Modes and Specificity of Interaction between the Bacterial Partners

Theoretically, phototrophic consortia could form randomly from bacterial cells which encounter each other just by chance. Under these conditions, morphologically identical consortia would be expected to harbor different types of green sulfur bacteria and, secondly, free-living epibiont cells should also be detected in the habitat of phototrophic consortia. However, when the epibionts associated with a particular phototrophic consortium were analyzed, all epibi-ont cells invariably contained the same 16S rRNA gene sequence. Evidently, phototrophic consortia with the same morphology that share the same habitat contain only one single type of epibiont (Glaeser and Overmann 2004). Furthermore, none of the 16S rRNA gene sequences of epibionts has so far been detected in free-living green sulfur bacteria, suggesting that the bacterial interaction within phototrophic consortia is highly specific, and that epibionts have adapted to a life in association.

In line with the results of the phylogenetic analyses, the epibiont numbers per consortium show a nonrandom frequency distribution in consortia from natural samples, as well as from enrichment cultures. Thus, most consortia of one particular type contain the same number of epibionts (Overmann et al. 1998). Consortia multiply as a whole. After a simultaneous doubling of epibi-onts and the central bacterium, the elongated consortium divides by a transverse constriction through the aggregate. Evidently, the cell division of all epibiont cells proceeds in a highly synchronized fashion and parallel to that of the central bacterium.

Several studies provide additional insight into the close interaction between the two partner bacteria. Evidence has accumulated that the two unrelated bacteria interact in a unique way with respect to signal transduction. When a microspectrum of light is projected onto a suspension of "Chlorochromatium aggregatum", the consortia accumulate at a wavelength of 740 nm. Since this value matches the position of the long wavelength absorption of bacterio-chlorophyll c in vivo, the antenna pigments are most likely the photoreceptors of the scotophobic response (Frostl and Overmann 1998). On the other hand, electron microscopy clearly demonstrated that the central bacterium is mo-nopolarly monotrichously flagellated, whereas the green sulfur bacterial epibi-onts are nonmotile (Glaeser and Overmann 2003b). Taken together, these experimental results clearly suggest that a rapid interspecies signal transfer occurs between the nonmotile, phototrophic component and the motile, colorless central bacterium. So far, this type of signal transfer between unrelated bacteria is unique in the microbial world.

Intact consortia exhibit chemotactic behavior toward sulfide (Frostl and Overmann 1998; Glaeser and Overmann 2003b, 2004). Since sulfide is utilized by the epibiont, a rapid signal exchange between the epibiont and the central motile bacterium could possibly occur also during chemotaxis, yet has to be confirmed experimentally.

Physiological experiments have provided additional information on a mutual signal exchange between the two bacterial partners. When a natural population of "Pelochromatium roseum" was incubated with radioactively labeled 2-oxoglutarate and the incorporation of radiolabel followed by microautoradiography of the samples, 87.5% of the consortia incorporated 2-oxoglutarate when both light and sulfide were present, whereas uptake was detected in less than 1.4% of the consortia if either light or sulfide were absent (Glaeser and Overmann 2003b). Because the epibionts in this population of "Pelochromatium roseum" had been shown to grow photoautotrophically, the metabolic state of the green sulfur bacterial epibiont seems to regulate the incorporation of 2-oxoglutarate by the central bacterium. Evidently, interaction between the two partner bacteria in phototrophic consortia is not limited to a rapid inter-species signal exchange during tactic behavior, but also occurs during the regulation of central metabolic processes. It may be speculated that such a mechanism serves to synchronize the growth and multiplication of the two partner bacteria.

It is to be expected that the tightly packed layer of epibionts impedes the diffusion of essential substrates and nutrients to the embedded central bacterium. The diffusion coefficient in cytoplasma is moderately decreased (by a mean factor of 0.3; Koch 1996) as compared to typical aqueous environments. Accordingly, the epibiont may not only regulate the uptake of exogenous carbon compounds by the central bacterium, but may actually supply the central bacterium directly with certain carbon substrates. Such an exchange of organic carbon compounds, vitamins and chelators has been suggested for other consortia, like the associations of heterotrophic bacteria with filamentous cyano-bacteria (Paerl and Pinckney 1996). Free-living Chlorobiaceae have been shown to excrete up to 30% of the photosynthetically fixed carbon (Czeczuga and Gradski 1972), mainly as 2-oxo-3-methylvalerate and 2-oxoglutarate (Sirevag and Ormerod 1970). The latter compound is a direct intermediate of the reverse tricarboxylic acid cycle which is employed by green sulfur bacteria for CO2 fixation (Evans et al. 1966; Sirevag and Ormerod 1970). Similar to other green sulfur bacteria, the epibionts may thus excrete 2-oxo acids, which could then be utilized as carbon substrates by the central bacterium.

A typical syntrophic sulfur cycle is less likely to occur within phototro-phic consortia (see above). Theoretically, interspecies hydrogen transfer may occur between the central bacterium and the epibionts in a manner similar to the aggregates of fermentative bacteria with methanogenic archaea or sul-fate-reducing bacteria in activated sludge. Based on typical metabolic rates for bacteria, it has been calculated that the single layer of epibionts is not sufficient to shield the central bacterium from high concentrations of exogenous, easily diffusible compounds, such as hydrogen gas, oxygen or sulfide (Overmann 2001a).

0 0

Post a comment