Sequencing the Gene Encoding Manganese-Dependent Superoxide Dismutase for Rapid Species Identification of Enterococci

Enterococci, although not highly virulent microorganisms, have emerged worldwide in the last decade as one of the leading causes of nosocomial bacteremia, surgical wound infections, and urinary tract infections (9, 10, 13, 24). This evolution is mainly due to the appearance of multiresistant strains of enterococci that are resistant to most antibiotics used in treatment (e.g., ampicillin, aminoglycosides, and glycopeptides). Most human enterococcal infections (≥90%) are caused by Enterococcus faecalis and Enterococcus faecium; however, the incidence of other species, such as Enterococcus casseliflavus and Enterococcus gallinarum, could be underestimated because of bacterial misidentification. In clinical laboratories, accurate identification of enterococcal species is required to carry out a proper epidemiologic surveillance and may help in the management of infected patients in case of relapse. This is usually done by testing tolerance to bile esculine and tellurite, growth in 6.5% NaCl broth, and specific carbohydrate utilization (2, 6); by characterizing bacterial motility and pigment production (1); and by using commercial biochemical test systems, such as the API 20STREP or Rapid ID 32 Strep systems. However, these phenotypic methods are often not reliable and the automated systems, such as the Vitek and MicroScan systems, do not properly identify enterococci other than E. faecalis and E. faecium in the absence of additional tests (11). Consequently, several genotypic methods based on the analysis of PCR products derived from selected target DNA have been developed for the species identification of enterococci (3, 14, 22). This includes the determination of the 16S ribosomal DNA (rDNA) sequence (18), a strategy which is now greatly facilitated by the use of universal 16S PCR primers associated with the development of simplified, partially automated, and cost-effective sequencing technologies. However, the interpretation of these data may be complicated by the fact that divergent 16S rDNA sequences may exist within a single organism (23) or, alternatively, that closely related species may have identical 16S rDNA sequences (8), as recently shown in the genera Enterococcusfor E. casseliflavus and E. gallinarum(18). To solve this problem, it is possible to use alternative monocopy target sequences which exhibit a higher divergence than that of the 16S rDNA. The sodA gene of the gram-positive cocci which encodes the manganese-dependent superoxide dismutase fulfills these criteria and we recently reported that sequencing of the sodA PCR product with the use of a single pair of degenerate primers constitutes a valuable approach to the genotypic identification of the 29 streptococcal species (20). In this work, the same universal primers (19) were used to construct a sodA database of 19 enterococcal species, including E. casseliflavus andE. gallinarum. We show the usefulness of this library for a rapid sequence-based identification method of enterococcal isolates.

The main characteristics of the bacterial strains used in this study, including the type strains, are listed in Table1 and 2. Rapid extraction of bacterial genomic DNA was carried out by using the InstaGene Matrix (Bio-Rad, Hercules, Calif.) on cells collected from 2 ml of an overnight culture. The sodA degenerate primersd1 (5′-CCITAYICITAYGAYGCIYTIGARCC-3′) andd2 (5′-ARRTARTAIGCRTGYTCCCAIACRTC-3′) were used to amplify an internal fragment, designatedsodA_int , representing approximately 85% of their sodA genes. PCRs were performed on a Gene Amp System 9600 instrument (Perkin-Elmer Cetus, Roissy, France) in a final volume of 50 μl containing 250 ng of DNA as template, 0.5 μM each primer, 200 μM each deoxynucleoside triphosphate, and 1 U of AmpliTaq Gold DNA polymerase (Perkin-Elmer) in a 1× amplification buffer (10 mM Tris-HCl [pH 8.3], 50 mM KCl, 1.5 mM MgCl₂). The PCR mixtures were denatured (3 min at 95°C) and then subjected to 30 cycles of amplification (60 s of annealing at 37°C, 60 s of elongation at 72°C, and 30 s of denaturation at 95°C) and 7 min at 72°C for the last elongation cycle. A single DNA fragment corresponding to the expected 480-bp amplification product,sodA_int , was observed in all cases following agarose gel electrophoresis and ethidium bromide staining (data not shown). PCR products were purified on an S-400 Sephadex column (Pharmacia, Uppsala, Sweden) and directly sequenced on both strands with the oligonucleotides d1 and d2 by using the ABI-PRISM Big Dye terminator sequencing kit on a Genetic ABI-PRISM 310 Sequencer Analyzer (Perkin-Elmer). The cycle sequencing protocol was optimized as follows: the sequencing mixtures were subjected to 40 cycles of amplification consisting of 10 s of denaturation at 96°C, 5 s of annealing at 40°C, and 4 min of elongation at 60°C.

