Terminal restriction fragment length polymorphism analysis for human fecal microbiota and its application for analysis of complex bifidobacterial communities

Terminal restriction fragment length polymorphism (T-RFLP) analysis was used to characterize and compare human fecal microbiota among individuals. T-RFLP patterns of fecal 16S ribosomal DNA (rDNA) PCR products from three adults revealed host-specific bacterial communities and were in good agreement with those reported in our previous study. In addition, we applied T-RFLP analysis for the analysis of complex bifidobacterial communities in human fecal samples. The developed method based on Bifidobacterium genus-specific PCR and T-RFLP could identify more than one bifidobacterial species. T-RFLP patterns of Bifidobacterium genus-specific PCR products from the fecal samples were host-specific as well as those of fecal 16S rDNA PCR products. These results were confirmed by PCR-denaturing gradient gel electrophoresis (DGGE) with primers specific for the genus Bifidobacterium and Bifidobacterium species- and group-specific PCR. Our study demonstrates that T-RFLP analysis is useful for assessment of the diversity of the human fecal microbiota and rapid comparison of the community structure among individuals, and that the applied method is useful for rapid and sensitive analysis of bifidobacterial community.     

Advances in the use of terminal restriction fragment length polymorphism (T-RFLP) analysis of 16S rRNA genes to characterize microbial communities

Terminal restriction fragment length polymorphism (T-RFLP) analysis is a popular high-throughput fingerprinting technique used to monitor changes in the structure and composition of microbial communities. This approach is widely used because it offers a compromise between the information gained and labor intensity. In this review, we discuss the progress made in T-RFLP analysis of 16S rRNA genes and functional genes over the last 10 years and evaluate the performance of this technique when used in conjunction with different statistical methods. Web-based tools designed to perform virtual polymerase chain reaction and restriction enzyme digests greatly facilitate the choice of primers and restriction enzymes for T-RFLP analysis. Significant improvements have also been made in the statistical analysis of T-RFLP profiles such as the introduction of objective procedures to distinguish between signal and noise, the alignment of T-RFLP peaks between profiles, and the use of multivariate statistical methods to detect changes in the structure and composition of microbial communities due to spatial and temporal variation or treatment effects. The progress made in T-RFLP analysis of 16S rRNA and genes allows researchers to make methodological and statistical choices appropriate for the hypotheses of their studies.     

Relative diversity and community structure analysis of rumen protozoa according to T-RFLP and microscopic methods

Protozoa are common inhabitants of the rumen where they play roles in host nutrition and methanogenesis. Knowledge of how changes in the composition of protozoa communities affect these processes is limited in part due to a lack of efficient methods for protozoa community analysis. In this study, a terminal-restriction fragment length polymorphism (T-RFLP) assay targeting the 18S rRNA gene was developed for comparative analysis of rumen protozoa communities. Comparison of diversity and structure of protozoa communities from hay-fed versus silage/grain-fed cattle via T-RFLP analysis yielded similar overall results to microscopy analysis. According to both methods, Entodinium spp. were more abundant in the silage/grain-fed cattle and protozoa diversity (as calculated using the Shannon index) was higher for the hay-fed cattle due to greater species evenness. Type B protozoa were more prevalent in the hay-fed cattle, whereas Type A protozoa were more prevalent in the silage/grain-fed cattle. Analysis of similarity (ANOSIM) indicated that the protozoa communities from hay-fed and silage/grain-fed cattle were different, and multivariate analysis indicated that pen mates (i.e., cattle fed the same diet and housed together) tended to have similar protozoa communities types. In summary, we present a T-RFLP method for analyzing rumen protozoa communities which complements traditional microscopy approaches but has the advantage of being amenable to high-throughput.     

New threshold and confidence estimates for terminal restriction fragment length polymorphism analysis of complex bacterial communities

Terminal restriction fragment length polymorphism (T-RFLP) analysis has the potential to be useful for comparisons of complex bacterial communities, especially to detect changes in community structure in response to different variables. To do this successfully, systematic variations have to be detected above method-associated noise, by standardizing data sets and assigning confidence estimates to relationships detected. We investigated the use of different standardizing methods in T-RFLP analysis of PCR-amplified 16S rRNA genes to elucidate the similarities between the bacterial communities in 17 soil and sediment samples. We developed a robust method for standardizing data sets that appeared to allow detection of similarities between complex bacterial communities. We term this the variable percentage threshold method. We found that making conclusions about the similarities of complex bacterial communities from T-RFLP profiles generated by a single restriction enzyme (RE) may lead to erroneous conclusions. Instead, the use of multiple REs, each individually, to generate multiple data sets allowed us to determine a confidence estimate for groupings of apparently similar communities and at the same time minimized the effects of RE selection. In conjunction with the variable percentage threshold method, this allowed us to make confident conclusions about the similarities of the complex bacterial communities in the 17 different samples.     

