Antibiotic testing and screening system

This application claims priority to U.S. Provisional Application No. 62/463,168, filed on Feb. 24, 2017, the content of which is hereby incorporated by reference in its entirety.

The present invention relates to a platform technology for evaluating the antimicrobial activity of a known antibiotic against specified resistance mechanisms and also for screening for new antimicrobial agents.

The emergence of antibiotic resistance has become a major clinical and public health threat worldwide. 1.2 According to a recent report from the United Kingdom, if antimicrobial resistance continues to increase, 10 million people will die due to it annually by 2050, a mortality rate greater than that of cancer. 1 Correct use of antibiotics and the development of novel antibiotics are both necessary to overcome this crisis.

World Health Organization (WHO) reported an extremely high resistant rates in Klebsiella pneumoniae, Escherichia coli and Staphylococcus aureus , which are common causes of hospital and community infections. 4 K. pneumoniae and E. coli both belong to Enterobacteriaceae and have similar antibiotic resistance mechanisms. 5 Understanding the antibiotic effectiveness against each resistant mechanism can help us to develop and use antibiotics. Traditionally, a large number of clinical drug-resistance isolates are collected and used for testing and screening antibiotics before going into clinical trials. 6 However, the resistance mechanisms in clinical isolates usually are complex and not fully identified. It is difficult to verify which mechanism is actually involved or plays the major role in conferring the drug resistance. Without the certain information, it is also difficult to modify the antibiotic candidates for improving its antimicrobial activity in an efficient manner.

The present design provides a platform technology for testing, screening, selecting and evaluating antibiotics by using genetically engineered strains with identified, individual or combined, resistance mechanisms, prepared from a multidrug (or preferably fully) susceptible clinical isolate.

Specifically, the present invention provides a platform technology for screening for antimicrobial agents by using genetic engineering bacterial strains, wherein these bacterial strains before genetic engineering are in nature highly susceptible to antibiotic treatment (preferably susceptible to multiple antibiotic treatment) and after genetic engineering it is given one or more mutations or antibiotic resistance genes based on the major types of drug-resistant mechanisms, selected form the group consisting of (i) decrease antibiotic permeability by loss of outer membrane proteins, (ii) pump out the antibiotics by overexpression of efflux pumps, (iii) eliminates or reduces binding of antibiotic by modification of antibiotic target or by acquirement of antibiotic-resistant target, and (iv) inactivate antibiotic by enzymatic cleavage or modification, and any combinations thereof. Since the drug-resistant mechanism given in the genetic engineering bacterial strains is known and identified, once an agent is cultured with the genetic engineering bacterial strain and determined to have an antimicrobial activity against it, we can thus figure out that the agent is specifically active against the drug-resistant mechanism given in the genetic engineering bacterial strains. Accordingly, the platform technology of the invention is feasible and efficient to identify an antimicrobial agent targeting a particular drug-resistant mechanism as desired. In addition, the platform technology of the invention, as compared with a conventional screening method using clinical resistant isolates, is less labor intensive and also is workable for carrying out a large-scale screening for an antimicrobial agent, especially for an antimicrobial agent active against multiple drug-resistant mechanisms, and moreover is meaningful to apply the selected antimicrobial agent to subsequent drug modification.

In particular, the present invention provides a method for screening for an antimicrobial agent, comprising:

(1) providing a wild-type (WT) susceptible bacterial strain which has a WT-referenced resistance level to a reference antibiotic;

(2) providing a first genetic engineering bacterial strain generated from the WT susceptible bacterial strain which is given a first mutation or antibiotic resistance gene (ET1) via a genetic engineering manner to confer drug resistance to the reference antibiotic, exhibiting a ET1-referenced resistance level to the reference antibiotic, wherein the ET1-referenced resistance level is higher than the WT-referenced resistance level;

(3) culturing the wild-type susceptible bacterial strain in the presence of a first test agent and analyzing the first test agent for its activity against the wild-type susceptible bacterial strain to obtain a WT-Test1 resistance level;

(4) culturing the first engineered bacterial strain in the presence of the first test agent and analyzing the first test agent for its activity against the first engineered bacterial strain to obtain a ET1-Test1 resistance level;

(5) obtaining a ET1-Test1 ratio of the ET1-Test1 resistance level to the WT-Test1 resistance level; and

(6) determining if the first test agent is a potentially effective antimicrobial agent against the drug resistance associated with ET1 in the first engineered bacterial strain based on the ET1-Test1 ratio.

