A. Palin´ska Institute for Chemistry and Biology of the Marine Environment, AG Geomicrobiology, Carl von Ossietzky University of Oldenburg, PO Box 2503, 26111 Oldenburg, Germany e-mail: [email protected]
surface of the spheres, trapping diatoms inside. The formation of “spheres” was the result of radial and synchronous movements of the cyanobacteria. In old cultures, the direction of the cyanobacterial movement has turned in the opposite direction, away from the sphere. The integrity of large “spheres” was influenced by chemotactic phenomena and maintained by some type of trichome–trichome interaction. This suggests the presence of metabolic secondary products, which attract cyanobacteria and influence their movement in a form of chemotactic response.
Introduction Filamentous cyanobacteria from the LPP-group (Lyngbya–Plectonema–Phormidium), diatoms and chemoorganotrophic bacteria are the dominant microorganisms found in North Sea microbial mats (Gerdes et al. 1986). Cultivated on solid (0.5% agar) or liquid media (Rippka et al. 1979), the cyanobacteria produce motile trichomes. They become motile in order to optimize their positions in their microhabitat using a number of external factors. Light is the major factor in directing their search for a suitable survival and growth niche (Castenholz 1982). Organic substances and their concentrations may be another trigger. However, chemotaxis has been very poorly described in cyanobacteria. A large potential area for research has therefore been neglected (Kangatharalingam et al. 1991; Budrene and Berg 1995). Aggregates of simpler geometry have been often observed in planktonic Aphanizomenon flos-aquae (Janson et al. 1999). The integrity of large flakes in this case is probably influenced by chemotactic phenomena and is maintained by some type of chemical trichome–trichome interaction. The coordinated movement and previously unknown three-dimensional spherical arrangement of Phormidium sp. filaments in close connection with other organisms is reported here. The possible strategies for this new type of
trichome aggregation and entrapment of diatoms are presented and discussed.
Materials and methods The isolates were grown in artificial seawater medium ASNIII, prepared according to Rippka et al. (1979). Liquid cultures were maintained at 24C in Erlenmeyer flasks, without further aeration and illuminated with Osram tungsten light tubes providing a photosynthetic photon flux density (PPFD) of 120 mol photons m2 s1 (measured with a LICOR LI-185B quantum/radiometer/ photometer equipped with a LI-190SB quantum sensor) and with a light/dark cycle of 12/12 h (photosynthetic photon flux density of 120 mol photons m2 s1). The cultures used in our experiments are not axenic, as cyanobacteria, diatoms, and bacteria are always present in the culture medium. The diatoms belong to the genera Navicula perminota and Nereneis sp. However, filamentous cyanobacteria are clearly the dominant species regulating the development of the spheres. In order to verify the chemotactic hypothesis, experiments were set up using the following well known signal substances: cAMP (cyclic adenosine monophosphate), BHL (butyryl homoserine lactone), and OHHL (oxohexanoyl homoserine lactone). All chemicals were added separately in a final concentration of 10 mM to solid (0.5% agar) and liquid ASN III media inoculated with small amounts of the mixed consortia culture. BHL was additionally used at 1 and 5 mM concentrations. In this set of experiments, three different temperatures were tested (17, 24, and 27C). Samples for TEM (transmission electron microscopy) were prepared according to Luft (1961) and analysed using a Zeiss TEM. Samples for SEM (scanning electron microscopy) were prepared after Sargent (1988). Light microscopy was performed with an inverted microscope (Zeiss Axiovert). Motility was quantitatively determined with the help of the same microscope by taking a series of photographs every 15 s over defined periods of time (20 min).
