Conservative fragments in bacterial 16S rRNA genes and primer design for 16S ribosomal DNA amplicons in metagenomic studies

Abstract

Bacterial 16S ribosomal DNA (rDNA) amplicons have been widely used in the classification of uncultured bacteria inhabiting environmental niches. Primers targeting conservative regions of the rDNAs are used to generate amplicons of variant regions that are informative in taxonomic assignment. One problem is that the percentage coverage and application scope of the primers used in previous studies are largely unknown. In this study, conservative fragments of available rDNA sequences were first mined and then used to search for candidate primers within the fragments by measuring the coverage rate defined as the percentage of bacterial sequences containing the target. Thirty predicted primers with a high coverage rate (>90%) were identified, which were basically located in the same conservative regions as known primers in previous reports, whereas 30% of the known primers were associated with a coverage rate of <90%. The application scope of the primers was also examined by calculating the percentages of failed detections in bacterial phyla. Primers A519-539, E969-983, E1063-1081, U515 and E517, are highly recommended because of their high coverage in almost all phyla. As expected, the three predominant phyla, Firmicutes, Gemmatimonadetes and Proteobacteria, are best covered by the predicted primers. The primers recommended in this report shall facilitate a comprehensive and reliable survey of bacterial diversity in metagenomic studies.

Figure 1. Coverage rates of candidate primers within a conservative fragment.

The coverage rates (%) of eight candidate primers within the conservative fragment 5′-CAAGDMTGAAACTTAAAGGAAT-3′ were determined using all archaeal 16S rDNA sequences (>1,200 nt) as the reference dataset. The coverage rate is the percentage of the rDNA sequences that have a target fragment matching a given candidate primer. One mismatch is allowed in the match.

Figure 2. Phylum specificity of predicted primers for Archaea.

Information on the primers is listed in Table 2. The primers were used to find their targets in 1544 Crenarchaeota, 4153 Euryarchaeota, 37 Korarchaeota, 3 Nanoarchaeota, and 878 unclassified species. The sequences without a target were classified into different phyla.

Figure 3. Phylum specificity of predicted primers for Eubacteria.

Information on the primers is listed in Table 2. The primers were used to find their targets in 778 Aquificae, 325 Thermotogae, 550 Deinococcus, 1922 Chloroflexi, 651 Nitrospira, 245 Deferribacteres, 4655 Cyanobacteria, 195 Chlorobi, 91629 Proteobacteria, 94475 Firmicutes, 23266 Actinobacteria, 2274 Planctomycetes, 216 Chlamydiae, 2646 Spirochaetes, 202 Fibrobacteres, 4040 Acidobacteria, 33924 Bacteroidetes, 737 Fusobacteria, 2851 Verrucomicrobia, 441 Gemmatimonadetes, 108 Lentisphaerae, 114 OP10, 327 TM7, 1683 Tenericutes, 108 Dehalococcoides, and 6333 unclassified bacteria. The phyla with less than 100 sequences were ignored. The sequences without a target were classified into different phyla.

Figure 4. Primer distributions and distances to variant regions.

The V3, V5, and V6 regions are variant regions of the 16S rRNA genes. Their locations on the E. coli 16S rRNA gene were shown as shadowed regions. The relative distances of the primers (including the known and predicted primers) to the three regions were shown.