ABSTRACT
Tigecycline is regarded as a last-resort treatment for carbapenem-resistant Klebsiella pneumoniae (CRKP) infections, but increasing numbers of tigecycline-resistant K. pneumoniae isolates have been reported. The tigecycline resistance mechanisms in CRKP are undetermined. This study aimed to elucidate the mechanisms underlying tigecycline resistance in 16 tigecycline- and carbapenem-resistant K. pneumoniae (TCRKP) isolates. Mutations in tigecycline resistance determinant genes [ramR, acrR, oqxR, tet(A), tet(L), tet(X), tet(M), rpsJ] were assessed by PCR amplicon sequencing, and mutations in ramR and tet(A) exhibited high prevalences individually (81%) and in combination (63%). Eight functional ramR mutation profiles reducing tigecycline sensitivity were verified by plasmid complementation of wild-type and mutant ramR. Using a site-specific mutant, the most frequent RamR mutation, A19V (60%), had no significant effect on tigecycline susceptibility or the upregulation of ramA and acrA. Two tet(A) variants with double frameshift mutations, type 1 and type 2, were identified; type 2 tet(A) is novel. A parent strain transformed with a plasmid carrying type 1 or type 2 tet(A) increased the tigecycline MIC by 8-fold or 4-fold, respectively. Synergistic effects were observed in strains harboring no ramR gene and a mutated tet(A), with an 8-fold increase in the tigecycline MIC compared with that in strains harboring only mutated tet(A) being seen. Overall, mutations in the ramR and tet(A) efflux genes constituted the major tigecycline resistance mechanisms among the studied TCRKP isolates. The identification of strains exhibiting the combination of a ramR deficiency and widespread mutated tet(A) is concerning due to the possible dissemination of increased tigecycline resistance in K. pneumoniae.
INTRODUCTION
Tigecycline exhibits expanded-spectrum activity against most clinically significant multidrug-resistant organisms, including carbapenem-resistant Enterobacteriaceae (CRE) (1), with the exception of Proteus and Pseudomonas spp., which are intrinsically tigecycline resistant (2). Carbapenem-resistant Klebsiella pneumoniae (CRKP), which has increased in global prevalence over the last decade (3, 4), is the most commonly encountered CRE species (5). Tigecycline is regarded as a last-resort treatment for CRKP infections. However, the emergence of tigecycline-resistant K. pneumoniae is a serious problem for the treatment of severe CRKP infections (5, 6).
Tigecycline resistance in K. pneumoniae is believed to be primarily mediated by the overexpression of genes encoding the AcrAB-TolC efflux pump, which are controlled by the local repressor acrR as well as global transcriptional activators, such as ramA (1, 2). RamR is a local negative regulator of ramA; loss-of-function mutations occurring in this protein cause significant increases in ramA expression and acrAB-dysregulated tigecycline nonsusceptibility (7). The OqxAB efflux pump, which is also associated with tigecycline resistance, is downregulated by the local repressor oqxR (8). In addition to efflux-mediated resistance mechanisms, an alteration in the tigecycline target site of ribosomal protein S10, which is encoded by the rpsJ gene, has been described (9, 10). There are other known tigecycline resistance mechanisms conferred by Tet proteins, including those associated with the tigecycline-modifying enzyme Tet(X) (11), the mutated efflux pumps of Tet(A) and Tet(L), as well as the ribosomal protective protein Tet(M), that have not yet been identified in K. pneumoniae (12, 13).
Cooperative tigecycline resistance mechanisms have emerged and are a cause for concern (14, 15). The combination of mutated ramR and tet(A) genes was discovered in a tigecycline-resistant Salmonella enterica clinical isolate (15), and mutations in the ramR and rpsJ genes were identified in the whole-genome sequence of a Klebsiella pneumoniae carbapenemase (KPC)-producing K. pneumoniae isolate (9). To date, only a few functional mutations in tigecycline resistance determinant genes have been identified in studies examining CRKP clinical isolates (16). In addition, there have been no functional studies investigating combined tigecycline resistance mechanisms.
