Abstract
The three chromosomally located clustered genes vanC1, vanXYc, and vanT confer intrinsic resistance to vancomycin and are used for species identification of Enterococcus gallinarum. In this study, 28 strains belonging to the E. gallinarum/casseliflavus group isolated from cloacal swabs from laying hens were screened for the presence of vanC1. As confirmed by species-specific multiplex PCR, 11 vanC1-positive strains were identified as E. gallinarum. Surprisingly, one yellow pigmented strain, verified as E. casseliflavus by species-specific multiplex PCR, was also vanC1 positive; vanXYc and vanT were additionally detectable in this strain. To our knowledge, this is the first report of vanC1, vanXYc, and vanT in E. casseliflavus. The minimum inhibitory concentration of vancomycin was 4 mg/L. Real-time reverse transcription-PCR revealed that none of the clustered genes was expressed in this strain. Even if the genes seem not to be active, there is a certain risk that they will be transferred to other bacteria where they might be functionally expressed. Therefore, it may be advisable to expand the search for vanC1, vanXYc, and vanT from E. gallinarum to other (enterococcal) species. This study confirms that enterococci live up to their name as being reservoir bacteria and should therefore always be closely monitored.
Introduction
E
Materials and Methods
Bacterial strains
In a previous study, a total of 923 Enterococcus species were isolated after pre-enrichment in buffered peptone water (37°C, 24 hr; Merck, Darmstadt, Germany) on citrate azide tween carbonate agar (37°C, 24 hr; Merck,) from cloacal swabs from conventionally and organically kept laying hens in Bavaria, Germany. 15 Of these, 35 strains were biochemically identified as Enterococcus belonging to the gfmc group, consisting of E. gallinarum, E. flavescens, E. mundti, and E. casseliflavus. 15 All enterococci were frozen at -70°C until needed; the strains belonging to the species E. gfmc group were revived on Standard I Nutrient (Merck) agar for further investigation in the present study.
Species verification
After subculturing on Standard I Nutrient agar and evaluation of pigmentation, DNA was extracted from pure colonies by using the cetyltrimethylammonium bromide method as previously described. 6 Species verification of the 35 E. gfmc group strains was performed by species-specific multiplex PCR as described elsewhere. 8 Briefly, a genus-specific primer pair (E1/2; synthesized by Metabion, Martinsried, Germany) targeted an 800 bp fragment of Enterococcus spp., and species-specific primer pairs targeted a 173 bp (E. gallinarum; GA1/2) and a 288 bp (E. casseliflavus; CA1/2) fragment, respectively. E. gallinarum DSM 20628 and E. casseliflavus DSM 20680 were used as reference strains (DSMZ, German Collection of Microorganisms and Cell Cultures). Reaction products were visualized in UV light after electrophoresis through a 1.5% agarose gel (BioRad, München, Germany) containing 0.285 μg/ml ethidium bromide (Sigma-Aldrich, Hamburg, Germany).
Antimicrobial susceptibility testing
Of the strains identified as E. gallinarum and E. casseliflavus (n=28), minimum inhibitory concentrations (MICs) of vancomycin (VAN; concentration range: 0.5–64 mg/L) and teicoplanin (TPL; concentration range: 0.25–32 mg/L) were determined by the broth microdilution method according to DIN 58940-81 (2002) and as previously described.4,15
Detection of vanC1 by PCR
The 28 strains that were identified as E. gallinarum/casseliflavus by species-specific multiplex PCR (see “Species verification” section) were screened for the presence of the vanC1 gene. An 822 bp fragment of the vanC1 gene was detected by PCR as described elsewhere. 5 Each PCR run contained one positive control (E. gallinarum BM 4174; Institut Pasteur, Paris, France) and one negative control (master mix without template DNA). Reaction products were visualized in UV light after electrophoresis through a 1.5% agarose gel (BioRad) containing 0.285 μg/ml ethidium bromide (Sigma-Aldrich). Extraction of overnight grown colonies and PCR runs were repeated three times for the vanC1-positive strain to verify the results and to exclude contamination.
