Development of alternate ssu-rRNA probing strategies for characterizing aquatic microbial communities

Plastids in phytoplankton retain prokaryote-like DNA sequences that may generate false-positive signals from eubacterial small subunit (ssu) rRNA oligonucleotide probes, resulting in the overestimation of bacterial activity in aquatic microbial communities. To assess the extent of possible plastid-associated binding to eubacterial signals, we performed an extensive database search, flask experiments using algal and cyanobacterial pure cultures, and field trials on five common eubacterial probes: S-D-Bact-008-a-A-19, S-D-Bact-338-a-A-18, S-D-Bact-785-a-A-19, S-D-Bact-927-a-A-17, and S-D-Bact-1088-a-A20. The database search and laboratory tests showed significant potential for binding among most bacterial probes and organelle ssu-rRNA. However, we propose two probing strategies to overcome this problem. First, one could use Bact-785 and Bact-338 in tandem, with the plastid component being estimated as the difference between the two signals (Bact-338 has similar to70% overlap with known plastid sequences). Alternately, one might use Bact-338 as the primary eubacterial probe, but then use Cyan-785-a-A-19 (a probe that binds significantly to plastid rRNA) to correct for the plastid-associated false-positive signal. Both strategies would use a eukaryotic probe (S-D-Euca-1379-a-A-16) and Cyan-785-b-A-19 (a probe for most cyanobacteria) to further segregate rRNA signals. Trials were successfully performed using the strategies on samples from a recent field study. (C) 2003 Elsevier B.V. All rights reserved.