In this study, we collected 16 tigecycline- and carbapenem-resistant Klebsiella pneumoniae (TCRKP) clinical isolates to elucidate their tigecycline resistance mechanisms. The two resistance determinants most commonly identified in these isolates, mutated efflux genes ramR and tet(A), were cloned and transformed into a tigecycline-susceptible wild-type (WT) K. pneumoniae clinical isolate to clarify their contributions to tigecycline resistance individually and in combination.
(This study was presented in part as a poster at the 7th International Congress of the Asia-Pacific Society of Infection Control [ASPIC], Taipei, Taiwan, 2015 [17].)
RESULTS
Tigecycline susceptibility testing of TCRKP clinical isolates.The tigecycline MICs and clinical profiles of the 16 TCRKP clinical isolates used in this study are described in Table 1. The strains demonstrating tigecycline MICs of ≥4 μg/ml, which were interpreted to be tigecycline resistant following the European Committee on Antimicrobial Susceptibility Testing (EUCAST) criteria, were selected for this study. Of the 16 isolates, 10 (63%) had tigecycline MICs of ≥8 μg/ml and 4 (25%) had tigecycline MICs of ≥16 μg/ml, including 1 with an MIC of 32 μg/ml.
Tigecycline resistance and related characteristics of TCRKP clinical isolates examined in the present studyd
Identification of tigecycline resistance determinants in TCRKP clinical isolates.Potential tigecycline resistance determinants, specifically, ramR, acrR, oqxR, tet(A), tet(L), tet(X), tet(M), and rpsJ, were identified in the 16 TCRKP isolates by PCR and sequencing (Table 1). Amplicons for ramR were obtained from 14 TCRKP isolates (the exceptions were isolates TR9 and TR12) using published or newly designed primers (see Table S1 in the supplemental material). The genes acrR, oqxR, and rpsJ were detected in all of the isolates; tet(A) was found in 13 (81%) isolates (Table 1).
Thirteen isolates (13/16, 81%) showed nucleotide changes in ramR or the upstream region in comparison to the sequence of the wild-type K. pneumoniae reference strain MGH78578 (GenBank accession number CP000647 ). The 13 distinct mutation profiles for ramR and its upstream sequence are illustrated schematically in Fig. S1 and S2. Six isolates (6/16, 38%) harbored point mutations in the nucleotide sequences upstream of ramR, including three mutations in the RamR binding region (G133A in TR8, C126T in TR13, and T125C in TR3) (Fig. S1). Ten isolates (10/16, 63%) harbored frameshift or substitution mutations in RamR (Fig. S2). The most frequent mutation, an A19V substitution, in RamR was observed in six isolates (6/10, 60%) (Table 1). Among isolates with a ramR mutation, four isolates harbored substitutions in a DNA-binding domain (A19V and L44R) or a dimerization domain (L111R and E113K). Three isolates with A19V substitutions in RamR also harbored another point mutation (L44R in TR11, L111R in TR1, and E113K in TR2). Truncated RamR proteins were produced by four isolates, including TR4 (139 amino acids [aa]), TR7 (134 aa), TR6 (121 aa), and TR10 (68 aa) (Fig. S2).
The acrR, oqxR, and rpsJ sequences with mutations were also compared to the MGH78578 reference sequence (GenBank accession number CP000647 ). Three isolates with acrR mutations (F204L in TR6, A151V in TR16, and an IS5 insertion in TR10) harbored both a ramR mutation and tet(A), whereas another two isolates with oqxR mutations (V130A in TR10 and V124D in TR14) harbored ramR mutations only. Among these strains, no mutations in rpsJ were detected.
In addition, two Tet(A) variants were identified on the basis of the reference sequences for Escherichia coli plasmid RP1 tetracycline resistance determinants (GenBank accession number X00006 ) and designated the type 1 and type 2 Tet(A) variants. The type 1 Tet(A) variant was found in 12 TCRKP isolates and was identical to that found in the E. coli D20-15 RP1 plasmid (GenBank accession number X61367 ). Only one isolate (TR7) exhibited a novel type 2 Tet(A) variant with a mutation profile of I5R, A93V, G151S, S201A, F202S, V203F, and G268A (Table 1). Both Tet(A) variants harbored double frameshift mutations (S201A, F202S, V203F) (Fig. S3). All 16 TCRKP isolates contained either ramR or tet(A) mutations, and 13 (81%) TCRKP isolates harbored mutations in both ramR and tet(A) (Table 1).
