A new in situ DNA amplification technique for microscopic detection of bacteria carrying a specific gene is described. Loop-mediated isothermal amplification (LAMP) was used to detect stxA 2 in Escherichia coli O157:H7 cells. The mild permeabilization conditions and low isothermal temperature used in the in situ LAMP method caused less cell damage than in situ PCR. It allowed use of fluorescent antibody labeling in the bacterial mixture after the DNA amplification for identification of E. coli O157:H7 cells with an stxA 2 gene. Higher-contrast images were obtained with this method than with in situ PCR.

An improved in situ hybridization technique, HNPP-FISH, using 2-hydroxy-3-naphthoic acid 2'-phenylanilide phosphate (HNPP) and Fast Red TR was applied to analyse the community structure of planktonic bacteria in river water. Oligonucleotide probes specific for the domain Bacteria (EUB338) and five bacterial groups [Ha wobacterium-Cytophaga; Burkholderia~seudomonas (rRNA Ill)-authentic Alcaligenes; VibriwAeromonas; Pseudomonas (rRNA I); the genus Acinetobacter] were used to investigate the bacterial community structure at two sites differing in organic carbon pollution level. A t the eutrophic site, 54-68% of all cells visualized by staining with DAPl (4',6-diamidino-2-phenylindole) could be detected with probe EUB338. In samples from the oligotrophic site, 3 9 4 5 % of the total cells hybridized with EUB338. At the eutrophic site, approximately 50% of the total cells were identified with the five group-specific probes; the bacterial community structure was dominated by the Flawobacteriu~Cytophaga group and BurkholderiaPseudomonas (rRNA Ill)-authentic Alcaligenes group. A t the oligotrophic site, only 2638% of the total cells were identified with the five group-specific probes. The community structure at the oligotrophic site was similar to that at the eutrophic site, although the percentage of EUB338-detectable cells differed. No appreciable change was found in the community structure during the sampling period at either site. The improved HNPP-FISH technique should be a useful tool for the analysis of microbial community composition.

Lateral gene transfer by phages has contributed significantly to the genetic diversity of bacteria. To accurately determine the frequency and range of phage-mediated gene transfer, it is important to understand the movement of DNA among microbes. Using an in situ DNA amplification technique (cycling primed in situ amplification-fluorescent in situ hybridization; CPRINS-FISH), we examined the propensity for phage-mediated gene transfer in freshwater environments at the single-cell level. Phage P1, T4 and isolated Escherichia coli phage EC10 were used as vectors. All E. coli phages mediated gene transfer from E. coli to both plaque-forming and non-plaque-forming Enterobacteriaceae strains at frequencies of 0.3-8 Â 10À3 per plaque-forming unit (PFU), whereas culture methods using selective agar media could not detect transductants in non-plaque-forming strains. The DNA transfer frequencies through phage EC10 ranged from undetectable to 9 Â 10 À2 per PFU (undetectable to 2 Â 10 À3 per total direct count) when natural bacterial communities were recipients. Direct viable counting combined with CPRINS-FISH revealed that more than 20% of the cells carrying the transferred gene retained their viability in most cases. These results indicate that the exchange of DNA sequences among bacteria occurs frequently and in a wide range of bacteria, and may promote rapid evolution of the prokaryotic genome in freshwater environments.

Bluegill (Lepomis macrochirus) in Lake Biwa, Japan, feed on benthic invertebrates (benthivorous type), aquatic plants (herbivorous type), and zooplankton (planktivorous type). To evaluate the effect of food on intestinal bacterial microbiota, we characterized and compared the intestinal microbiota of these three types of bluegill in terms of community-level physiological profile (CLPP) and genetic structure. The CLPP was analyzed using Biolog MicroPlates (Biolog, Inc., Hayward, CA, USA), and multivariate analysis of variance revealed that the CLPP of intestinal microbiota differed significantly between any pairs of the three types of bluegill. The genetic profiles were analyzed by temperature gradient gel electrophoresis of polymerase chain reaction (PCR)-amplified 16S rDNA fragments, and multidimensional scaling indicated the existence of specific intestinal bacterial structures for both the benthivorous and the planktivorous types. These results suggest that the host's feeding habit can be one factor controlling the intestinal microbiota of fish in the natural environment.