|Home | About | Journals | Submit | Contact Us | Français|
The combination of β-lactams and β-lactamase inhibitors has been shown to have potent in vitro activity against multidrug-resistant tuberculosis (MDR-TB) isolates. In order to identify the most potent β-lactam–β-lactamase inhibitor combination against MDR-TB, we selected nine β-lactams and three β-lactamase inhibitors, which belong to different subgroups. A total of 121 MDR-TB strains were included in this study. Out of the β-lactams used herein, biapenem was the most effective against MDR-TB and had an MIC50 value of 8 μg/ml. However, after the addition of clavulanate or sulbactam, meropenem exhibited the most potent anti-MDR-TB activity with an MIC50 value of 4 μg/ml. For meropenem, 76 (62.8%), 41 (33.9%), and 22 (18.2%) of the 121 MDR-TB strains were subjected to a synergistic effect when the drug was combined with sulbactam, tazobactam, or clavulanate, respectively. Further statistical analysis revealed that significantly more strains experienced a synergistic effect when exposed to the combination of meropenem with sulbactam than when exposed to meropenem in combination with tazobactam or clavulanate, respectively (P < 0.01). In addition, a total of 10.7% (13/121) of isolates harbored mutations in the blaC gene, with two different nucleotide substitutions: AGT333AGG and ATC786ATT. For the strains with a Ser111Arg substitution in BlaC, a better synergistic effect was observed in the meropenem-clavulanate and in the amoxicillin-clavulanate combinations than that in a synonymous single nucleotide polymorphism (SNP) group. In conclusion, our findings demonstrate that the combination of meropenem and sulbactam shows the most potent activity against MDR-TB isolates. In addition, the Ser111Arg substitution of BlaC may be associated with an increased susceptibility of MDR-TB isolates to meropenem and amoxicillin in the presence of clavulanate.
The emergence of multidrug-resistant tuberculosis (MDR-TB), which is resistant to at least isoniazid and rifampin (RIF); pre-extensively drug-resistant tuberculosis (pre-XDR-TB), which is additionally resistant to either fluoroquinolone (FQ) or at least one of the three injectable second-line drugs; XDR-TB, which is additionally resistant to any FQ and at least one of the three injectable second-line drugs; and totally drug-resistant TB (TDR-TB), which is resistant to all first- and second-line drugs tested, is the major obstacle to global tuberculosis control (1,–4). According to the estimation of the World Health Organization (WHO), there were 480,000 new cases of MDR-TB and 210,000 deaths from MDR-TB globally in 2013 (1). Due to the resistance of MDR-TB to the most important backbone anti-TB drugs, MDR-TB patients have a higher risk for treatment failure than non-MDR-TB patients, and the treatment success rate for XDR-TB is even lower (5, 6). Hence, the global epidemic of MDR-TB has highlighted the urgent need for new and effective therapeutic options to overcome the shortcomings of current chemotherapy regimens, especially for XDR-TB (7, 8).
The β-lactam class of antibiotics has exhibited good bacteriostatic activity against Gram-negative and Gram-positive bacterial infection (9), while Mycobacterium tuberculosis is naturally resistant to most of these antibiotics in vitro (10). The intrinsic β-lactam resistance of M. tuberculosis has been attributed to the presence of the highly active β-lactamase BlaC (11). BlaC belongs to Ambler class A β-lactamases, which are susceptible to the β-lactamase inhibitors widely used in the clinic, including clavulanate (CLAV), sulbactam (SUB), and tazobactam (TAZ) (12). Therefore, resistance can be overcome by the combination of β-lactams and β-lactamase inhibitors, and several previous studies have demonstrated that the use of amoxicillin-clavulanate was active in vitro and had early bactericidal activity in patients with MDR-TB (10, 13,–15). As the clinical evidence for amoxicillin-clavulanate for the treatment of MDR-TB and XDR-TB is limited, this combination is considered a group 5 antituberculous drug (15).