Results and discussion A synergistic microbial consortium was isolated from North Sea microbial mats showing a number of different behavioural patterns in continued subculture for 3 years at transfer intervals of 4–6 weeks. One cyanobacterium (Phormidium sp.), one major diatom (Navicula perminota) and chemoorganotrophic bacteria invariably form spherical synergistic aggregates. Tightly intertwined filaments of cyanobacteria form the outer thin layer of the spheres. Increasing numbers of diatoms fill the inner parts of the spheres. Cyanobacteria can enter and leave the spheres and sometimes form hairy outer shapes because of some of the trichomes stick out from the spheres in tapering bundles. When the spheres get completely filled with diatoms and their extracellular polymeric substances (EPS) a mature sphere may spontaneously disintegrate. Principles of movement Different types of gliding trichomes were observed on agar, displaying a different but constant velocity in repeated microscopic measurements. The majority of the cyanobacteria when outside the spheres were slowly moving and
packed close together, all moving in the same direction. In contrast, the trichomes, which constructed the spheres, were constantly and actively moving in and on the surfaces of the spheres. The highest velocity of the Phormidium sp. trichomes spreading across the agar plate at room temperature was about 0.5 mm per minute (8 m/s). Motility in some cyanobacteria, e.g. the heterocystous forms, is periodic, and occurs in some phases of their morphogenetic cycles (hormogonia). Motility in microorganisms implies selective advantage. From our observations, specific substances are the triggers for movement acceleration. The possession of motility in microorganisms implies selective advantage. Motility in some cyanobacteria, e.g. the heterocystous forms, is occasional, and occurs in some phases of their morphogenetic cycles (hormogonia). Movement and/or growth can serve as a mean of spreading an otherwise attached organism over a wider area. Tactic responses could explain the directional orientation to many of the movements. Gliding is movement in contact with a solid or semi-solid surface without flagellalike organs (Castenholz 1982). This movement can be continuous in one direction for a long time or can occur at frequent intervals. Most members of the Oscillatoriaceae rotate when gliding. Trichomes moving on the surface of the agar often move in wide arcs, resulting in flat circles of many trichomes with non-uniform diameters. Castenholz (1982) explained circling of trichomes by a lateral force caused by a rotating trichome. In our studies this phenomenon was observed at a late stage of cultivation. Such trichome “deformations” or the occurrence of unexpected colony forms may occur with collisions and with portions of the same trichome gliding together in opposite directions. Assemblages of biofilm communities The filamentous cyanobacteria studied here are characterized by single segments of 3 m width and 4–4.5 m length. They belong to the LPP group (Lyngbya, Plectonema, Phormidium) as defined by Rippka et al. (1979). Twelve days after incubation a number of cyanobacteria trichomes formed circles. During the next few days the circles remained stationary but developed into spheres composed of a small number of individuals, which could be clearly identified. Cyanobacterial trichomes seemed to glide inside the spherical envelopes, apparently overcoming mechanical resistance of a chemical barrier, and then arranged themselves inside underneath the surface. The spheres were located within the agar, or randomly in liquid medium. The number of associated diatoms was found to be considerably higher inside the sphere than outside. This process was a result of accumulation of diatoms and their daughter cells trapped inside the sphere. The improved propagation and protection from grazing are potential advantages of the sphere formation and organization.
Fig. 1 a Radial movements of Phormidium trichomes towards and against the sphere (light microscopy). b Envelope appearance of a sphere (arrows) (light microscopy). c Sphere with diatoms inside (arrows) (light microscopy)
In several cultures a radial arrangement of cyanobacteria emerged from the spheres in form of tapered reemerging trichome bundles (Fig. 1a). The surfaces of the spheres were composed of permanently moving trichomes. The position of the spheres did not change with time, but the number of trichomes on their surface was variable. After 4–6 weeks the spheres grew pale, appearing as dense concentric circles. Starting at the surface of the spheres, the cyanobacteria formed dense spokes radiating from the centre of the spheres. The number of trichomes joining the circles increased over time. The dark green colour of the rings was conspicuous in comparison with the pale greenish hue of cyanobacteria in trichomes in the adjacent agar or liquid medium. The spheres maintain their original shape and position, although the trichomes are continuously moving. The movement of the aggregated trichomes appears to be coordinated, which in some respects is reminiscent of the E. coli “slugs” described by Budrene and Berg (1995). This means: the trichomes move together as a sphere. In cases where trichomes move away, the sphere turns. According to the literature this pattern of aggregation and locomotion of aggregates has not been reported. This is the first report of coordinated aggregation and movement.
