Stefan Schillberg , Detlef Schumann and Rainer Fischer RWTH

In prokaryotes, the ribosomal gene locus contains three rRNA species: 16S, 23S and 5S. The potential of sequencing 16S rDNA of bacteria and archaea for establishing phylogenetic relationships and for use in molecular diagnostics is well-documented. Currently, over 3000 16S rRNA sequences are available in the databases. However, the potential of the 23S rRNA gene as a diagnostic tool has not been fully explored. Strong secondary structure and a larger size (3 vs. 1.5 kb for 16S rDNA) have hindered efficient sequencing of the 23S rRNA gene. So, there are just over 100 bacterial 23S rRNA sequences available for bacteria (4,5,8). Commonly, 23S rRNA gene sequence is determined either by sequencing of cloned fragments (1) or by amplification of partial fragments and subsequent direct sequencing of these products (2,9). In a recent report on direct sequencing of the PCR-amplified 23S rDNA, Sallen et al. (9) described the preparation of the 23S rDNA by three amplification steps to obtain overlapping fragments and the subsequent use of 33 sequence primers to achieve the complete sequencing of the 23S rRNA gene. We describe a simple protocol that allows amplification of almost the complete 23S rRNA gene and the subsequent direct sequencing of the doublestranded fragment, thereby eliminating the need for cloning or preparation of single-stranded template (7). We adapted a sequencing strategy, developed by Amersham (Amersham International plc, Little Chalfont, Bucks, England, UK) for sequencing DNA clones, for direct sequencing of PCR-amplified 23S rDNA. We have chosen sequencing of the 23S rRNA gene of the human pathogen Francisella tularensis as a model to demonstrate the potential of the described strategy. Fresh cultures of Francisella strains were pelleted and resuspended in 1 mL TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0). The suspension was heated to 80°C for 30 min and centrifuged at 10 000× g for 15 min, and 1 μL of the supernatant was added to a polymerase chain reaction (PCR) master mixture containing (in a final volume of 50 μL) 1× PCR buffer [20 mM Tris-HCl, 10 mM KCl, 10 mM (NH4)2SO4, 1% Triton X-100, 1.5 mM MgSO4 and 1 mg/mL of bovine serum albumin], 200 μM of each dNTP and 10 pmol of each primer. The mixture was overlaid with 50 μL of mineral oil and heated to 80°C before 1 U each of Taq DNA polymerase and Taq extender (Stratagene, La Jolla, CA, USA) was added. Immediately after adding the enzymes, the tubes were subjected to 30 cycles of denaturation for 1 min at 94°C, annealing for 1 min at 60°C and extension for 2.5 min at 72°C on a Model 480 Thermal Cycler (PerkinElmer, Norwalk, CT, USA). The amplification process was terminated by a 10-min extension at 72°C, and tubes were rapidly cooled to 4°C. The primers used were chosen so that the 16S–23S spacer region and virtually the complete 23S rRNA gene would be amplified, thereby enabling determination of the 5′ end of the gene. The resulting fragment is approximately 3 kb (because of a rather short spacer of 400 bp in F. tularensis, Figure 1). The primers are complementary to positions 1522–1538 and 2669–2654 of the 16S and 23S rRNA genes, respectively (Table 1). Fragments were separated by electrophoresis on a 1% agarose gel. Amplification products were excised