Introduction of the ramR A19V mutation into the wild-type parent strain.The predominant mutation A19V was introduced into ramR in the tigecycline-susceptible parent strain NVT2001, and the isogenic mutant NVT2001::ramR-A19V demonstrated a tigecycline MIC of 0.38 μg/ml. The A19V site-specific mutant showed the same MIC values for tigecycline as its parent strain, NVT2001.
Complementation with wild-type ramR in TCRKP clinical isolates.To ascertain the effects of ramR mutations on tigecycline susceptibility, the 13 TCRKP isolates harboring mutations in ramR and/or the region upstream were complemented with wild-type ramR and its upstream region. All TCRKP isolates with the exception of three (TR4, TR6, and TR10) showed less than a 4-fold decrease in their tigecycline MICs after introduction of the wild-type ramR (Table S2).
Complementation with TCRKP ramR in the ramR deletion strain.To further elucidate the contribution of ramR mutations to tigecycline susceptibility in the 13 TCRKP isolates on an isogenic K. pneumoniae strain background, the ramR deletion strain NVT2001 ΔramR was constructed. The ramR deletion strain had a tigecycline MIC of 3 μg/ml, which was an 8-fold increase compared to the tigecycline MIC for parent strain NVT2001. Eight TCRKP ramR-complemented strains showed the same tigecycline MICs as the ramR deletion strain, suggesting loss-of-function mutations in their ramR genes. Conversely, the other five TCRKP ramR-complemented strains had tigecycline MICs that decreased from 3 μg/ml to 0.25 μg/ml, suggesting that their ramR mutations had little or no effect on RamR function. The eight functional ramR mutation profiles consisted of four nonsense mutations causing premature RamR products, three missense mutations leading to combined point substitutions (A19V and L44R, A19V and L111R, and A19V and E113K), and one nucleotide change in the ramR upstream region (C126T) (Table 2).
Tigecycline susceptibility and relative ramA and acrA expression levels of NVT2001 ramR deletion strains harboring wild-type or TCRKP ramR plasmidsc
Relative expression levels of ramA and acrA in TCRKP isolates and ramR mutation strains derived from NVT2001.To speculate on the different genetic changes in ramR, the relative ramA and acrA expression levels of 16 TCRKP isolates were determined by quantitative reverse transcription-PCR (qRT-PCR). Compared to the results for wild-type strain NVT2001, 13 TCRKP isolates overexpressed ramA-specific transcripts (9.15- to 59.05-fold) and 3 overexpressed acrA (2.18- to 4.1-fold). Among those isolates with the A19V mutation (TR1, TR2, TR3, TR7, TR11, and TR16), not all isolates increased their expression level of ramA (0.49- to 59.05-fold) and acrA (0.31- to 4.1-fold). In isolate TR15 with wild-type ramR, its ramA level was high (11.47-fold), while its acrA level remained unaffected (Table S3). To further investigate the direct association between the ramR mutation profiles and the increased levels of expression of ramA and acrA, qRT-PCR analysis was performed with a strain into which A19V was incorporated by site-directed mutagenesis and TCRKP ramR-complemented ramR deletion strains. Increased levels of expression of ramA and acrA in the A19V isogenic mutant compared to those in its NVT2001 parent strain with wild-type ramR were not detected. In addition, in all 8 TCRKP ramR-complemented strains that did not show at least a 2-fold reduction in tigecycline MICs, both their ramA and acrA transcription levels were increased (Table 2).
