In summary, FISH probes represent tools that are rather tedious to construct but easy to apply. The specificity of oligonucleotides for a particular range of phylotypes is deduced from the sequence data that are available at the time of their design. Due to the exponential growth of sequence databases, some probes may lose their originally envisaged specificity or target group coverage with time. For example, one commonly used probe for Bacteria (EUB338) has been constructed on a set of 6). At the time of this review, the Ribosomal Database Project (42) has collected >170,000 full and partial bacterial sequences. Thus, it is not surprising that the coverage of probe EUB338 is incomplete, and it has been modified accordingly (51). This points to the need to check the target range of existing FISH probes on the latest data set before applying them to unknown samples. A useful database about the current specificity and coverage of many published probes is provided by the Molecular Ecology department of the University of Vienna at http://www.microbial-ecology.de/probebase (162).
However, the replacement of some of the first generation of available FISH probes should be seen as a necessary optimization process in a rapidly progressing field rather than as an irresolvable problem of the FISH approach. Analyses of rRNA gene sequences from aquatic habitats indicate that only a limited number of well-defined phylogenetic clades of microorganisms might actually be common in the pelagic zones of marine and freshwater environments (55, 85, 91, 121, 240, 328). A substantial amount of diversity within several of these clades appears to be covered adequately by the presently available sequence data (108). Such knowledge will eventually provide a reliable base for a new generation of more “habitat-specific” FISH probes that discriminate well-established lineages of microbes in a particular environment (Table 1), e.g., the various marine SAR clades (62, 188, 189, 325) or the freshwater actinobacteria (91, 310).
Another drawback of FISH with fluorescently monolabeled oligonucleotide probes is the low fluorescence intensity of hybridized bacteria from natural water samples (214). Bacteria in oligotrophic water are often small, slow growing, or in stationary phase (187), and their ribosome content is typically low (65). Consequently, there are few rRNA target molecules for FISH staining. The fraction of microbial cells that can be visualized microscopically may thus vary with the physiological state of the studied assemblage. For example, a significantly smaller percentage of bacteria could be stained by FISH in coastal North Sea surface waters during the winter months than during the spring and summer seasons (63). In environments such as offshore marine waters, sometimes only a minor fraction of microbes can be visualized by FISH with fluorescently monolabeled oligonucleotide probes (214). Therefore, it is likely that the abundances of some slowly-growing microorganisms with small cell sizes, e.g., of members of the marine SAR86 clade, are underestimated by the standard FISH approach (210).
During the last decade, efforts have been made to increase the sensitivity of FISH, e.g., with peptide nucleic acid probes (320), brighter fluorochromes (88), image intensified video microscopy (78), preincubation with chloramphenicol (200), hybridization with more than one fluorescently labeled oligonucleotide probe (188), and helper probes (70). Two particularly promising alternative strategies to FISH with fluorescently monolabeled oligonucleotides are polynucleotide probes and enzymatic signal amplification (57, 149, 210). Fluorescently multilabeled rRNA-targeted polyribonucleotide probes yield significantly higher signal intensities than oligonucleotide probes (214). They have been successfully applied to discriminate between bacteria and different groups of archaea in coastal and open ocean environments (40, 136, 214). Limitations of polyribonucleotide probes as a routine tool for the identification of aquatic microbes are the relatively high effort and variability of enzymatic probe synthesis, and the rather low phylogenetic resolution. Only three large taxa have been distinguished by this technique in marine waters (57), although potentially a resolution at the genus level is possible (300).
A working alternative is catalyzed reporter deposition (CARD)-FISH (210, 211, 263), which combines in situ hybridization with horseradish peroxidase labeled oligonucleotide probes and enzymatic signal amplification with fluorescently labeled tyramides (24). This allows the quantitative detection of marine and freshwater pelagic bacteria with low ribosome content that are insufficiently visualized by fluorescently monolabeled probes (210, 263) (Fig. 4).
Answers to all your Biology Questions