The size of the spheres did not change over the whole time of observation. Finally, the rings were reduced in width and number of trichomes. Formation of spherical aggregates of trichomes The steps of sphere formation are as follows (according to microscopic observations). First, a marine microbial community grown in liquid or solid medium excretes recognizable substances aggregating into spheres of 0.8–3 mm diameter. The number of spheres occurring in a given culture time is increased by signal substances. In certain places bacterial accumulations appear, which are surrounded by microscopically visible envelopes (Fig. 1b). These spherical structures are recognized by filamentous cyanobacteria. They approach the spheres and penetrate through the envelopes at certain specific places only. The envelopes seem to be permeable for specific microorganisms. Bacteria and diatoms cannot penetrate through the flexible but resistant envelopes. Once trichomes of Phormidium sp. have penetrated into a sphere, they still can glide inward and outward. After they enter completely into the sphere and have arranged themselves
Fig. 2 a Two types of sphere: 1 a smooth one and 2 a hairy one (light microscopy). b Section detail of a sphere (cryo-SEM). c Sphere formation. Trichomes of Phormidium form the surface layer (cryo-SEM)
at the internal surface, the trichomes do not glide back out. The colonization of the spheres leads to recognizable stratifications of: 1. trichomes of Phormidiumsp. arranged beneath the surface of the sphere, 2. randomly distributed increasingly dense masses of the diatom Navicula perminotarich in EPS (extracellular polymeric substances) located in the inner part of the sphere (Fig. 1c), and 3. chemoorganotrophic bacteria. During the sphere formation the number of diatoms inside increases continuously. At the initial stages of
colonization of the spheres by trichomes of Phormidium sp. only a few diatoms are present, occasionally reaching 10–20 or more diatoms. At the end of colonization, the space inside the spheres is completely filled by the diatoms. Sometimes cyanobacteria leave the spheres and form hairy outer rims (Fig. 2a). Chemoorganotrophic bacteria and diatoms cannot leave the mature spheres. Our microscopic studies clearly demonstrated the to-and-fro movement of diatoms and bacteria within the spheres, often showing rebounding effects from the envelope without breaking or passing through it. The surface, as well as the inside of a sphere, is always very rich in EPS (Fig. 2b, c). Phormidium sp. trichomes in an advanced stage of sphere formation created a non-
transparent cover. Diatoms inside the sphere are visible only in sections after cutting the sphere. In mature spheres (8–10 weeks of cultivation) the structure of the envelope seems to change, the growth conditions inside the sphere become less favourable and the diatoms abandon this microbial community, leaving the cyanobacterial surface network intact in form and shape. Only the cyanobacterial framework of the spheres remains visible. This phase remains stable for several weeks. Sometimes the number of cyanobacteria increases again, but it never reaches same the level of concentration as in the initial spherical phase. Initiation of sphere formation The activity of the diatoms and the presence of the diatom–cyanobacterial community were obvious. In order to elucidate the potential role of the chemoorganotrophic bacteria in this community, typical bacterial signal substances were tested for their influence on aggregate and sphere formation. In order to potentially enhance the development of the spheres and their maturation, a number of different signal substances were added separately to the media (cAMP, OHHL, BHL). Bacteria are able to sense their environment and react appropriately, communicating with each other. This phenomenon of cell-to-cell signalling is a generic phenomenon described for many bacteria (e.g. Greenberg 1997; Surette et al. 1999). At low cell density, an autoinducer is synthesized at basal levels and this is thought to diffuse into the surrounding media, where it becomes diluted. The autoinducer allows the bacteria to communicate with each other, to sense their own density and, together with a transcriptional activator, to express specific genes as a population instead of individual cells (van Delden and