Detection of vanXYc and vanT by real-time PCR
The vanC1-positive strain was tested for the presence of vanXYc and vanT, which are normally clustered with vanC1,1 as previously described. 14 Briefly, the primers vanXYcFW 5′-GCAAAACAATGGGAACGACT-3′ and vanXYcRV 5′-ATCTCGAAAATGAGGGCAGA-3′, targeted a 246 bp fragment of the vanXYc gene, and the primers vanTcFW 5′-CTCAAAGACAGCCCTTTTGC-3′ and vanTRV 5′-TTCCGCAATCGAATACCTTC-3′, targeted a 189 bp fragment of the vanT gene. PCR mixtures (total volume 20 μl) consisted of 1.6 μl MgCl2, 1 μl forward primer, 1 μl reverse primer, 2 μl SYBR Mix (Roche, Mannheim, Germany), and 2 μl template DNA. LightCycler® conditions were as follows: denaturation at 95°C for 10 min; 35 cycles denaturation at 95°C for 10 sec, amplification at 56°C/57°C and elongation at 72°C for 10/8 sec (vanXYc/vanT); cooling at 40°C for 30 sec. For positive and negative control, (see “Detection of vanC1 by PCR” section).
Analysis of gene expression by real-time reverse transcription-PCR
The vanC1-positive E. casseliflavus, one vanC1-positive E. gallinarum (serving as a positive control) and two further, randomly selected E. casseliflavus strains (no-vanC1 controls) were examined for the expression of vanC1, vanXYc, and vanT by two-step quantitative real-time reverse transcription-PCR (RT-PCR) as previously described. 14 Briefly, bacterial suspensions were adjusted to OD600=0.8; total RNA was extracted with the NucleoSpin® RNAII kit (Macherey-Nagel, Düren, Germany) using 5 mg/ml lysozyme (Sigma-Aldrich). One microgram of each total RNA sample was reverse transcribed in a total volume of 60 μl, containing 12 μl 5×buffer (Promega, Mannheim, Germany), 10 mM dNTPs (Fermentas, St. Leon-Rot, Germany), 50 μM hexamer primers (Invitrogen, Carlsbad, CA), and 1 μl M-MLV RT Enzyme RNase H(-)Point Mutant incl. buffer (Promega). DNA contamination of the RNA extracts was excluded by performing a pooled no-RT control. The resulting cDNA was amplified by real-time PCR as described above and previously. 14 Briefly, each mixture contained 1 μl forward and reverse primer (primer sequences for detection of vanC1, vanXYc, and vanT, see “Detection of vanC1 by PCR” and “Detection of vanXYc and vanT by real-time PCR” sections). LightCycler conditions were as follows: denaturation at 95°C for 10 min; 35 cycles denaturation at 95°C for 10 sec, annealing at 54°C/56°C/57°C and elongation at 72°C for 33/10/8 sec (vanC1/vanXYc/vanT); cooling at 40°C for 30 sec.
Results
Using species-specific multiplex PCR, 11 strains were confirmed as E. gallinarum and 17 as E. casseliflavus; 7 of the 35 tested strains were only positive in the genus-specific Enterococcus PCR, but not in the species-specific PCR and were thus excluded for further investigations. Of the 28 remaining strains confirmed as E. gallinarum/casseliflavus, 12 were positive for vanC1. The colonies of 11 vanC1-containing cultures were white on Standard I Nutrient agar, but the colonies of one pure culture were yellow pigmented. Noticeably, the white vanC1-positive strains were confirmed as E. gallinarum, whereas the yellow vanC1-positive strain was confirmed as E. casseliflavus by species-specific multiplex PCR. Additionally, the clustered genes vanXYc and vanT were also detectable in the vanC1-positive E. casseliflavus by real-time PCR. The 17 yellow, vanC1-negative strains were identified as E. casseliflavus as well (Table 1). MICs for vancomycin ranged from ≤0.5 to 8 mg/L (median 4 mg/L). All strains were susceptible to teicoplanin (MIC ≤0.5 mg/L; Table 1). The genes vanC1, vanXYc, and vanT were amplified by real-time RT-PCR in E. gallinarum, proving the validity of the method (positive control). However, neither vanC1 nor vanXYc or vanT were detectable by real-time RT-PCR in the vanC1-positive E. casseliflavus strain, revealing that none of the genes of the cluster was expressed in E. casseliflavus. The no-vanC1 controls (E. casseliflavus) were also negative by real-time RT-PCR, which excludes contamination.