Analysis of T-RFLP data using analysis of variance and ordination methods: a comparative study

The analysis of T-RFLP data has developed considerably over the last decade, but there remains a lack of consensus about which statistical analyses offer the best means for finding trends in these data. In this study, we empirically tested and theoretically compared ten diverse T-RFLP datasets derived from soil microbial communities using the more common ordination methods in the literature: principal component analysis (PCA), nonmetric multidimensional scaling (NMS) with Sørensen, Jaccard and Euclidean distance measures, correspondence analysis (CA), detrended correspondence analysis (DCA) and a technique new to T-RFLP data analysis, the Additive Main Effects and Multiplicative Interaction (AMMI) model. Our objectives were i) to determine the distribution of variation in T-RFLP datasets using analysis of variance (ANOVA), ii) to determine the more robust and informative multivariate ordination methods for analyzing T-RFLP data, and iii) to compare the methods based on theoretical considerations. For the 10 datasets examined in this study, ANOVA revealed that the variation from Environment main effects was always small, variation from T-RFs main effects was large, and variation from T-RF × Environment (T × E) interactions was intermediate. Larger variation due to T × E indicated larger differences in microbial communities between environments/treatments and thus demonstrated the utility of ANOVA to provide an objective assessment of community dissimilarity. The comparison of statistical methods typically yielded similar empirical results. AMMI, T-RF-centered PCA, and DCA were the most robust methods in terms of producing ordinations that consistently reached a consensus with other methods. In datasets with high sample heterogeneity, NMS analyses with Sørensen and Jaccard distance were the most sensitive for recovery of complex gradients. The theoretical comparison showed that some methods hold distinct advantages for T-RFLP analysis, such as estimations of variation captured, realistic or minimal assumptions about the data, reduced weight placed on rare T-RFs, and uniqueness of solutions. Our results lead us to recommend that method selection be guided by T-RFLP dataset complexity and the outlined theoretical criteria. Finally, we recommend using binary or relativized peak height data with soil-based T-RFLP data for ordination-based exploratory microbial analyses.     

Application of a new PCR primer for terminal restriction fragment length polymorphism analysis of the bacterial communities in plant roots

Contamination with plastid small subunit (SSU) rDNA is a major drawback when analyzing the bacterial communities of plant roots using culture-independent methods. In this study, a polymerase chain reaction (PCR) primer, 783r, was designed and tested to specifically amplify the SSU rDNA of various bacterial species without amplifying the SSU rDNA of plant plastids. To confirm how useful the community analysis of rhizobacteria is using 783r, the terminal restriction fragment length polymorphism (T-RFLP) method was performed with wheat (Triticum aestivum) and spinach (Spinacea oleracea) root samples. Using the standard T-RFLP method, a large T-RF peak of plant plastid SSU rDNA interfered with the bacterial community analysis. In contrast, the T-RFLP method using the 783r primer was able to detect the bacterial DNA while directly eliminating the influence of the plant-derived DNA extracted from the plant roots. Primer 783r might, therefore, be a useful PCR primer for the culture-independent analysis of bacterial communities in plant roots using SSU rDNA.     

Optimization of terminal-restriction fragment length polymorphism analysis for complex marine bacterioplankton communities and comparison with denaturing gradient gel electrophoresis.

The potential of terminal-restriction fragment length polymorphism (T-RFLP) and the detection of operational taxonomic units (OTUs) by capillary electrophoresis (CE) to characterize marine bacterioplankton communities was compared with that of denaturing gradient gel electrophoresis (DGGE). A protocol has been developed to optimize the separation and detection of OTUs between 20 and 1, 632 bp by using CE and laser-induced fluorescence detection. Additionally, we compared T-RFLP fingerprinting to DGGE optimized for detection of less abundant OTUs. Similar results were obtained with both fingerprinting techniques, although the T-RFLP approach and CE detection of OTUs was more sensitive, as indicated by the higher number of OTUs detected. We tested the T-RFLP fingerprinting technique on complex marine bacterial communities by using the 16S rRNA gene and 16S rRNA as templates for PCR. Samples from the Northern and Middle Adriatic Sea and from the South and North Aegean Sea were compared. Distinct clusters were identifiable for different sampling sites. Thus, this technique is useful for rapid evaluation of the biogeographical distribution and relationships of bacterioplankton communities.     