Recently, another β-lactam–β-lactamase inhibitor combination, meropenem-clavulanate, also showed high antimycobacterial activity in vivo against MDR-TB and XDR-TB strains of M. tuberculosis (16). Several clinical trials have provided preliminary evidence of the effectiveness and safety of meropenem-clavulanate in the treatment of MDR-TB and XDR-TB (16, 17). The promising bactericidal activity of the meropenem-clavulanate combination suggests that other β-lactam–β-lactamase inhibitor combinations are suitable for the clinical management of MDR-TB and XDR-TB patients. With this notion in mind, we selected nine β-lactams and three β-lactamase inhibitors, which belong to different subgroups in this study. The broth dilution method based on Middlebrook 7H9 broth was used to evaluate the in vitro activity of β-lactams in combination with β-lactamase inhibitors against MDR-TB isolates. Our aim was to provide more potential β-lactam–β-lactamase inhibitor options for the clinical treatment of MDR-TB cases.
All of the MDR-TB strains were isolated from smear-positive tuberculosis patients registered at local TB dispensaries in Chongqing as previously reported (18). Drug susceptibility for isoniazid, rifampin, ethambutol, streptomycin, kanamycin, and ofloxacin was tested with the proportion method endorsed by WHO (3). Mycobacterium species identification was performed by growth test on a medium containing p-nitrobenzoic acid (PNB) and 2-thiophenecarboxylic acid hydrazide (TCH). Prior to performing phenotypic drug susceptibility testing, the strains were recovered on Lowenstein-Jensen (LJ) medium for 4 weeks at 37°C.
A total of nine β-lactams and three β-lactamase inhibitors were selected to perform the MIC experiments. The nine β-lactams were from three different classes: penicillin, cephalosporin and cephamycin, and carbapenem. Out of these, amoxicillin (AMX) belonged to the penicillin group. Five other β-lactams represented different generations of cephalosporin and cephamycin group. In addition, the most frequently used carbapenems in clinical practice, including imipenem (IMI), meropenem (MERO), and biapenem (BIA), were enrolled in this study. For the β-lactamase inhibitors, clavulanate (CLAV), tazobactam (TAZ), and sulbactam (SUB) were included in the drug panel for each strain (Table 1). All of the reagents were purchased from Sigma-Aldrich Chemical Company (St. Louis, MO, USA).
The MICs of MDR-TB isolates against β-lactams were determined by using a microplate alamarBlue assay as previously reported (19). The inoculum was prepared from the 4-week-old cultures grown in LJ medium. The turbidity of the cultures was adjusted to a 1.0 McFarland standard. Then, the 1.0-McFarland standard cell suspension was diluted to 1:20 in Middlebrook 7H9 broth containing 10% oleic acid-albumin-dextrose-catalase (OADC), and 100 μl of this inoculum was added to the wells of the 96-well plate. Following 7 days of incubation, 70 μl of alamarBlue solution was added to each well, was incubated for 24 h at 37°C, and was assessed for color development. A change from blue to pink indicated bacterial growth. MIC was defined as the lowest concentration of antibiotic that prevented the color change from blue to pink. Final drug concentrations were 0.125 to 512 μg/ml. M. tuberculosis H37Rv (ATCC 27249) was tested in each round as a quality control strain. All experiments were done in duplicate to access reproducibility.
The ratio of the MIC of the test isolate against a β-lactam in combination with a β-lactamase inhibitor compared to a β-lactam without a β-lactamase inhibitor was calculated. A synergistic effect was declared if the ratio was less than 0.25, indicating that the MIC of the β-lactam showed at least a 4-fold decrease in combination with a β-lactamase inhibitor.
Genomic DNA was extracted from freshly cultured bacteria as previously reported (2). The bacterial cells were transferred into a 1.5-ml microcentrifuge tube with 500 μl Tris-EDTA (TE) buffer. This was followed by centrifugation at 13,000 rpm for 2 min. After the supernatant was discarded, the pellet was resuspended in 500 μl TE buffer and then heated in a 95°C water bath for 1 h. The cellular debris was separated by centrifugation, and the DNA in the supernatant was used for the PCR amplification.
The blaC, dacB2, and ldtMt1 genes were amplified by PCR, respectively. The primers used in this procedure are listed in Table S1 in the supplemental material. The genomic DNA prepared above was used as the template to perform PCR amplification. The PCR mixture was prepared in a volume of 50 μl containing 25 μl 2× PCR mixture, 5 μl of DNA template, and 0.2 μM (each) primer set. PCR products were purified using a Mag-Bind PCR purification kit (CWBio, Beijing, China) and were then sent to Qingke Company (Beijing, China) for DNA sequencing services. DNA sequences were aligned with the homologous sequences of the reference M. tuberculosis H37Rv strains using multiple sequence alignments (http://www.ncbi.nlm.nih.gov/BLAST).