In some embodiments, the ET1-referenced resistance level is higher than the WT-referenced resistance level by 1-fold, 2-fold, 3-fold, or 4-fold or more.

In some embodiments, the wild-type susceptible bacterial strain is susceptible to multiple antibiotics, selected from the group consisting of β-lactams, quinolones, aminoglycosides, tetracyclines, folate pathway inhibitors, polymyxins, phenicols, fosomycins, nitrofurans and any combinations thereof.

In some embodiments, the first test agent is determined as a potentially effective antimicrobial agent against the drug resistance associated with ET1 if the ET1-Test1 ratio is no more than 4 (e.g. 4 or less, 3 or less, 2 or less, 1 or less).

In more particular, the method of the invention further comprises:

(7) (17) culturing the wild-type susceptible strain in the presence of a second test agent and analyzing the second test agent for its activity against the wild-type susceptible strain to obtain a WT-Test2 resistance level;

(8) (18) culturing the first engineered bacterial strain in the presence of the second test agent and analyzing the second test agent for its activity against the first engineered bacterial strain to obtain a ET1-Test2 resistance level;

(9) (19) obtaining a ET1-Test2 ratio of the ET1-Test2 resistance level to the WT-Test2 resistance level; and

(10) (20) determining if the second test agent is a potentially effective antimicrobial agent against the drug resistance associated with ET1 in the first engineered bacterial strain based on the ET1-Test2 ratio.

In some embodiments, the second test agent is determined as a potentially effective antimicrobial agent against the drug resistance associated with ET1 if the ET1-Test2 ratio is no more than 4 (e.g. 4 or less, 3 or less, 2 or less, 1 or less).

In some embodiments, the second test agent is determined as a potentially effective antimicrobial agent against the drug resistance associated with ET1 if the ET1-Test2 resistance level, which is expressed by minimum inhibitory concentration (MIC), is no more than 10 μg/ml, 5 μg/ml, 2.5 μg/ml, 1 μg/ml or 0.5 μg/ml.

In more particular, the method of the invention further comprises:

(11) ranking the first test agent and the second test agent for a preferable antimicrobial activity according to the ET1-Test1 ratio and the ET1-Test2 ratio, the lower the ratio, the higher the ranking.

In more particular, the method of the invention further comprises:

(12) providing a second engineered bacterial strain generated from the WT susceptible bacterial strain which is given a second mutation or antibiotic resistance gene (ET2) in a genetic engineering manner to confer drug resistance to the reference antibiotic, exhibiting a ET2-referenced resistance level to the reference antibiotic, wherein the ET2-referenced resistance level is higher than the WT-referenced resistance level;

(13) culturing the second engineered bacterial strain in the presence of the first test agent and analyzing the first test agent for its activity against the second engineered bacterial strain to obtain a ET2-Test1 resistance level;

(14) obtaining a ET2-Test1 ratio of the ET2-Test1 resistance level to the WT-Test1 resistance level;

(15) determining if the first test agent is a potentially effective antimicrobial agent against the drug resistance associated with ET2 in the second engineered bacterial strain based on the ET2-Test1 ratio.

In some embodiments, the first test agent is determined as a potentially effective antimicrobial agent against the drug resistance associated with ET2 if the ET2-Test1 ratio is no more than 4 (e.g. 4 or less, 3 or less, 2 or less, 1 or less).

