Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, September 2007, p. 3205-3211, Vol. 51, No. 9
0066-4804/07/$08.00+0 doi:10.1128/AAC.00625-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Patricia A. Bradford,1 and
C. Hal Jones1*
Infectious Diseases Discovery Research,1 Vaccines Discovery Research, Wyeth Research, Pearl River, New York2
Received 11 May 2007/ Returned for modification 8 June 2007/ Accepted 2 July 2007
|
|
|---|
8 µg/ml) and 248 (15%) were tetracycline resistant (MIC
8 µg/ml) but susceptible to minocycline (MIC
4 µg/ml). A total of 452 tetracycline-resistant, nonduplicate isolates were positive by PCR for at least one of the six tetracycline resistance determinants examined. Over half of the isolates encoding a single determinant were positive for tet(A) (26%) or tet(B) (32%) with tet(C), tet(D), tet(E), and tet(M), collectively, found in 4% of isolates. Approximately 33% of the isolates were positive for more than one resistance determinant, with the tet(B) plus tet(E) combination the most highly represented, found in 11% of isolates. The susceptibilities of the tetracycline-resistant strains to tigecycline (MIC90, 0.5 µg/ml), regardless of the encoded tet determinant(s), were comparable to the tigecycline susceptibility of tetracycline-susceptible strains (MIC90, 0.5 µg/ml). The results provide a current (2002 to 2006) picture of the distribution of common tetracycline resistance determinants encoded in a globally sourced collection of clinical E. coli strains. |
|
|---|
Tetracycline resistance in bacteria is mediated by four mechanisms: efflux, ribosomal protection, enzymatic inactivation, and target modification (10). Tetracycline efflux, first identified in 1953 in Shigella dysenteriae (1), and ribosomal protection, first identified in Streptococcus spp. (7), are now prevalent in both gram-negative and gram-positive bacteria (10). These resistance mechanisms are widely distributed in bacteria due to their association with mobilizable DNA elements, which have facilitated the spread to more than 50 genera, and are often coupled with multidrug resistance (11, 31, 33). Tetracycline resistance mediated through enzymatic inactivation and target modification has only been identified in a few bacterial species (34, 42) and, at present, is of limited clinical importance.
At present, 23 genes encoding efflux pumps and 11 genes encoding ribosomal protection proteins, not including the recently described mosaic tetracycline resistance genes (27), have been identified in bacteria since the first report of transferable tetracycline resistance in 1960 (33). The leading tetracycline resistance mechanism in E. coli is the extrusion of drug from the cytoplasm via efflux (10). Tetracycline-specific efflux pumps are members of the major facilitator superfamily (MFS) of efflux pumps (28). MFS pumps specific for tetracycline operate by transporting tetracycline in an energy-dependent fashion, via proton exchange, thereby reducing the intracellular concentration of the drug (10). Tet pumps are divided into six groups based on amino acid sequence, with Tet(A), Tet(B), Tet(C), Tet(D), and Tet(E) placed in group 1 due to amino acid sequence similarity (10, 33). Most tetracycline-specific efflux pumps confer resistance to tetracycline only; however, tet(B) encodes a pump that is able to extrude both tetracycline and minocycline (18, 29).
Tigecycline is the novel 9-t-butyl glycylamido derivative of minocycline that has been approved for use in complicated skin and skin structure infections and complicated intra-abdominal infections (2, 14) (Tygacil package insert, http://www.fda.gov/cder/foi/label/2005/021821lbl.pdf; Wyeth Pharmaceuticals Inc., Collegeville, PA). During the course of the tigecycline phase 3 clinical trials, all bacterial isolates were screened for susceptibility to tetracycline, among a panel of antibiotics. Tetracycline-resistant E. coli isolates (MIC
8 µg/ml) were examined by PCR for the presence of 10 resistance determinants: tet(A), tet(B), tet(C), tet(D), tet(E), tet(G), tet(K), tet(L), tet(M), and tet(Y).
|
|
|---|
Susceptibility determination. Broth microdilutions were conducted in accordance with Clinical and Laboratory Standards Institute recommendations, using fresh Mueller Hinton II broth (MHB) for tigecycline (12).
Primers, template preparation, and PCR assay.