A chi-square test was used to compare the proportion of synergistic effect among different β-lactam–β-lactamase inhibitor combinations. Statistical analysis was performed in SPSS 11.5 (SPSS Inc., USA). Differences with a P value of ≤0.05 were considered statistically significant.
In order to determine the optimum concentrations of different β-lactamase inhibitors, five MDR-TB isolates were enrolled to evaluate the MICs for β-lactams alone or in the presence of β-lactamase inhibitors in fixed concentrations of 1.25 μg/ml, 2.5 μg/ml, 5 μg/ml, and 10 μg/ml. As shown in Table 2, the MIC values of MDR-TB isolates against amoxicillin, meropenem, and biapenem were significantly decreased along with the increased concentration of β-lactamase inhibitors. However, between the 5-μg/ml inhibitor group and the 10-μg/ml group, there was no difference in MIC of the β-lactams tested. In addition, the MICs of MDR-TB isolates against imipenem decreased 4-fold at most owing to the presence of β-lactamase inhibitors. For cephalosporins, we observed that they showed no anti-TB activity against MDR-TB with MIC values of 32 to 256 μg/ml. Even in the presence of any β-lactamase inhibitor at 10 μg/ml, the MICs of five cephalosporins did not decrease with any combination of β-lactamase inhibitors. In light of our findings, cephalosporins were excluded in further analysis, and 5 μg/ml of β-lactamase inhibitors were used in the latter panel testing.
A total of 121 MDR-TB strains were included in this study. We first evaluated MICs for β-lactams alone or in combination with clavulanate (5 μg/ml) against MDR-TB strains. Out of the four β-lactams used herein, biapenem showed the best effect against MDR-TB, with MIC50 and MIC90 values of 8 μg/ml and 32 μg/ml, respectively (Table 3). However, after the addition of clavulanate, meropenem exhibited the most potent anti-MDR-TB activity with an MIC50 value of 4 μg/ml. We also analyzed the increased susceptibility of MDR-TB isolates to β-lactams due to the presence of clavulanate. As shown in Table 3, meropenem and amoxicillin showed the best synergy in combination with clavulanate. In contrast, the presence of clavulanate seemed to have no effect on the MIC values of MDR-TB isolates against imipenem.
When adding tazobactam to β-lactams, the MIC values of strains against amoxicillin presented a striking difference compared with those of β-lactams alone. The MIC values of 95 isolates (78.5%) against amoxicillin were decreased by 75% between amoxicillin alone and amoxicillin plus tazobactam. In addition, biapenem plus tazobactam showed a favorable synergistic effect compared with that of biapenem used alone, and the MIC50 of strains against biapenem decreased from 8 μg/ml to 4 μg/ml. Similar to clavulanate, the supplement of tazobactam barely affected the MIC values of MDR-TB isolates against imipenem (Table 4).
Overall, the MIC50 values of the combination of sulbactam and β-lactams were lower than those of the combinations of the two other β-lactamase inhibitors and the β-lactams. When sulbactam was present, meropenem and biapenem were the most active against MDR-TB isolates, and each had an MIC50 value of 4 μg/ml. Moreover, the MIC values of amoxicillin fell by an average of 12.6 dilutions due to the addition of sulbactam, resulting in the MIC50 values of amoxicillin decreasing from 128 μg/ml to 8 μg/ml (Table 5).
We further compared the effect of different β-lactamase inhibitors on the MIC values of β-lactams. As listed in Table 6, for amoxicillin, the proportion of the synergistic effect against strains in combination with sulbactam (94.2%) was significantly higher than that of clavulanate (38.8%) and tazobactam (78.5%) (P < 0.01). For imipenem, sulbactam served as the most potent β-lactamase inhibitor, while only 23.1% of strains were subjected to a synergistic effect when adding the sulbactam. For meropenem, 76 (62.8%), 41 (33.9%), and 22 (18.2%) out of 121 MDR-TB strains were subjected to the synergistic effect when combined with sulbactam, tazobactam, and clavulanate, respectively. Further statistical analysis revealed that the percentage of synergistic effect among the meropenem-sulbactam group was significantly higher than that among the meropenem-tazobactam and meropenem-clavulanate groups, respectively (P < 0.01). In addition, the proportion of synergistic effect among the meropenem-clavulanate group was also higher than that of the meropenem-tazobactam group (P < 0.01). Similar to other β-lactams, biapenem-sulbactam (57.9%, 70/121) showed better synergistic effect compared with that of the other two combinations.