In some embodiments, the first test agent is determined as a potentially effective antimicrobial agent against the drug resistance associated with ET2 if the ET2-Test1 resistance level, which is expressed by minimum inhibitory concentration (MIC), is no more than 10 μg/ml, 5 μg/ml, 2.5 μg/ml, 1 μg/ml or 0.5 μg/ml.

In more particular, the method of the invention further comprises: (16) obtaining a first sum ratio value with respect to the first test agent by adding the ET1-Test1 ratio to the ET2-Test1 ratio.

In more particular, the method of the invention further comprises:

(17) culturing the wild-type susceptible strain in the presence of a second test agent and analyzing the second test agent for its activity against the wild-type susceptible strain to obtain a WT-Test2 resistance level;

(18) culturing the first genetic engineering bacterial strain in the presence of the second test agent and analyzing the second test agent for its activity against the first genetic engineering bacterial strain to obtain a ET1-Test2 resistance level;

(19) obtaining a ET1-Test2 ratio of the ET1-Test2 resistance level to the WT-Test2 resistance level; and

(20) determining if the second test agent is a potentially effective antimicrobial agent against the drug resistance associated with ET1 in the first genetic engineering bacterial strain based on the ET1-Test2 ratio.

In some embodiments, the second test agent is determined as a potentially effective antimicrobial agent against the drug resistance associated with ET1 if the ET1-Test2 ratio is no more than 4 (e.g. 4 or less, 3 or less, 2 or less, 1 or less).

In some embodiments, the second test agent is determined as a potentially effective antimicrobial agent against the drug resistance associated with ET1 if the ET1-Test2 resistance level, which is expressed by minimum inhibitory concentration (MIC), is no more than 10 μg/ml, 5 μg/ml, 2.5 μg/ml, 1 μg/ml or 0.5 μg/ml.

In more particular, the method of the invention further comprises:

(21) culturing the second engineered bacterial strain in the presence of the second test agent and analyzing the second test agent for its activity against the second engineered bacterial strain to obtain a ET2-Test2 resistance level;

(22) obtaining a ET2-Test2 ratio of the ET2-Test2 resistance level to the WT-Test2 resistance level;

(23) determining if the second test agent is a potentially effective antimicrobial agent against the drug resistance associated with ET2 in the second engineered bacterial strain based on the ET2-Test2 ratio.

In some embodiments, the second test agent is determined as a potentially effective antimicrobial agent against the drug resistance associated with ET2 if the ET2-Test2 ratio is no more than 4 (e.g. 4 or less, 3 or less, 2 or less, 1 or less).

In some embodiments, the second test agent is determined as a potentially effective antimicrobial agent against the drug resistance associated with ET2 if the ET2-Test2 resistance level, which is expressed by minimum inhibitory concentration (MIC), is no more than 10 μg/ml, 5 μg/ml, 2.5 μg/ml, 1 μg/ml or 0.5 μg/ml.

In more particular, the method of the invention further comprises: (24) obtaining a second sum ratio value with respect to the second test agent by adding the ET1-Test2 ratio to the ET2-Test2 ratio.

In more particular, the method of the invention further comprises: (25) ranking the first test agent and the second test agent for a preferable antimicrobial activity according to the first sum ratio value and the second sum ratio value, the lower the value, the higher the ranking.

In more particular, the method of the invention further comprises: (26) ranking the first test agent and the second test agent for a preferable antimicrobial activity according to the ET2-Test1 ratio and the ET2-Test2 ratio, the lower the ratio, the higher the ranking.

In some embodiments, the preferable antimicrobial activity includes an antimicrobial activity against multiple drug-resistance mechanisms.

In some embodiments, the first mutation or antibiotic resistance gene, or the second mutation or antibiotic resistance gene is given based on one or more drug-resistant mechanisms selected form the group consisting of (i) decrease antibiotic permeability by loss of outer membrane proteins, (ii) pump out the antibiotics by overexpression of efflux pumps, (iii) eliminates or reduces binding of antibiotic by modification of antibiotic target or by acquirement of antibiotic-resistant target, and (iv) inactivate antibiotic by enzymatic cleavage or modification, and any combinations thereof.

