Fluorescence in situ hybridization probes targeting members of the phylum Candidatus Saccharibacteria falsely target Eikelboom type 1851 filaments and other
Tadashi Nittami,1† Lachlan B. M. Speirs,2†
Junji Fukuda,1 Masatoshi Watanabe1 and
Robert J. Seviour2* 1Division of Materials Science and Chemical
Engineering, Faculty of Engineering, Yokohama
National University, Yokohama, Japan. 2Department of Microbiology, La Trobe University,
Bundoora, VIC 3086, Australia.
The FISH probe TM7-305 is thought to target the filamentous Eikelboom morphotype 0041 as a member of the Candidatus ‘Saccharibacteria’ (formerly TM7) phylum. However, with activated sludge samples in both Japan and Australia, this probe hybridized consistently with filamentous bacteria fitting the description of the morphotype 1851, which also responded positively to the CHL1851 FISH probe designed to target Chloroflexi members of this morphotype. 16S rRNA clone libraries from samples containing type 1851 TM7-305-positive filaments yielded Chloroflexi clones with high sequence similarity to Kouleothrix aurantiaca. These contained a variant TM7-305 probe target site possessing weakly destabilizing mismatches insufficient to prevent probe hybridization.
Furthermore, the TM7-905 FISH probe, designed to target members of the entire Candidatus ‘Saccharibacteria’ phylum, also hybridized with the filament morphotypes 0041/0675, which responded also to the phylum level Chloroflexi probes. Many
Chloroflexi sequences have only a single base mismatch to the TM7-905 probe target sequence. When competitor probes for both the TM7-305 and TM7-905
Chloroflexi non-target sites were applied, no fluorescent signal was seen in any of the filamentous organisms also hybridizing with the aforementioned
Chloroflexi probes. These data indicate that these competitor probes must be included in hybridizations when both the TM7-905 and TM7-305 FISH probes are applied, to minimize potential false positive FISH results.
Bulking and foaming, both caused by excessive proliferation of filamentous bacteria, are major global operational problems in activated sludge wastewater treatment plants (WWTPs) (Jenkins et al., 2004; Martins et al., 2004;
Kragelund et al., 2005). Many of these can be identified by applying 16S/23S rRNA targeted fluorescence in situ hybridization (FISH) probes (Daims et al., 2005;
Kragelund et al., 2009), and of these, the TM7-305 and
TM7-905 probes designed to target members of the
Candidatus ‘Saccharibacteria’ (formerly Candidate TM7) phylum (Hugenholtz et al., 2001; Albertsen et al., 2013) have been applied widely in FISH-based surveys (e.g.Thomsen et al., 2002). One filamentous bacterium
Eikelboom morphotype is type 0041, which is often the dominant filament seen in bulking and foaming incidents in treatment plants around the world (Seviour and
Nielsen, 2010), but has never been grown axenically.
Type 0041 is a Gram variable, unbranched, septate and sheathed filament, with abundant attached growth and square/rectangular cells, and is distinguished from the
Eikelboom filament morphotype type 0675 almost entirely on the basis of its larger filament diameter (Jenkins et al., 2004). No 16/23S rRNA sequences have been affiliated with type 0041 filaments, although the available FISH data would seem to suggest that it is probably polyphyletic (Hugenholtz et al., 2001; Thomsen et al., 2002; 2006;
Kragelund et al., 2007), containing members affiliated with each of the Candidatus ‘Saccharibacteria’,
Chloroflexi and Proteobacteria phyla.
When biomass samples from Australian and Japanese treatment plants were subjected to routine FISH analysis (data not shown) using the division level TM7-905 FISH probe (Hugenholtz et al., 2001), filamentous organisms, including the type 0041 morphotype responding to this probe, also hybridized with other FISH probes designed to target members of the Chloroflexi. Similarly, when samples containing the morphotype type 1851 filaments
Received 18 April, 2013; accepted 12 April, 2014. *For correspondence. E-mail firstname.lastname@example.org; Tel. (+61) 2 49844523; Fax (+61) 2 49844523. †These authors contributed equally to the experimentation undertaken. bs_bs_banner
Environmental Microbiology Reports (2014) 6(6), 611–617 doi:10.1111/1758-2229.12172 © 2014 Society for Applied Microbiology and John Wiley & Sons Ltd were subjected to FISH analysis, these hybridized not only with the type 1851 Chloroflexi probe CHL1851 (Beer et al., 2002), but also with the TM7-305 probe designed by
Hugenholtz and colleagues (2001). Such outcomes suggest that applying these FISH probes may lead to an overestimation in activated sludge samples of the relative abundances of the filaments against which it was originally designed. In this report, we attempt to understand the reasons for this and suggest how they may be prevented experimentally. This information is considered important as Saccharibacteria appear widespread in nature, found in other environments including soils, human gut, freshwater, marine and coral ecosystems, and are often identified and studied there using in situ hybridization techniques (e.g. Brinig et al., 2003; Ferrari et al., 2005; Kuehbacher et al., 2008; Dinis et al., 2011).
Results and discussion
Preliminary FISH observations
FISH-based filament surveys were performed on paraformaldehyde-fixed activated sludge samples collected from 35 municipal WWTPs in Australia and Japan (Supporting Information Table S1). Many of these contained high abundances of the Eikelboom filament morphotype 1851, differentiated microscopically from type 0041, with which it shares some features in common, by its distinctive long, thin slightly bent sheathed filaments, which are frequently arranged in characteristic bundles (Jenkins et al., 2004). These filaments responded, as expected to the GNSB941 (Gich et al., 2001), CFX1223 (Björnsson et al., 2002), CFX109 (Björnsson et al., 2002) and CHL1851 FISH probes designed to target most Chloroflexi, subgroup 1