To investigate the genetic mutations associated with the synergistic effect, we examined three potential target genes in all MDR isolates, blaC, dacB2, and ldtMt1. BlaC is the β-lactamase in M. tuberculosis (11) while d,d-carboxypeptidase DacB2 and 3→3 cross-linking transpeptidase LdtMt1 serve as the target of carbapenems (20, 21). As shown in Table 7, a total of 10.7% (13/121) of isolates harbored mutations in the blaC gene, with two different nucleotide acid substitutions, AGT333AGG and ATC786ATT. ATC786ATT was the most frequent mutation; this genetic mutation resulted in no amino acid substitution (Ile262Ile), indicating that this mutant type serves as a synonymous single nucleotide polymorphism (SNP). We also identified that 4 (3.3%) out of 121 isolates carried the Ser111Arg (AGT333AGG) mutation. In addition, sequencing of dacB2 revealed that 8 (6.6%) isolates had a mutation at position 659, resulting in an amino acid substitution from Leu to Gln. We analyzed the effect of genetic mutation on the synergy between β-lactams and β-lactamase inhibitors. When setting the synonymous mutation ATC786ATT as the control group, we found that the Leu220Gln mutation in DacB2 did not affect the MIC values of synergistic effect, indicating that this acid substitution does not play a role in mediating β-lactam resistance in M. tuberculosis. For the strains with Ser111Arg substitution in BlaC, better synergistic effects were observed in the MERO-CLAV and AMX-CLAV combinations compared with those in a synonymous SNP group, while the other combinations did not show a difference between these two groups. In addition, no mutation was observed in the ldtMt1 gene among any of the MDR-TB isolates.
β-Lactam–β-lactamase inhibitor combinations attract more attention due to their effective in vivo and in vitro activity against MDR-TB strains (9, 15, 16). The present study showed that different β-lactam–β-lactamase inhibitor combinations vary in their anti-MDR-TB efficiency in vitro. Out of three β-lactam families, carbapenem alone or in the presence of β-lactamase inhibitors showed the most potent activity against MDR-TB isolates. Consistent with our findings, a recent article by Horita et al. demonstrates that clinical TB isolates are more susceptible to carbapenem plus β-lactamase inhibitors when considering the MIC range and MIC90 (22). Compared with penicillin and cephalosporins, carbapenems are more resistant to hydrolysis by β-lactamase, which may result in further improved bactericidal activity when combined with a β-lactamase inhibitor.
In addition, different carbapenems exhibited various anti-MDR-TB efficiencies in vitro. Overall, the MIC values of meropenem or biapenem alone were lower than those of imipenem. The difference in anti-MDR-TB activity among them may be explained by the mechanism of action in relation to the Ambler class A β-lactamase BlaC. A recent biochemical study demonstrated that meropenem is more resistant to the hydrolysis of BlaC than imipenem due to lower catalytic efficiency (kcat/Km), resulting in M. tuberculosis being relatively less sensitive to imipenem (11). In addition, carbapenems can inhibit the biosynthesis of peptidoglycan by blocking the 3→3 (diaminopimelic acid [DAP]-DAP) and/or 4→3 (d-Ala-DAP) cross-links (20, 23). It has been shown that meropenem binds to several targets involved in peptidoglycan synthesis, including d,d-carboxypeptidases (Dacs) (part of the classic penicillin-binding proteins) and l,d-transpeptidases (Ldts), the latter being involved in the synthesis of alternatively cross-linked peptidoglycan (20). Hence, the addition of meropenem in the culture medium may exhibit better efficiency in the inhibition of biosynthesis of the cell wall, thereby enhancing the permeability of the bacterial envelope. This slow penetration has been considered a major determinant of M. tuberculosis resistance to β-lactam antibiotics. Hence, the increased permeability of M. tuberculosis due to the addition of meropenem may increase the susceptibility of M. tuberculosis against carbapenem and other antibiotics. In agreement with our hypothesis, the synergy between meropenem and amoxicillin-clavulanate has been pronounced in recent literature (10).
