Environmental Microbiology (2005) 7(3), 419–433 doi:10.1111/j.1462-2920.2004.00711.x © 2004 Society for Applied Microbiology and Blackwell Publishing Ltd Blackwell Science, LtdOxford, UKEMIEnvironmental Microbiology 1462-2912Society for Applied Microbiology and Blackwell Publishing Ltd, 20047 3419433Original ArticleBiogeography of marine sponge-associated bacteriaM. W. Taylor et al. Received 18 March, 2004; revised 28 July, 2004; accepted 28 July, 2004. *For correspondence. E-mail taylor@microbial-ecology.net; Tel. (+43) 1 4277 54207; Fax (+43) 1 4277 54389. † Present address: Lehrstuhl für Mikrobielle Ökologie, Institut für Ökologie und Naturs- chutz (IECB), Universität Wien, Althanstr. 14, A-1090 Wien, Austria. Biogeography of bacteria associated with the marine sponge Cymbastela concentrica Michael W. Taylor, 1 * † Peter J. Schupp, 2,3 Rocky de Nys, 4 Staffan Kjelleberg 2,5 and Peter D. Steinberg 1,2 1 School of Biological, Earth and Environmental Sciences, 2 Centre for Marine Biofouling and Bio-Innovation, University of New South Wales, Sydney, NSW 2052, Australia. 3 University of Guam Marine Laboratory, UOG Station, Mangilao, Guam 96923, USA. 4 School of Marine Biology and Aquaculture, James Cook University, Townsville, QLD 4811, Australia. 5 School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, NSW 2052, Australia. Summary Recent debate regarding microbial biogeography has focused largely on free-living microbes, yet those microbes associated with host organisms are also of interest from a biogeographical perspective. Marine eukaryotes and associated bacteria should provide ideal systems in which to consider microbial bioge- ography, as (i) bacteria in seawater should be able to disperse among individuals of the same host species, yet (ii) potential for adaptation to particular hosts (and thus speciation) also exists. We used 16S rDNA-DGGE (denaturing gradient gel electrophoresis) to examine geographic variability in bacterial community compo- sition in the marine sponge Cymbastela concentrica. Denaturing gradient gel electrophoresis banding patterns (and phylogenetic analysis of excised DGGE bands) indicated different communities in Cymbastela concentrica from tropical versus temper- ate Australia. In contrast, communities were very sim- ilar over a 500-km portion of the sponge’s temperate range. Variation in bacterial community composition was also considered with respect to ocean current patterns. We speculate that the divergent communi- ties in different parts of the sponge’s range provide evidence of endemism attributed to host association, although variation in environmental factors such as light and temperature could also explain the observed results. Interestingly, bacterial communities in seawa- ter varied much less between tropical and temperate locations than did those in C. concentrica, supporting the concept of widespread dispersal among these free-living microbes. Introduction The distribution of organisms is a fundamental issue in ecology, with implications for key biological indices such as species diversity. In the case of free-living (i.e. not host- associated) protozoans, for example, there is consider- able evidence for their ubiquitous (worldwide) dispersal (Fenchel, 1993; Finlay and Clarke, 1999; Finlay, 2002), which in turn is thought to limit their global diversity. It has been argued that the sheer abundance of these eukary- otic microbes drives their dispersal. Given the even higher abundance of bacteria, then prokaryotic microbes are even more likely to be distributed globally (Finlay and Clarke, 1999). The assumed result of this is an absence of endemism and, consequently, a relatively low global prokaryotic diversity, at least among free-living species. Such logic suggests that uniformity of microbial commu- nities should be most likely in a contiguous, relatively homogeneous environment such as the ocean. Indeed, closely related microbes have been found in widely sep- arated oceanic areas (Mullins et al., 1995; Hagström et al., 2000), and it has been suggested that marine microbial diversity is relatively low (Hagström et al., 2002). In contrast, several recent terrestrial studies have indicated that endemism does exist among bacteria (Fulthorpe et al., 1998; Cho and Tiedje, 2000; Papke et al., 2003) and archaea (Whitaker et al., 2003), but that methods offering high levels of genetic resolution are required for its detection. The nature of the habitat of different bacteria (e.g. free- living versus host-associated) is an important factor in this debate. Many bacteria live on or within eukaryotic hosts (e.g. as symbionts or parasites), and the potentially impor- tant contribution of these organisms to marine microbial diversity was recently highlighted (Taylor et al., 2004). Host organisms such as marine sponges, corals, bryozo- ans and algae provide a unique set of environmental conditions for microbial colonization, quite different from