Characterization of microbial communities found in the human vagina by analysis of terminal restriction fragment length polymorphisms of 16S rRNA genes

To define and monitor the structure of microbial communities found in the human vagina, a cultivation-independent approach based on analyses of terminal restriction fragment length polymorphisms (T-RFLP) of 16S rRNA genes was developed and validated. Sixteen bacterial strains commonly found in the human vagina were used to construct model communities that were subsequently used to develop efficient means for the isolation of genomic DNA and an optimal strategy for T-RFLP analyses. The various genera in the model community could best be resolved by digesting amplicons made using bacterial primers 8f and 926r with HaeIII; fewer strains could be resolved using other primer-enzyme combinations, and no combination successfully distinguished certain species of the same genus. To demonstrate the utility of the approach, samples from five women that had been collected over a 2-month period were analyzed. Differences and similarities among the vaginal microbial communities of the women were readily apparent. The T-RFLP data suggest that the communities of three women were dominated by a single phylotype, most likely species of Lactobacillus. In contrast, the communities of two other women included numerically abundant populations that differed from Lactobacillus strains whose 16S rRNA genes had been previously determined. The T-RFLP profiles of samples from all the women were largely invariant over time, indicating that the kinds and abundances of the numerically dominant populations were relatively stable throughout two menstrual cycles. These findings show that T-RFLP of 16S rRNA genes can be used to compare vaginal microbial communities and gain information about the numerically dominant populations that are present.     

Reliability for detecting composition and changes of microbial communities by T-RFLP genetic profiling

Terminal restriction fragment length polymorphism (T-RFLP) analysis is commonly used for profiling microbial communities in various environments. However, it may suffer from biases during the analytic process. This study addressed the potential of T-RFLP profiles (1) to reflect real community structures and diversities, as well as (2) to reliably detect changing components of microbial community structures. For this purpose, defined artificial communities of 30 SSU rRNA gene clones, derived from nine bacterial phyla, were used. PCR amplification efficiency was one primary bias with a maximum variability factor of 3.5 among clones. PCR downstream analyses such as enzymatic restriction and capillary electrophoresis introduced a maximum bias factor of 4 to terminal restriction fragment (T-RF) signal intensities, resulting in a total maximum bias factor of 14 in the final T-RFLP profiles. In addition, the quotient between amplification efficiency and T-RF size allowed predicting T-RF abundances in the profiles with high accuracy. Although these biases impaired detection of real community structures, the relative changes in structures and diversities were reliably reflected in the T-RFLP profiles. These data support the suitability of T-RFLP profiling for monitoring effects on microbial communities.     

Community structure analyses are more sensitive to differences in soil bacterial communities than anonymous diversity indices

Changes in the diversity and structure of soil microbial communities may offer a key to understanding the impact of environmental factors on soil quality in agriculturally managed systems. Twenty-five years of biodynamic, bio-organic, or conventional management in the DOK long-term experiment in Switzerland significantly altered soil bacterial community structures, as assessed by terminal restriction fragment length polymorphism (T-RFLP) analysis. To evaluate these results, the relation between bacterial diversity and bacterial community structures and their discrimination potential were investigated by sequence and T-RFLP analyses of 1,904 bacterial 16S rRNA gene clones derived from the DOK soils. Standard anonymous diversity indices such as Shannon, Chao1, and ACE or rarefaction analysis did not allow detection of management-dependent influences on the soil bacterial community. Bacterial community structures determined by sequence and T-RFLP analyses of the three gene libraries substantiated changes previously observed by soil bacterial community level T-RFLP profiling. This supported the value of high-throughput monitoring tools such as T-RFLP analysis for assessment of differences in soil microbial communities. The gene library approach also allowed identification of potential management-specific indicator taxa, which were derived from nine different bacterial phyla. These results clearly demonstrate the advantages of community structure analyses over those based on anonymous diversity indices when analyzing complex soil microbial communities.     