Inhibitor combinations provide one strategy to overcome β-lactamase-mediated resistance (24, 25). Previous studies have demonstrated that the β-lactamase inhibitors show species-specific differences in activity against microorganisms, which is ascribable to the classification and the three-dimensional structure of β-lactamase in vivo (25, 26). In this study, we found that sulbactam served as the more effective β-lactamase inhibitor against MDR-TB isolates than clavulanate and tazobactam when in combination with β-lactams. Clavulanate is a mechanism-based inhibitor to class A β-lactamase with high potency (typically Ki is <0.1 mM), while a previous article by Horita and colleagues reveals that sulbactam harbors higher potency than clavulanate against M. tuberculosis when in combination with β-lactams, which confirms our findings (22). On one hand, the MIC values of this study were determined by the broth dilution method, which requires 7 days to reach the endpoint (19). The half-life of clavulanate (1.4 days) is significantly lower than that of sulbactam (>6 days) (27). Therefore, the relative instability of clavulanate will result in the loss of active drug during incubation, thereby limiting the anti-TB activity of synergetic β-lactams. One the other hand, efficient penetration of the β-lactamase inhibitor through the cell envelope is essential to enhance the antibacterial activity of β-lactams in vivo (28). Considering that clavulanate is more hydrophilic than sulbactam, we hypothesized that clavulanate might penetrate the thick layer of lipid on the outer surface of M. tuberculosis with greater difficulty, resulting in an unsatisfactory inhibitory effect. Further study is warranted to investigate the potential role of in vivo penetration of different β-lactamase inhibitors on its efficacy in combination with β-lactams for combating TB.
We further analyzed the genetic mutations associated with the synergistic effect between β-lactams and β-lactamase inhibitors. Interestingly, one nonsynonymous mutation in the blaC gene attracts our attention because it may be associated with an increased susceptibility of MDR-TB isolates to meropenem and amoxicillin in the presence of clavulanate. The binding affinity between a ligand and its receptor protein is majorly dependent on the structural complement and intermolecular interaction (29). The protein structure analysis of the wild type reveals that the serine at position 111 is located near the binding motif of BlaC with β-lactams–β-lactamase inhibitors, including the amino acids R220, A244, S130, and T237 of M. tuberculosis BlaC. Therefore, we hypothesized that the substitution of Ser111 by Arg111 introduced conformational changes in BlaC, thereby influencing inhibitor activity. However, there is no direct in vitro biochemical evidence that Ser111 is involved in substrate binding as part of the carboxylate-binding region. Further molecular analysis will provide new insight on the active sites of BlaC in M. tuberculosis. In contrast, the Ser111Arg substitution in BlaC seems to have no role in the susceptibility of MDR-TB strains against imipenem due to the supplement of clavulanate. One possible explanation is that imipenem is particularly unstable, with a half-life of 0.4 days (27), which may be responsible for the loss of synergy between imipenem and clavulanate. In addition to the amino acid substitution in BlaC, we also identified that the Leu220Gln mutation in DacB2 did not affect the MIC values of synergistic effect. DacB2 has been considered an important target for meropenem, which serves as part of the penicillin-binding protein (PBP) enzyme machinery. Chambers and colleagues have demonstrated that β-lactams have to bind to at least three of four identified PBPs to be effective. Hence, the complicated mechanism may serve as a major determinant of M. tuberculosis resistance to β-lactam antibiotics (27).
In conclusion, our findings demonstrate that the combination of meropenem and sulbactam shows the most potent activity against MDR-TB isolates. In contrast, the MDR-TB isolates are resistant to imipenem and cephalosporins alone or in combination with β-lactamase inhibitors in vitro. In addition, the Ser111Arg substitution in BlaC may be associated with an increased susceptibility of MDR-TB isolates to meropenem and amoxicillin owing to the presence of clavulanate. Further investigation is urgently needed to evaluate the potential uses of meropenem-sulbactam against MDR-TB in clinical practice.
We thank the staff from the National Tuberculosis Reference Laboratory (Beijing, China) for their technical support.
Supplemental material for this article may be found at http://dx.doi.org/10.1128/AAC.01035-15.