The nucleotide sequences of the sodA_int fragments from the type strains of Enterococcus avium,E. casseliflavus, Enterococcus cecorum,Enterococcus columbae, Enterococcus dispar,Enterococcus durans, E. faecalis, E. faecium, Enterococcus flavescens, E. gallinarum, Enterococcus hirae, Enterococcus malodoratus, Enterococcus mundtii, Enterococcus pseudoavium, Enterococcus raffinosus,Enterococcus saccharolyticus, Enterococcus seriolicida, Enterococcus solitarius,Enterococcus sulfureus, and Lactococcus garvieaewere determined (Table 1). We assumed that the PCR products sequenced were actual sodA_int fragments, since the corresponding deduced polypeptides all contained the amino acids characteristic of the manganese-dependent superoxide dismutase (16, 17) at the expected positions (data not shown). Multiple alignment of these sodA_int DNA sequences plus those of L. garvieae (Table 1),Lactococcus lactis (19), Streptococcus bovis (20), and Streptococcus pyogenes(20) was carried out with the Clustal X program (12), and an unrooted phylogenetic tree was constructed by the neighbor-joining (NJ) method (21). The sequences of the degenerate oligonucleotides d1 and d2 and alignment gaps were not taken into consideration for calculations. The reliability of the tree nodes was evaluated by calculating the percentage of 1,000 bootstrap resamplings that support each topological element. Only the nodes having a bootstrap value greater than 95% are shown in Fig. 1, since this critical value could be used to define the monophyly of a clade of related organisms (7). This analysis revealed that, as expected, the members of the genus Enterococcus, with the exception ofE. seriolicida, were clustered within a clade supported by 99.5% of the bootstrap replicates. The sodA_int sequences of E. seriolicida and of L. garvieaewere almost identical (99.5% sequence identity) and were clustered with that of L. lactis within a clade supported by 96.3% of the bootstrap confidence limit (Table 3and Fig. 1). These results are consistent with the redesignation ofE. seriolicida as L. garvieae (4). The phylogenetic tree representing the enterococcalsodA_int sequences (Fig. 1) has the same topology as the NJ tree constructed from the analysis of their 16S rDNA sequences (18). It is worth noting that thesodA_int sequences of the E. casseliflavus and E. gallinarum type strains displayed 16.9% sequence divergence, a value similar to the 19.7% sequence divergence observed between the ddl genes encoding thed-Ala–d-Ala ligases in these species (5). These results do not support the suggestion thatE. casseliflavus and E. gallinarum comprise a single species (18). By contrast, the fact that the 16S rDNA sequence (18) and the ddl (15),vanC (3), and sodA_int (Table 3) genes of the E. casseliflavus and E. flavescens type strains were almost identical (99.9, 99.5, 96, and 98% sequence identity, respectively) suggests that they should be associated in a single species.

• Open in new tab
• Download powerpoint

Fig. 1.

Phylogenetic unrooted tree showing relationships among the sodA_int fragments from various enterococcal type strains. The tree was established from an analysis of the sequences listed in Table 1 by using the NJ method. ThesodA_int sequences of L. lactis,L. garvieae, S. bovis, and S. pyogenestype strains were included in this work. The value on each branch is the estimated confidence limit (expressed as a percentage) for the position of the branch as determined by bootstrap analysis. Only bootstrap values greater than 95%, which were considered significant, are indicated. The scale bar (NJ distance) represents a 10% difference in nucleotide sequences.

The phylogenetic tree shown in Fig. 1 revealed the presence of two major clusters within the enterococcal species which we have designated the faecium group (E. faecium,E. durans, E. hirae, and E. mundtii) and the avium group (E. avium,E. malodoratus, E. pseudoavium, andE. raffinosus). Within each group, the 16S rDNA sequences exhibited more than 99% sequence identity (18), whereas the highest percentage of similarity found between twosodA_int sequences was 87.9% (Table 3). These results confirm that the gene sodA constitutes a more discriminative target sequence than 16S RNA in differentiating closely related bacterial species.

Fifteen enterococcal isolates were identified by using conventional microbiological tests, Rapid ID 32 Strep, and thesodA_int systems (Table 2). In all cases, thesodA_int sequences of the isolates displayed less than 1.5% divergence from the corresponding type strain. For 10 strains (NEM1616, NEM1617, NEM1621, NEM1623, NEM1624, NEM1625, NEM1626, NEM1627, NEM1628, and NEM1630), the two methods gave the same results. Four isolates (NEM1618, NEM1620, NEM1622, and NEM1629) were identified at the species level with thesodA_int system but not with conventional microbiological tests or the Rapid ID 32 Strep system. The remaining isolate, NEM1619, was identified with the Rapid ID 32 Strep system asE. hirae but was identified with thesodA_int system as E. durans (Table2). The reliability of the molecular identification of NEM1164 was based on the fact that its sodA_int fragment displays 99.5 and 85% sequence identity with the type strains ofE. durans and E. hirae, respectively.

In conclusion, we have determined thesodA_int sequences of the type strains ofE. avium, E. casseliflavus, E. flavescens, E. cecorum, E. columbae,E. dispar, E. durans, E. faecalis,E. faecium, E. gallinarum, E. hirae, E. malodoratus, E. mundtii,E. pseudoavium, E. raffinosus, E. saccharolyticus, E. seriolicida, E. solitarius, and E. sulfureus and demonstrated the usefulness of this database for the species identification of enterococcal isolates. The identification method presented in this study is not accessible to routine clinical microbiology laboratories, but it may become the “gold standard” technique in reference and large research hospital laboratories for epidemiologic purposes and/or identifying problematic strains.

Nucleotide sequence accession number. Representative nucleotide sequences have been submitted to the EMBL database and have been given accession no. AJ387906 to AJ387925 and AJ387927 to AJ387941.

• Copyright © 2000 American Society for Microbiology