In some embodiments, the first mutation or antibiotic resistance gene, or the second mutation or antibiotic resistance gene is selected from the group consisting of: ΔompK35, ΔompK36, ΔramR, GyrA S83I, GyrA S83L, GyrA S83F, GyrA S83Y, GyrA D87N, ParC S80I , CTX-M-14, CTX-M-15, SHV-12, CMY-2, DHA-1-AmpR, KPC-2, KPC-3, IMP-8, NDM-1, VIM-1, OXA-48, QnrB, QnrS, AAC(6′)-Ib-cr, AAC(6′)-Ib-cr, AAC(3)-IId, AAC(3)-IVa, ANT(2″)-Ia, ANT(3″)-Ia, APH(3′)-Ia, APH(3′)-IIa, StrA-StrB, ArmA, RmtB, Tet(A), Tet(B), Tet(C), Tet(D), Tet(M), Sul1, Sul2, DfrA1, DfrA16, AdeR D20N, AdeR A91V, AdeR P116L, AdeS G30D, AdeS A94V, AdeS R152K, AdeS T153M, ParC G78C, ParC S80L, ParC S80W, ParC S80Y, ParC E84K, VEB-3, ADC-30, IMP-1, OXA-23, OXA-58, OXA-66, OXA-72, AAC(3)-IIa, APH(3′)-VIa, and any combinations thereof.

In some embodiments, the first mutation or antibiotic resistance gene, or the second mutation or antibiotic resistance gene is given by a chromosome-mediated approach or a plasmid-mediated approach.

The foregoing summary, as well as the following detailed description of the invention, will be better understood when read in conjunction with the appended drawings. For the purpose of illustrating the invention, there are shown in the drawings embodiments which are presently preferred. It should be understood, however, that the invention is not limited to the precise arrangements and instrumentalities shown.

In the drawings:

FIGS. 1A and 1B show in vivo efficacy of ceftazidime ( FIG. 1A ) and cefotaxime ( FIG. 1B ) against K. pneumoniae NVT1001 harboring the pACYC177 plasmid with bla OXA-48 in a mouse peritonitis-sepsis model.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by a person skilled in the art to which this invention belongs.

As used herein, the singular forms “a”, “an”, and “the” include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to “a sample” includes a plurality of such samples and equivalents thereof known to those skilled in the art.

As used herein, the term “drug resistance” can refer to a cell's increasing resistance to drug treatment, such as antibiotic treatment.

As used herein, the terms “susceptible” and “antibiotic susceptibility” can indicate that the growth of a microorganism is inhibited by a normal achievable concentration of an antimicrobial agent when the recommended dosage is used.

As used herein, the terms “resistant” and “antibiotic resistance” can indicate that microorganism growth is not inhibited by a normal achievable concentration of the agent and clinical efficacy of the agent against the microorganism has not been shown in treatment studies.

As used herein, a resistance level with reference to a bacterial strain for a particular agent can indicate a minimal amount of the particular agent that will inhibit the visible growth of a microorganism, such as a minimum inhibitory concentration (MIC). A bacterial strain with a lower resistance level to an agent can indicate that the bacterial strain is relatively susceptible to the agent (reflecting a relatively lower MIC), while a bacterial strain with a higher resistance level to an agent can indicate that the bacterial strain is relatively resistant to the agent (reflecting a relatively higher MIC).

According to the present invention, a bacterial strain can be from any bacterial species, particularly a species leading to common infectious diseases in human, including but not limited to, Gram-negative bacteria, such as K. pneumonia, Escherichia coli, Salmonella spp., Acinetobacter baumannii and Pseudomonas aeruginosa , and Gram-positive bacteria, such as Staphylococcus aureus and Enterococcus spp.