Introduction of mutated tet(A) into the wild-type or ramR deletion strain.To evaluate the tigecycline resistance effects of mutated tet(A) alone or in combination with a ramR mutation, plasmids harboring type 1 or type 2 mutated tet(A) were transformed individually into the wild-type or ramR deletion strain. The tigecycline MIC of parent strain NVT2001 was 0.38 μg/ml, whereas that of the type 1 tet(A)-complemented strain was 3 μg/ml, an 8-fold increase. The type 2 tet(A)-complemented strain showed an MIC of 1.5 μg/ml, which was a 4-fold increase. The tigecycline MICs of the type 1 and type 2 tet(A)-complemented strains were 24 μg/ml and 12 μg/ml, respectively, which were 63-fold and 32-fold increases, respectively, relative to the tigecycline MIC of the ramR-deficient non-tet(A)-complemented NVT2001 strain. The ramR deficiency increased the tigecycline MICs of the mutants carrying only type 1 or type 2 tet(A) by 8-fold, indicating a synergistic mechanism of resistance for tigecycline (Table 3).
Tigecycline susceptibility of NVT2001 harboring the tet(A) plasmid alone or in combination with a ramR deletion
DISCUSSION
In this study, the tigecycline resistance determinants detected in 16 TCRKP clinical isolates were efflux-altering mutations, including mutations in ramR, acrR, oqxR, and tet(A). Mutations in ramR or tet(A) were specifically identified in all of the examined strains, suggesting that efflux plays a major role in tigecycline resistance in K. pneumoniae. Although various efflux-mediated mechanisms are associated with a low level of resistance to tigecycline (18), it is reasonable to infer that a combination of several efflux mechanisms may exert a synergistic effect on resistance. This study demonstrates, for the first time, that the combination of mutated tet(A) and a deficiency in ramR has synergistic effects on tigecycline resistance in K. pneumoniae.
Mutations in ramR are commonly detected in clinical strains of fluoroquinolone-resistant S. enterica (19), as well as tigecycline-resistant strains of Enterobacter cloacae, Enterobacter aerogenes (20), and K. pneumoniae (21). ramR was not detected in two isolates (TR9 and TR12), suggesting that their ramR genes may have been truncated or excised, as has already been observed in S. enterica clinical strain 75 (19) as well as complete ramR deletion mutants of the K. pneumoniae KPBj1 M3 Lev mutant (22). This study identified six mutations located upstream of ramR, including one nucleotide deletion at position −21 and five point mutations (A9T, T55A, T125C, C126T, and G133A) that have not been previously identified. Two novel point mutations (L44R and L111R) in the RamR protein were identified, in addition to the previously described A19V (18, 23, 24) and E113K (16) mutations.
The A19V RamR substitution mutation was dominant (4/23, 17%) in a study in China (18); it was also the most frequent mutation (6/10, 60%) found in TCRKP strains harboring mutated RamR. Though at least a 2-fold reduction in the tigecycline MICs was observed in TCRKP strains harboring the A19V change in ramR upon complementation with a functional ramR (see Table S2 in the supplemental material), not all strains showed a significant upregulation of ramA and acrA relative to the level of regulation in wild-type strain NVT2001 (Table S3), suggesting that other regulation mechanisms may be involved (8). Our site-directed mutagenesis experiments indicated that the A19V substitution may have no effect on the tigecycline MIC, which is in agreement with the qRT-PCR data showing insignificant ramA upregulation in that A19V isogenic mutant relative to its levels of expression in its parent strain with wild-type ramR (Table S3), as described in K. pneumoniae clinical strain TS238, which harbors this mutation (24).
ramR inactivation appears to be the most prevalent mechanism of tigecycline resistance in K. pneumoniae (9), although several natural functional mutations of ramR, such as I106F, E113K (16), and S41N plus P46E (24), have been shown to affect ramA expression or tigecycline MIC levels in wild-type ramR-complemented strains (23–25). Following complementation with a plasmid containing wild-type ramR, a high proportion of complemented strains (10/13, 77%) showed less than a 4-fold decrease in their tigecycline MICs (Table S2). In TR15, which has no ramR mutation, a 2-fold decrease in the tigecycline MIC value was observed after complementation with wild-type ramR (Table S2). The results suggest the presence of other tigecycline resistance determinants (2). Thus, an alternative complementation assay was carried out in which mutated TCRKP ramR was introduced into an ramR deletion strain. This approach identified eight functional mutations in ramR in an isogenic background by determination of tigecycline susceptibility and the relative transcriptional levels of ramA and acrA (Table 2).