The primer sets, PCR assay conditions, and positive control strains known to harbor the specific determinants for each primer set were previously described (22). In addition, for the present work, PCR assays were designed to detect the tet(G) (GenBank accession no. AF133139), tet(Y) (GenBank accession no. AB089608), and tet(L) (GenBank accession no. U17153) genes. The primer set used for the amplification of tet(G) was F, 5'CAT TGC CCT GCT GAT CG; and R, 5' TTG GTG AGG CTT GTA AGC. The following primers were used for tet(Y): F, 5' CCG CAC TCA TTG TTG TCG; R, 5' TTT TCA TCG CAA ACA AGA CC. The primer set used for the detection of tet(L) was F, 5' ATA AAT TGT TTC GGG TCG GTA AT; and R, 5' AAC CAG CCA ACT AAT GAC AAT (8). The primers and conditions for tet(K) (GenBank accession no. AJ888003) amplification were previously described (21). The positive control strains were Salmonella enterica serovar Typhimurium strain DT104 (5, 37) for tet(G) and Enterococcus faecalis strain BM4253 for tet(L) (8) and a recombinant E. coli DH5
strain transformed with plasmid pIE1120 containing tet(Y), kindly provided by E. Tietze (38, 39). The expected size of the amplification product for each target gene was as follows: tet(G), 993 bp; tet(K), 1.08 kb; tet(L), 1.07 kb; and tet(Y), 949 bp. Lysate production and PCR amplifications were carried out as previously described (22).
|
|
|---|
8 µg/ml), and 248 (15%) were tetracycline resistant but susceptible to minocycline (MIC
4 µg/ml). The 653 resistant isolates were screened by PCR for the common tetracycline determinants tet(A), tet(B), tet(C) tet(D), tet(E), and tet(M). Ribotyping determined that 488 isolates were nonduplicates on the patient level (data not shown). Of these, 452 (93%) were positive for one or more of the six tested determinants, whereas 36 (7%) isolates were negative for all six tetracycline resistance determinants on the original screening panel. These 36 isolates were further screened for the presence of the tet(G) and tet(Y) genes, which have been previously described in E. coli (5, 39), as well as the tet(K) and tet(L) genes, which have been detected in some gram-negative bacteria (33). All 36 isolates were negative for tet(G), tet(K), tet(L), and tet(Y) (data not shown); the basis of tetracycline resistance in these isolates is the focus of ongoing research. Worldwide prevalence of tetracycline resistance determinants. An examination of single tetracycline resistance determinants revealed that tet(A) and tet(B) accounted for 58% of all isolates, with 26% of isolates encoding only tet(A) and 32% of isolates encoding only tet(B) (Table 1). The other efflux pump-encoding genes were found much less frequently, with 3% of isolates encoding tet(D), 1% of isolates encoding tet(C), and only a single strain encoding the tet(E) gene. Interestingly, the tet(E) determinant in combination with tet(B) (11%) was found more frequently than the tet(E) determinant alone (0.2%); in fact this was the most common combination of multiple efflux determinants identified (Table 2). The other two most frequent combinations were the tet(A) plus tet(D) pair and the tet(B) plus tet(D) pair seen in 5.1% and 6.6% of isolates, respectively. Overall, 33% of isolates encoded more than one tetracycline resistance determinant, with 2.5% of isolates encoding three or more determinants.
|
View this table: [in a new window] |
TABLE 1. Regional distribution of E. coli isolates with a single tetracycline resistance determinant
|
|
View this table: [in a new window] |
TABLE 2. Regional distribution of E. coli isolates with multiple tetracycline resistance determinants
|
Multiple-determinant isolates encoding tet(B) plus tet(E) accounted for >16% of isolates from Europe (Eastern and Western) and South Africa. Similarly, multiple-determinant isolates from North America encoding the tet(B) plus tet(D) pairing accounted for 17% of isolates from that region and isolates encoding the tet(A) plus tet(D) pair accounted for 10% of isolates from the Asia-Pacific region and Western Europe (Table 2).
The study also identified seven isolates, two from Taiwan and one each from Belgium, China, France, Guatemala, and Romania, expressing tet(M) homologues encoding a ribosomal protection protein. Although reported previously in commensal E. coli isolates (25), tet(M) had not been reported in human clinical E. coli isolates until 2006 (22). All of the isolates, with the exception of the isolate from Guatemala, also encoded the tet(A) resistance determinant.
Antibiotic susceptibility.
Tetracycline and minocycline susceptibility data for the strain collection are presented in Table 3 and Fig. 1. The tet(B) gene was detected in 272 (56%) isolates, either alone or in combination with additional determinants. The minocycline susceptibility pattern of this group of strains (MIC50, 8 µg/ml; MIC90, 32 µg/ml; MIC range, 0.25 to >64 µg/ml) reflects the unique ability of TetB to efflux minocycline (18, 29). The remainder of the strains, encoding tet(A), tet(C), tet(D), tet(E), or tet(M) or a combination of these determinants, were highly resistant to tetracycline (MIC90,
64 µg/ml) and would be categorized as intermediate with respect to minocycline (MIC90, 8 µg/ml).