As used herein, a wild-type susceptible bacterial strain can indicate a bacterial strain, the genome of which is not modified by a genetic engineering manner and which is susceptible to antibiotic treatment. Preferably, a wild-type susceptible bacterial strain as used herein is susceptible to multiple antibiotics, selected from the group consisting of β-lactams, quinolones, aminoglycosides, tetracyclines, folate pathway inhibitors, polymyxins, phenicols, fosomycins, nitrofurans and any combinations thereof. A particular example of a wild-type susceptible bacterial strain as used herein includes K. pneumoniae strain NVT1001, which can be obtained for example from Division of Infectious Diseases and Tropical Medicine, Department of Internal Medicine, Tri-Service General Hospital, Taipei 114, Taiwan.

As used herein, a genetically engineered bacterial strain can be generated from a wild-type susceptible bacterial strain via a genetic engineering manner, which after the genetic engineering, is given mutation or antibiotic resistance gene to confer drug resistance. The genetic engineering can be conducted by a conventional method as known in the art, such as by a chromosome-mediated approach (e.g. mutating a gene in the chromosome of the strain to confer drug resistance) or a plasmid-mediated approach (e.g. adding a foreign gene to confer drug resistance in the strain). The genetically engineered bacterial strain can include one or more genetic changes based on one or more drug-resistant mechanisms selected form the group consisting of (i) decrease antibiotic permeability by loss of outer membrane proteins, (ii) pump out the antibiotics by overexpression of efflux pumps, (iii) eliminates or reduces binding of antibiotic by modification of antibiotic target or by acquirement of antibiotic-resistant target, and (iv) inactivate antibiotic by enzymatic cleavage or modification, and any combinations thereof. In some embodiments, the mutation or antibiotic resistance gene is selected from the group consisting of: ΔompK35, ΔompK36, ΔramR, GyrA S83I, GyrA S83L, GyrA S83F, GyrA S83Y, GyrA D87N, GyrA S80I, CTX-M-14, CTX-M-15, SHV-12, CMY-2, DHA-1-AmpR, KPC-2, KPC-3, IMP-8, NDM-1, VIM-1, OXA-48, QnrB, QnrS, AAC(6′)-Ib-cr, AAC(6′)-Ib-cr, AAC(3)-IId, AAC(3)-IVa, ANT(2″)-Ia, ANT(3″)-Ia, APH(3′)-Ia, APH(3′)-IIa, StrA-StrB, ArmA, RmtB, Tet(A), Tet(B), Tet(C), Tet(D), Tet(M), Sul1, Sul2, DfrA1, DfrA16, AdeR D20N, AdeR A91V, AdeR P116L, AdeS G30D, AdeS A94V, AdeS R152K, AdeS T153M, ParC G78C, ParC S80L, ParC S80W, ParC S80Y, ParC E84K, VEB-3, ADC-30, IMP-1, OXA-23, OXA-58, OXA-66, OXA-72, AAC(3)-IIa, APH(3′)-VIa, and any combinations thereof. In some embodiments, the mutation or antibiotic resistance gene is preferably include genetic changes as many as possible which can be used to screen for a strong antimicrobial agent against multiple drug-resistant mechanisms.

The present invention provides a method for screening for an antibiotic agent. In particular, the method of the invention creates a genetically engineered bacterial strain from a wild-type susceptible bacterial strain via a genetic engineering manner, which after the genetic engineering, is given mutation or antibiotic resistance gene to confer drug resistance to a reference antibiotic based on one or more drug-resistant mechanisms; a test agent is then added to a culture of the wild-type susceptible bacterial strain and a culture of the genetically engineered bacterial strain to analyze for its activity against the wild-type susceptible bacterial strain and the genetically engineered bacterial strain, respectively, wherein the activity of the test agent against the bacterial strains can be expressed by a minimum inhibitory concentration (MIC) (also referring to a resistance level as described herein; typically, a higher MIC indicates a higher resistance level and a lower antimicrobial activity, and a lower MIC indicates a lower resistance level and a higher antimicrobial activity); and if the activity of the test agent against the genetically engineered bacterial strain is similar to that against the wild-type susceptible bacterial strain, the test agent is selected as a potential antimicrobial agent against the same species of the bacterial strain. In general, the antimicrobial activity of a test agent is expressed by MIC, which is preferably no more than 10 μg/ml, 5 μg/ml, 2.5 μg/ml, 1 μg/ml or 0.5 μg/ml.