Community Structure Analyses Are More Sensitive to Differences in Soil Bacterial Communities than Anonymous Diversity Indices

Changes in the diversity and structure of soil microbial communities may offer a key to understanding the impact of environmental factors on soil quality in agriculturally managed systems. Twenty-five years of biodynamic, bio-organic, or conventional management in the DOK long-term experiment in Switzerland significantly altered soil bacterial community structures, as assessed by terminal restriction fragment length polymorphism (T-RFLP) analysis. To evaluate these results, the relation between bacterial diversity and bacterial community structures and their discrimination potential were investigated by sequence and T-RFLP analyses of 1,904 bacterial 16S rRNA gene clones derived from the DOK soils. Standard anonymous diversity indices such as Shannon, Chao1, and ACE or rarefaction analysis did not allow detection of management-dependent influences on the soil bacterial community. Bacterial community structures determined by sequence and T-RFLP analyses of the three gene libraries substantiated changes previously observed by soil bacterial community level T-RFLP profiling. This supported the value of high-throughput monitoring tools such as T-RFLP analysis for assessment of differences in soil microbial communities. The gene library approach also allowed identification of potential management-specific indicator taxa, which were derived from nine different bacterial phyla. These results clearly demonstrate the advantages of community structure analyses over those based on anonymous diversity indices when analyzing complex soil microbial communities.     

Identification of major subgroups of ammonia-oxidizing bacteria in environmental samples by T-RFLP analysis of amoA PCR products

A cloning-independent method based on T-RFLP (terminal restriction fragment length polymorphism) analysis of amoA PCR products was developed to identify major subgroups of autotrophic ammonia oxidizers of the beta-subclass of the class Proteobacteria in total community DNA. Based on a database of 28 partial gene sequences encoding the active-site polypeptide of ammonia monooxygenase (amoA), defined lengths of terminal restriction fragments (= operational taxonomic units, OTUs) of amoA were predicted to correlate in TaqI-based T-RFLP analysis with phylogenetically defined subgroups of ammonia oxidizers. Members of the genus Nitrosospira showed a specific OTU of 283 bp in length, while a fragment size of 219 bp was indicative of Nitrosomonas-like sequence types including N. europaea, N. eutropha, and N. halophila. Two amoA sequence clusters designated previously as the lineages 'PluBsee' and 'Sch?hsee' [Rotthauwe, J.-H., Witzel, K.-P., Liesack, W., 1997. Appl. Environ. Microbiol. 63, 4704-4712] shared a TaqI-based OTU with a fragment size of 48 bp, but sequence types of these two lineages could be differentiated by AluI-based T-RFLP analysis. A survey of various environmental samples and enrichment cultures by T-RFLP analysis and by comparative analysis of cloned amoA sequences confirmed the predicted correlations between distinct OTUs and phylogenetic information. Our data suggest that amoA-based T-RFLP analysis is a reliable tool to rapidly assess the complexity of ammonia-oxidizing communities in environmental samples with respect to the presence of major subgroups, i.e. nitrosospiras versus nitrosomonads.     

Comparison of molecular fingerprinting methods for analysis of soil microbial community structure

Denaturing gradient gel electrophoresis (DGGE), terminal-restriction fragment length polymorphism (T-RFLP) analysis, and automated ribosomal intergenic spacer analysis (ARISA) have been widely used as molecular fingerprinting methods for analysis of microbial communities. To find suitable methods, we compared the three fingerprinting methods by analyzing soil fungal communities in four differing land-use types: bare ground, crop fields, grasslands, and forests. We also examined optimal primer pairs for DGGE analysis by comparing single and mixed DNA samples of cultured fungal populations. Principal coordinate analysis (PCO), nonmetric multidimensional scaling method (NMDS), and analysis of similarities (ANOSIM), which are major multivariate statistical analyses for quantifying fingerprint patterns, were compared. All three fingerprinting methods yielded clear discrimination of soil fungal communities among the four land-use types, irrespective of statistical methods. The advantages and disadvantages of the three fingerprinting methods were discussed.     