|
View this table: [in a new window] |
TABLE 3. Tigecycline, minocycline, and tetracycline susceptibilities of E. coli isolates expressing various tetracycline resistance determinants
|
![]() View larger version (14K): [in a new window] |
FIG. 1. Susceptibility profile of 488 tetracycline-resistant E. coli isolates. The figure shows the cumulative percentage of strains versus the respective MICs of tetracycline ( ), minocycline ( ), and tigecycline (). The leftward shift for the minocycline trace is indicative of the subset of isolates that fail to efflux minocycline; the further leftward shift of the tigecycline trace underlies the inability of classical tetracycline resistance determinants to circumvent tigecycline.
|
|
|
|---|
E. coli strains collected during the tigecycline clinical trials showed an incidence of tetracycline resistance similar (39%) to those in recently reported worldwide surveillance studies (30 to 45%) (16, 17, 35, 36). In the current study, a large percentage of the resistant strains (33%) encoded multiple tetracycline efflux pumps. This confirms the recent upward trend in which >30% of environmental isolates were found to encode multiple tetracycline efflux pumps (6, 40), an increase of more than three times that reported recently from other sources (33). A correlation between the number of encoded tetracycline resistance genes and the level of tetracycline resistance (in MICs) has not been noted. In addition to the increasing burden of tetracycline in the environment, the increased use of various biocides, i.e., triclosan and chlorhexidine, may be responsible for the increase in the frequency of isolates encoding multiple tetracycline resistance determinants (4, 23). The continued genetic exchange of resistance determinants among various environmental, commensal, and clinical bacteria would also be expected to have a clinical impact.
There were 16 isolates in the strain collection that, although PCR positive for tet(B), were susceptible to minocycline (MIC,
4 µg/ml) and resistant to tetracycline (MIC,
64 µg/ml). Of these isolates, 13 encoded an additional tet determinant, tet(A) or tet(E). Therefore, it is possible that the tet(B) gene is either poorly expressed or not expressed in these strains. For the three strains that expressed only tet(B) and resulted in an MIC of
4 µg/ml for minocycline and an MIC of >64 µg/ml for tetracycline, it may be that the level of expression is too low to result in overt resistance to minocycline or that tetracycline resistance in these isolates is mediated by a gene not included in our PCR panel.
Somewhat surprisingly, 36 isolates in this clinical collection were negative for tet(A) to tet(E) and tet(M), as well as tet(G), tet(K), tet(L), and tet(Y). One possible explanation is that a number of these isolates encode a mosaic gene encoding a ribosomal protection protein, as has recently been described in a number of human and animal fecal samples (27). Our primer sets were not designed to capture these unique genetic elements that therefore would have been missed in this analysis.
Four E. coli isolates were collected that tested with a tigecycline MIC of 2 µg/ml. Two of these isolates were negative for all tet determinants on the PCR screening panel, one isolate encoded tet(A), and the fourth isolate was shown to encode tet(A) plus tet(D). As 24 other isolates encoded this combination and this group of strains had an MIC90 of 1 µg/ml, there was no correlation between this combination of efflux pumps and reduced susceptibility to tigecycline. Recent studies have implicated the overexpression of AcrAB, a member of the RND multidrug efflux family, in E. coli as contributing to reduced susceptibility to tigecycline (D. Keeney, A. Ruzin, F. McAleese, and P. A. Bradford, submitted for publication).
Tigecycline, a new broad-spectrum antibiotic, is the first antibiotic of the glycylcycline class approved for clinical use (Tygacil package insert, Wyeth Pharmaceuticals Inc., Collegeville, PA). Tigecycline evades both the ribosomal protection and efflux pump resistance mechanisms that are the two most common mechanisms of tetracycline resistance in clinical strains (15, 29). In vitro studies have demonstrated that tigecycline is not a substrate for the TetB efflux pump and does not induce proton transport across membranes (19). Ribosomal binding studies have demonstrated that tigecycline binds more tightly than tetracycline and minocycline to the ribosome; therefore, tigecycline may be less efficiently displaced from the 30S ribosomal subunit by ribosomal protection proteins than the classical tetracyclines (3, 26). The results of this study confirm that the presence of one or more tetracycline resistance determinants does not affect tigecycline susceptibility in clinical isolates of E. coli.
These studies provide a current (2002 to 2006) picture of the burden of tetracycline resistance determinants among clinical E. coli isolates, as well as the utility of the novel broad-spectrum agent, tigecycline, against these pathogens. Moreover, these results support the general approach of reengineering existing antimicrobial agents with acceptable safety profiles to evade the resistance mechanisms posed by bacterial pathogens.
Published ahead of print on 9 July 2007. ![]()
Present address: Discovery Neurosciences, Wyeth Research, Princeton, New Jersey. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»