Each was also subject to surface sterilization (designated by an s) to determine just the endophytic community. Numbers are the % of the total

number of sequences (mean 2,515 per sample) for each sample that were classified as a particular taxa, and only taxa accounting for > 0.1% of the sequences across all samples are shown. *indicates taxa that accounted for significantly different (p < 0.05) percentages of the total community between either sterilized and non-sterilized samples (S) or conventional versus organic production (O). While buy JNK-IN-8 sequences corresponding to 23 taxa were detected at a frequency that was > 0.1% of all of the sequences examined, other “rare” OTUs were Milciclib detected at low levels. Of the 634 different OTUs RGFP966 recognized, 319 were represented by just one sequence read in a single sample, and a further 104 by just two sequence reads. The number of OTUs detected in each sample, when standardized to the same number of reads, was used as a simple measure of bacterial community diversity. An average of 47 OTUs were detected in each sample, but this varied from 17 (the samples from surface-sterilized and non-sterilized

organic romaine lettuce) to 92 (the organic red leaf lettuce sample; Table  3). These values are in the same range as those reported for the leaf surface bacterial communities on store-bought lettuce and spinach [19], and are similar or slightly lower than diversity estimates reported for stems and leaves of alfalfa [3]. However, they are an order of magnitude lower than estimates of bacterial endophyte diversity derived from pyrosequencing of potato roots [2], although that study relied on diversity statistics (e.g. the Chao statistic) rather than directly assessing the number of distinct OTUs. Bacterial densities in leaves are also thought to be lower than those in roots or the rhizosphere [5, 20],

which may account for less diverse bacterial communities in above-ground plant structures. There were Dapagliflozin no consistent patterns in OTU richness in regards to organic versus conventional produce or in terms of surface-sterilized versus non-sterilized samples (p > 0.05 for both comparisons), but surface-sterilized (i.e. endophyte) diversity was moderately correlated with overall bacterial diversity determined from the non-sterilized samples (R = 0.68). It should be noted, that these diversity estimates are likely to be low given that sequences were grouped into OTUs based on the more conservative 97% similarity criterion and that rarefaction curves (Additional file 1) did not always reach an asymptote.