Though all three of these substitutions (C126T in TR13, T125C in TR3, and G133A in TR8) exist in the RamR binding site, only the C126T mutation upstream of ramR in TR13 compromised RamR function, resulting in ramA and acrA overexpression in the TR13 ramR-complemented strain relative to their levels of expression in NVT2001 ΔramR/pramR-WT (Table 2). The result indicated that specific nucleotide mutations in conserved inverted repeat sequences of regulatory elements in the romA-ramA locus, such as a 1-bp substitution (26) or a 2-bp deletion (27), may exert various effects on the DNA-binding affinity of RamR (Fig. S1). In addition, four truncated RamR proteins of different lengths (from 68 to 139 aa) in TR4, TR6, TR7, and TR10 may have lost function due to the absence of the α8-α9 region and the interruption of dimerization (Fig. S2) (28). Therefore, the function of the upstream substitution T55A in TR6 with an incomplete RamR lacking the α8-α9 region could not be ascertained in this study. Moreover, further study will be required to determine whether the three novel combined substitutions, A19V and L44R, A19V and L111R, and A19V and E113K, disrupt the RamR protein structure for dimerization or DNA binding.
Additional mutations detected in the Tet(A) major efflux pump of TCRKP type 1 and type 2 tet(A)-complemented strains were found to confer tigecycline resistance that may have resulted from double frameshift mutations (S201A, F202S, V203F) within the interdomain loop C3 region of the Tet(A) protein, affecting the substrate specificity of tigecycline (13–15), as previously identified in Salmonella species (13). However, in another study, the interdomain loop mutation detected in E. coli isogenic mutants had no significant effect on tigecycline susceptibility (29), which may have been attributable to the expression plasmid system or the plasmid copy numbers. Furthermore, type 1 Tet(A) mutations were detected at a high frequency (75%) in the TCRKP isolates and were found in 1,257 K. pneumoniae sequences in GenBank (described in Materials and Methods), raising concern over the evolution of tigecycline resistance.
Strains complemented with mutated tet(A) and deficient in ramR had tigecycline MICs greater than 8 μg/ml and showed an 8-fold increase in tigecycline MICs compared with those of strains deficient in ramR only, which is higher than the 2-fold increase reported in S. enterica (14). Thus, the coexistence of two mechanisms contributing to low-level resistance may be associated with an increased risk for high-level tigecycline resistance in K. pneumoniae clinical isolates.
Conclusion.According to the findings of this study, mutant ramR and tet(A) tigecycline resistance determinants are prevalent in TCRKP isolates, and their functional gene mutations were identified. Furthermore, an ramR deficiency in the presence of a mutated tet(A) exerts synergistic effects on tigecycline resistance in K. pneumoniae. The combination of a deficiency in ramR and widespread mutated tet(A) should be of concern due to the potential evolution and dissemination of high-level tigecycline resistance in K. pneumoniae.
MATERIALS AND METHODS
Bacterial strains and growth conditions.The bacterial strains used in this study are listed in Table 4. TCRKP clinical isolates were selected from a nationwide surveillance study of carbapenem-nonsusceptible K. pneumoniae isolates in Taiwan (30). The TCRKP isolates were collected from patients between January 2012 and December 2014. With the exception of isogenic mutants, the strains were cultured in Luria-Bertani (LB) or brain heart infusion (BHI) medium at 37°C. Transformants harboring antibiotic resistance-conferring plasmids were grown in medium supplemented with 50 μg/ml kanamycin (Sigma-Aldrich, St. Louis, MO) or 100 μg/ml amikacin (Sigma-Aldrich).