Statistical methods for characterizing diversity of microbial communities by analysis of terminal restriction fragment length polymorphisms of 16S rRNA genes

The analysis of terminal restriction fragment length polymorphisms (T-RFLP) of 16S rRNA genes has proven to be a facile means to compare microbial communities and presumptively identify abundant members. The method provides data that can be used to compare different communities based on similarity or distance measures. Once communities have been clustered into groups, clone libraries can be prepared from sample(s) that are representative of each group in order to determine the phylogeny of the numerically abundant populations in a community. In this paper methods are introduced for the statistical analysis of T-RFLP data that include objective methods for (i) determining a baseline so that 'true' peaks in electropherograms can be identified; (ii) a means to compare electropherograms and bin fragments of similar size; (iii) clustering algorithms that can be used to identify communities that are similar to one another; and (iv) a means to select samples that are representative of a cluster that can be used to construct 16S rRNA gene clone libraries. The methods for data analysis were tested using simulated data with assumptions and parameters that corresponded to actual data. The simulation results demonstrated the usefulness of these methods in their ability to recover the true microbial community structure generated under the assumptions made. Software for implementing these methods is available at http://www.ibest.uidaho.edu/tools/trflp_stats/index.php.     

New Threshold and Confidence Estimates for Terminal Restriction Fragment Length Polymorphism Analysis of Complex Bacterial Communities

Terminal restriction fragment length polymorphism (T-RFLP) analysis has the potential to be useful for comparisons of complex bacterial communities, especially to detect changes in community structure in response to different variables. To do this successfully, systematic variations have to be detected above method-associated noise, by standardizing data sets and assigning confidence estimates to relationships detected. We investigated the use of different standardizing methods in T-RFLP analysis of PCR-amplified 16S rRNA genes to elucidate the similarities between the bacterial communities in 17 soil and sediment samples. We developed a robust method for standardizing data sets that appeared to allow detection of similarities between complex bacterial communities. We term this the variable percentage threshold method. We found that making conclusions about the similarities of complex bacterial communities from T-RFLP profiles generated by a single restriction enzyme (RE) may lead to erroneous conclusions. Instead, the use of multiple REs, each individually, to generate multiple data sets allowed us to determine a confidence estimate for groupings of apparently similar communities and at the same time minimized the effects of RE selection. In conjunction with the variable percentage threshold method, this allowed us to make confident conclusions about the similarities of the complex bacterial communities in the 17 different samples.     

Novel phylogenetic assignment database for terminal-restriction fragment length polymorphism analysis of human colonic microbiota

Various molecular-biological approaches using the 16S rRNA gene sequence have been used for the analysis of human colonic microbiota. Terminal- restriction fragment length polymorphism (T-RFLP) analysis is suitable for a rapid comparison of complex bacterial communities. Terminal-restriction fragment (T-RF) length can be calculated from a known sequence, thus one can predict bacterial species on the basis of their T-RF length by this analysis. The aim of this study was to build a phylogenetic assignment database for T-RFLP analysis of human colonic microbiota (PAD-HCM), and to demonstrate the effectiveness of PAD-HCM compared with the results of 16S rRNA gene clone library analysis. PAD-HCM was completed to include 342 sequence data obtained using four restriction enzymes. Approximately 80% of the total clones detected by 16S rRNA gene clone library analysis were the same bacterial species or phylotypes as those assigned from T-RF using PAD-HCM. Moreover, large T-RFs consisted of common species or phylotypes detected by both analytical methods. All pseudo-T-RFs identified by mung bean nuclease digestion could not be assigned to a bacterial species or phylotype, and this finding shows that pseudo-T-RFs can also be predicted using PAD-HCM. We conclude that PAD-HCM built in this study enables the prediction of T-RFs at the species level including difficult-to-culture bacteria, and that it is very useful for the T-RFLP analysis of human colonic microbiota.     

Molecular analysis of bacterial communities associated with the roots of Douglas fir (Pseudotsuga menziesii) colonized by different ectomycorrhizal fungi

We studied the effect of ectomycorrhizal fungi on bacterial communities colonizing roots of Douglas fir (Pseudotsuga menziesii). Mycorrhizal tips were cleaned of soil and separated based on gross morphological characteristics. Sequencing of the internal transcribed spacers of the nuclear rRNA gene cluster indicated that the majority of the tips were colonized by fungi in the Russulaceae, with the genera Russula and Lactarius comprising 70% of the tips. Because coamplification of organellar 16S rRNA genes can interfere with bacterial community analysis of root tips, we developed and tested a new primer pair that permits amplification of bacterial 16S rRNA genes but discriminates more effectively against organellar sequences than commonly used bacterial primer sets. We then used terminal restriction fragment length polymorphism (T-RFLP) and sequence analysis of the 16S rRNA gene to examine differences in bacterial communities associated with the mycorrhizal tips. Cluster analysis of T-RFLP profiles indicated that there were different bacterial communities among the root tips; however, the communities did not seem to be affected by the taxonomic identity of the ectomycorrhizal fungi. Terminal restriction fragment profiling and sequencing of cloned partial 16S rRNA genes indicated that most bacteria on the ectomycorrhizal tips were related to the Alphaproteobacteria and the Bacteroidetes group.     