Iglewski 1998). In our experiments, the best results were obtained with BHL. Samples without signal substances had only up to four spheres per plate. After adding BHL the number of the spheres rose as high as 43, with an average of 25 in parallel. Temperatures between 16 and 27C and different light intensities were also tested. These changes in environmental conditions had no influence on the sphere growth. The size of the spheres differed from 0.08 mm up to a maximum of 3 mm. The spheres were also repeatedly observed in fresh samples from the North Sea microbial mats. However, the most important aspect of our work was that were able to reproducibly generate different amounts of spheres in the laboratory and stimulate the number of spheres produced in one experiment by signal substances. Experiments with 1, 5 and 10 mM BHL showed that the probability of sphere appearance significantly increased with the higher concentrations. A similar kind of trichome to trichome communication was observed in Pseudanabaena galeata (Castenholz 1982). He observed the formation of swarms of trichomes, and called them “comets”. Similar to the phenomenon observed there, an elaborate gliding of trichomes
and trichome aggregates also took place in our cultures. In the case of the comets, light triggered complex movements. In the case of cyanobacteria/diatom sphere formation, light has to be excluded as a factor causing movement and complex organization. The origin of the spheres cannot be connected with the existence of air bubbles or agar heterogeneity, as no differences in density inside and outside the sphere within the solid medium were visible. Similar to our spheres, the “comets” formed and maintained their integrity during their movement. The manner in which trichome-to-trichome communication is going on with respect to coordinated movement is, in both cases entirely unknown. Bacterial chemotaxis is a motile response to chemical gradients and may explain many consortia of cyanobacteria and bacteria. However, direct evidence for the role of chemotaxis in forming such consortia is inadequate. Our observations on the formation of spherical aggregates of trichomes of cyanobacteria followed by massive development of diatoms inside the spheres provide evidence for cyanobacterial chemotactic response to bacterial and/or algal products, implying that chemotactic responses between different taxa may play a role in the formation of certain microbial consortia in nature. Further effects such as the waxing and waning of the consortia, maintenance of the principle both in culture and in nature (now observed for over 3 years), and subsequent biomineralization in the form of calcispheres will be followed using new field material as well as the consortia in “pure” culture.
References Budrene EO, Berg HC (1995) Dynamics of formation of symmetrical patterns by chemotactic bacteria. Nature 376:49–53 Castenholz RW (1982) Motility and taxes. In: Carr NG, Whitton BA (eds) The biology of cyanobacteria. Blackwell Scientific, Oxford, pp 413–438 Delden van C, Iglewski BH (1998) Cell-to-cell signalling and Pseudomonas aegurinosa infections. Emerging Infectious Diseases 4:551–560 Gerdes G, Krumbein WE, Reineck H-E (1986) Das Farbstreifensandwatt. In: Gerdes G, Krumbein WE, Reineck H-E (eds) Mellum: Portrait einer Insel. Kramer, Frankfurt, pp 170–187 Greenberg EP (1997) Quorum sensing in gram-negative bacteria. ASM News 63:371–377 Janson S, Carpenter EJ, Bergman B (1999) Fine structure and immunolocalisation of proteins in Aphanizomenon sp. from the Baltic Sea. Eur J Phycol 29:203–211 Kangatharalingam N, Wang L, Priscu JC (1991) Evidence for bacterial chemotaxis to cyanobacteria from a radioassay technique. Appl Environ Microbiol 57:2395–2398 Luft JH (1961) Improvements in epoxy resin embedding methods. J Biophys Biochem Cytol 9:409–414 Rippka R, Deruelles J, Waterbury JB, Herdman M, Stanier RY (1979) Generic assignments, strain histories and properties of pure cultures of cyanobacteria. J Gen Microbiol 111:1–61 Sargent JA (1988) Low temperature scanning electron microscopy: advantages and applications. Scanning Microsc 2:835–849 Surette MG, Miller MB, Bassler BL (1999) Quorum sensing in Escherichia coli, Salmonella typhimurium and Vibrio harveyi: A new family of genes responsible for autoinducer production. Proc Natl Acad Sci USA 96:1639–1644