As determined by species-specific multiplex PCR according to Jackson et al. 8 The 800 bp genus-specific fragment was amplified in all strains; E. gallinarum=presence of the 173 bp band and absence of the 288 bp amplicon; E. casseliflavus=presence of the 288 bp band and absence of the 173 bp amplicon.
MIC, minimum inhibitory concentration; VAN, vancomycin; TPL, teicoplanin.
Discussion
Most human enterococcal infections are caused by E. faecalis and E. faecium—however, there seems to be an increase in infections caused by other enterococci, including E. gallinarum and E. casseliflavus. 9 Therefore, there is a need for accurate identification of enterococci at the species level.9,11
The VanC phenotype of E. gallinarum and E. casseliflavus is characterized by intrinsic low-level resistance to vancomycin (MIC 2–32 mg/L) and susceptibility to teicoplanin (MIC 0.5–1 mg/L).3,10,12 Consistently, all strains of the present study were susceptible to teicoplanin (MIC ≤0.5 mg/L). On the contrary, 10 strains of E. casseliflavus were inhibited by low concentrations of vancomycin (MIC ≤1 mg/L; see Table 1), leading to the classification “susceptible to vancomycin.” 4 Therefore, the VanC resistance phenotype is not always reliable for the identification of E. casseliflavus.
The chromosomally located vanC1 gene is used as a species marker for E. gallinarum.5,10,13 The vanC1 gene and two further clustered genes, vanXYc and vanT, are responsible for intrinsic low-level resistance to vancomycin in E. gallinarum. 1 To our knowledge, vanC1 has not yet been detected in another species than E. gallinarum and, in rare cases, in E. faecalis: we have recently found vanC1 in a phenotypically vancomycin-susceptible E. faecalis strain, 14 and there is, until now, one further report on the presence of vanC1 in a resistant E. faecalis strain. 7
To our knowledge, vanXYc and vanT have only been found in E. gallinarum so far. 1 Therefore, the present study is the first report on the presence of vanC1 in E. casseliflavus and, moreover, of vanXYc and vanT outside E. gallinarum. A species identification error is unlikely because the yellow pigmentation and biochemical reactions were typical, 16 and the species-specific multiplex PCR 8 confirmed the results of the conventional bacteriological investigation. Real-time RT-PCR revealed that none of the genes of the vanC gene cluster was expressed. Nevertheless, it is a cause for concern that E. casseliflavus is able to naturally acquire this cluster, since there is a risk that these genes will be transferred to other bacteria where they might be functionally expressed.
The presence of vanC1 in E. faecalis as well as the presence of the gene cluster vanC1/vanXYc/vanT in E. casseliflavus in the present study might cast doubts on the reliability of vanC1 for species identification of E. gallinarum in the future. On the other hand, these findings seem to be highly exceptional. Either way it may be advisable and interesting to search for vanC1/vanXYc/vanT not only for species identification purposes in E. gallinarum, but rather to expand the screening to other (enterococcal) species. This study confirms once again that enterococci live up to their name as being reservoir bacteria. 14 Therefore, this genus should always be closely monitored, regardless whether the isolates are pathogenic or commensal. 14
Footnotes
Acknowledgments
This work was supported by the Bavarian State Ministry of the Environment and Public Health for funding the project. Thanks go to Heike Kliem and technicians for providing the laboratory and equipment for real-time RT-PCR. We also gratefully thank Barbara Fritz, Cornelia Oehme, Thomas Korbica, and Angelika Barreiro-Gebhard for excellent technical assistance.
Disclosure Statement
No competing financial interests exist.