Strains used in this study
Antimicrobial susceptibility testing.Strains were tested for susceptibility to tigecycline by performing an Etest (bioMérieux, Marcy l'Etoile, France) following the manufacturer's instructions, and Escherichia coli ATCC 25922 was used as the control strain. Tigecycline resistance was defined as a MIC of at least 4 μg/ml, on the basis of the European Committee on Antimicrobial Susceptibility Testing (EUCAST) criteria (≤1 μg/ml for susceptible, 2 μg/ml for intermediate, ≥4 μg/ml for resistant) (31).
Detection and sequence analysis of tigecycline resistance determinant genes.The TCRKP clinical isolates were screened for the tigecycline resistance determinants ramR, acrR, oqxR, rpsJ, tet(A), tet(L), tet(M), and tet(X) by performing PCR with gene-specific primers (see Table S1 in the supplemental material). The PCR products of the open reading frames (ORFs) and the flanking regions of the target genes were purified with a QIAquick PCR purification kit (Qiagen, GmbH, Germany) and were sent to Genomics Corporation (New Taipei City, Taiwan) for sequencing. Mutations were characterized by comparing the sequences with those of wild-type reference strains [K. pneumoniae MGH78578 for ramR, acrR, and oqxR (GenBank accession number CP000647 ) and E. coli plasmid RP1 for tet(A) (GenBank accession number X00006 )]. The homology and identity of Tet(A) sequences were analyzed using the BLAST online tool at http://www.ncbi.nlm.nih.gov/BLAST/ (accessed 21 January 2017) (32).
Construction of ramR deletion mutants.The suicide vector pUT-KB (33), which contains an R6K origin of replication, mobRP4 mobile transfer element, kanamycin resistance cassette, and sacB counterselection marker, was used to construct plasmids for ramR deletion and site-directed mutagenesis. The nucleotide sequences of ramR and its flanking regions in K. pneumoniae NVT2001 were identical to those of the wild-type reference strain MGH78578 (GenBank accession number CP000647 ). NVT2001 was used as the parent strain for the construction of the ramR deletion and site-specific mutants. Deletion of ramR in the NVT2001 wild-type strain was achieved via in-frame deletion mutagenesis as previously described (34). Briefly, two DNA fragments derived from the regions upstream and downstream of ramR were PCR amplified using the ramR-Del-AF/ramR-Del-AR and ramR-Del-BF/ramR-Del-BR primer sets, respectively (Table S1). The two amplicons were spliced to create a ramR deletion fragment by overlap extension PCR. The spliced fragment was digested with the ApaI and XbaI restriction enzymes and then cloned into pUT-KB, which was similarly digested, resulting in plasmid pramR-Del. For homologous recombination, pramR-Del was transformed into E. coli S17-1 λpir using the heat shock method and was then mobilized into NVT2001 via conjugation. The transconjugants were screened on brilliant green plates containing inositol-nitrate-deoxycholate (BIND) supplemented with kanamycin (50 μg/ml) to select K. pneumoniae strains harboring pramR-Del, while the growth of non-K. pneumoniae contaminants was effectively suppressed on the BIND plates (33). The selected K. pneumoniae strains were incubated for 6 h in 20 ml of BHI medium in the absence of kanamycin, and then the inocula were spread onto LB medium plates containing 10% sucrose. After double-crossover events had occurred, sucrose-resistant and kanamycin-sensitive colonies were selected and screened via PCR and sequencing to identify and verify the mutants with in-frame deletions.
Construction of site-specific mutants.The site-directed mutagenesis of ramR was accomplished using a QuikChange II site-directed mutagenesis kit (Agilent, CA, USA). Wild-type NVT2001 ramR and its flanking regions were PCR amplified with primers ramR-Sdm-WT-F and ramR-Sdm-WT-R (Table S1), digested with ApaI and XbaI, and then cloned into pUT-KB, which was similarly digested, generating plasmid pramR-Sdm-WT. A single point mutation (A19V) was introduced into ramR on pramR-Sdm-WT following the manufacturer's protocol using primers ramR-A19V-F and ramR-A19V-R (Table S1), resulting in plasmid pramR-A19V. Transformation, conjugation, and screening for double-crossover NVT2001::ramR-A19V mutants were performed similarly to the procedures for NVT2001 ΔramR, using plasmid pramR-A19V. The presence of the ramR A19V mutation in selected mutants was confirmed by DNA sequencing.