Links between plant and rhizoplane bacterial communities in grassland soils, characterized using molecular techniques

Molecular analysis of grassland rhizosphere soil has demonstrated complex and diverse bacterial communities, with resultant difficulties in detecting links between plant and bacterial communities. These studies have, however, analyzed "bulk" rhizosphere soil, rather than rhizoplane communities, which interact most closely with plants through utilization of root exudates. The aim of this study was to test the hypothesis that plant species was a major driver for bacterial rhizoplane community composition on individual plant roots. DNA extracted from individual roots was used to determine plant identity, by analysis of the plastid tRNA leucine (trnL) UAA gene intron, and plant-related bacterial communities. Bacterial communities were characterized by analysis of PCR-amplified 16S rRNA genes using two fingerprinting methods: terminal restriction fragment length polymorphisms (T-RFLP) and denaturing gradient gel electrophoresis (DGGE). Links between plant and bacterial rhizoplane communities could not be detected by visual examination of T-RFLP patterns or DGGE banding profiles. Statistical analysis of fingerprint patterns did not reveal a relationship between bacterial community composition and plant species but did demonstrate an influence of plant community composition. The data also indicated that topography and other, uncharacterized, environmental factors are important in driving bacterial community composition in grassland soils. T-RFLP had greater potential resolving power than DGGE, but findings from the two methods were not significantly different.     

Use of the T-RFLP technique to assess spatial and temporal changes in the bacterial community structure within an agricultural soil planted with transgenic and non-transgenic potato plants

The aim of this study was to examine whether the terminal restriction fragment length polymorphism (T-RFLP) analysis represents an appropriate technique for monitoring highly diverse soil bacterial communities, i.e. to assess spatial and/or temporal effects on bacterial community structure. The T-RFLP method, a recently described fingerprinting technique, is based on terminal restriction fragment length polymorphisms between distinct small-subunit rRNA gene sequence types. This technique permits an automated quantification of the fluorescence signal intensities of the individual terminal restriction fragments (T-RFs) in a given community fingerprint pattern. The indigenous bacterial communities of three soil plots located within an agricultural field of 110 m(2) were compared. The first site was planted with non-transgenic potato plants, while the other two were planted with transgenic GUS and Barnase/Barstar potato plants, respectively. Once prior to planting and three times after planting, seven parallel samples were taken from each of the three soil plots. The T-RFLP analysis resulted in very complex but highly reproducible community fingerprint patterns. The percentage abundance values of defined T-RFs were calculated for the seven parallel samples of the respective soil plot. A multivariate analysis of variance was used to test T-RFLP data sets for significant differences. The statistical treatments clearly revealed spatial and temporal effects, as well as spacextime interaction effects, on the structural composition of the bacterial communities. T-RFs which showed the highest correlations to the discriminant factors were not those T-RFs which showed the largest single variations between the seven-sample means of individual plots. In summary, the T-RFLP technique, although a polymerase chain reaction-based method, proved to be a suitable technique for monitoring highly diverse soil microbial communities for changes over space and/or time.     

Molecular and culture-dependent analyses revealed similarities in the endophytic bacterial community composition of leaves from three rice (Oryza sativa) varieties

The endophytic bacterial communities of the three most important rice varieties cultivated in Uruguay were compared by a multiphasic approach. Leaves of mature plants grown in field experiments for two consecutive crop seasons were studied. No significant differences were found in the heterotrophic bacterial density for the three varieties. Pantoea ananatis and Pseudomonas syringae constituted 51% of the total of the isolates. These species were always present regardless of the variety or the season. Molecular analysis based on the 16S rRNA gene was performed by terminal restriction fragment length polymorphism (T-RFLP) and cloning. T-RFLP analysis revealed that bacterial communities grouped according to the variety, although the three varieties presented communities that showed 74% or higher similarities. Brevundimonas, the dominant genus in the clone library (18% of the clones), which might be present in all varieties according to T-RFLP profiles, was not recovered by cultivation. Conversely, bacteria from the genus Pseudomonas were not detected in the clone library. These results indicate that communities established in leaves of physiologically different rice varieties were highly similar and composed by a reduced group of strongly associated and persistent bacteria that were partially recovered by cultivation.