ramR complementation.Plasmid complementation studies with wild-type or naturally mutated ramR were carried out to validate the functional mutations in ramR or the upstream nucleotide sequence. Although multidrug resistance was observed in our TCRKP clinical isolates, these strains remained susceptible to amikacin. The amikacin resistance gene armA and its flanking region in an amikacin-resistant extended-spectrum β-lactamase-producing K. pneumoniae strain obtained from the Taiwan Surveillance of Antimicrobial Resistance Project (35) were together amplified with primers armA-F and armA-R (Table S1) and then cloned into the pACYC184 vector using BamHI, generating plasmid pA. Wild-type NVT2001 ramR and its flanking regions were PCR amplified with primers ramR-Com-WT-F and ramR-Com-WT-R (Table S1) and cloned into pA using EcoRV, resulting in plasmid pAramR-Com-WT. pAramR-Com-WT was transformed into the TCRKP clinical isolates individually to restore their RamR function. Naturally mutant ramR genes and their flanking regions were PCR amplified from the TCRKP isolates with primers ramR-Com-TR-F and ramR-Com-TR-R (Table S1), and each product was individually cloned into pACYC177 using PstI, generating plasmids pramR-TR1, -2, -3, -4, -5, -6, -7, -8, -10, -11, -13, -14, and -16. Each pramR-TR plasmid was transformed into the NVT2001 ΔramR strain to restore the RamR function. Strains NVT2001/pAramR-WT and NVT2001 ΔramR/pramR-TR-WT were used as controls to establish the basal levels of tigecycline susceptibility for comparison with the levels of tigecycline susceptibility of the ramR-complemented isogenic mutants.
qRT-PCR.To assess the gene expression levels of ramA and acrA, quantitative reverse transcription-PCR (qRT-PCR) was performed using a Fast SYBR green PCR master mix (Applied Biosystems) and gene-specific primers (Table S1) in a 7500 Fast real-time PCR system (Applied Biosystems), as previously described (36). Total RNA of the mid-log-phase K. pneumoniae inoculum was extracted using an RNeasy kit (Qiagen). cDNA was synthesized from RNase-free DNase I (Qiagen)-treated total RNA using SuperScript III reverse transcriptase (Invitrogen). All qRT-PCR experiments were performed three times, with 23S rRNA being used as an internal control (37).
tet(A) plasmid transformation.Plasmids expressing type 1 and type 2 tet(A) variants from the TCRKP isolates were transformed into tet(A)-deficient strains NVT2001 and NVT2001 ΔramR to determine the synergism between dysfunctional RamR and mutated tet(A). Mutated tet(A) and the surrounding regions from isolates TR1 and TR7 were PCR amplified using the primers TetA-Add-F and TetA-Add-R (Table S1) and were cloned individually into pACYC177 with PstI, resulting in plasmids ptetA-TR1 and ptetA-TR7, respectively.
ACKNOWLEDGMENTS
We are grateful to Jen-Chang Chang, Chin-Lin Huang, and Ya-Wen Hsiao for their excellent technical assistance. We thank Shih-Feng Tsai for providing strain MGH78578. We acknowledge the technical services (real-time PCR system support) provided by the Instrument Center of the National Defense Medical Center.
We have no conflicts of interest to declare.
This study was supported by research grants from the Ministry of Science and Technology (MOST103-2314-B-016-020-MY3, MOST104-2314-B-016-019-MY3, MOST105-2314-B-016-050, MOST105-2320-B-400-016), the Tri-Service General Hospital (TSGH-C106-095, TSGH-C106-097), the Medical Affairs Bureau, Ministry of National Defense, Republic of China (MAB-106-080), and the National Health Research Institutes.
FOOTNOTES
- Received 24 February 2017.
- Returned for modification 14 March 2017.
- Accepted 15 May 2017.
- Accepted manuscript posted online 22 May 2017.
Supplemental material for this article may be found at https://doi.org/10.1128/AAC.00391-17 .
- Copyright © 2017 American Society for Microbiology.