Table A shows an embodiment of the method of the present invention.

In certain embodiments, the method of the invention is to identify a potential agent with antimicrobial activity against multiple drug-resistance mechanisms.

In particular, the method of the present invention is advantageous to develop a platform technology to screen for a large number of agents for their antimicrobial activity. In addition, once a test agent is determined as a potentially antimicrobial agent in a first run of the process, it can be subjected to a subsequent run of the process by creating another genetically engineered bacterial strain with a new mutation or antibiotic resistance gene and repeating the steps of culturing, obtaining a resistance level and analyzing for its activity against the wild-type susceptible bacterial strain and the genetically engineered bacterial strain. The more times the process is conducted, the more accurate the results, and an agent thus selected by more runs of the process is more effective against various types of drug-resistant mechanisms.

The present invention will now be described more specifically with reference to the following examples, which are provided for the purpose of demonstration rather than limitation.

The system developed in the present study provides a platform technology for testing, screening, selecting and evaluating antibiotics using genetically engineered bacterial strains with individual or combined resistance mechanisms that were constructed from the fully susceptible clinical K. pneumoniae NVT1001 isolate. The chromosome-mediated resistance mechanisms that were constructed in this study are decreased antibiotic permeability, generated by deletion of the genes encoding the outer-membrane proteins OmpK35 and/or OmpK36; 7 active antibiotic export, generated by deletion of the ramR regulatory gene, which leads to overexpression of the AcrAB-TolC efflux pump (Tables 1); 8-11 and modification of quinolone target sites via gyrA and/or parC mutations. 12-15 Most plasmid-mediated mechanisms are specific to one kind of antibiotic group, 16 with this study focusing on β-lactams, quinolones, aminoglycosides, tetracyclines and folate pathway inhibitors. Because many kinds of resistance are conferred by combined resistance mechanisms in the clinical setting, mechanisms related to the same antibiotic group were combined. Testing antibiotics against these combined mechanisms can be used to further evaluate their antimicrobial activities.

In addition, Enterobacteriaceae, Acinetobacter baumannii and Pseudomonas aeruginosa have been identified by the WHO as three critical priority pathogens for the research and development of new antibiotics in 2017. 3 Given that the genetic background of A. baumannii is quite different from that of K. pneumoniae , specific resistance genes can be found in this pathogen. Therefore, the strategic system has been constructed by using A. baumannii.

This study includes generation of genetically engineered strains from a fully susceptible clinical isolate of A. baumannii . The chromosome-mediated resistance mechanisms that were constructed in this study are active antibiotic export, generated by mutation of the adeR or adeS regulatory gene, which leads to overexpression of the AdeABC efflux pump;41-46 and modification of quinolone target sites via gyrA and/or parC mutations. 47-54 Most plasmid-mediated mechanisms are specific to one kind of antibiotic group, 55 with this study focusing on β-lactams, aminoglycosides and tetracyclines. All of these genetically engineered A. baumannii strains are with clear and simple resistance mechanisms and their usability was evaluated via antimicrobial susceptibility testing in this study.

Prior to the present invention, no complete system is available for testing antibiotics against specific resistance mechanisms. This study generated genetically engineered strains with clear and simple resistance mechanisms and evaluated their usability via both in vitro and in vivo assays.

1. Materials and Methods

1.1 Concept and Approach to Engineer the Resistant Strains for Antibiotic Screening System.

To create the bacterial strains for antibiotic screening system, four approaches with different target related antibiotic resistance were used.

For construction of chromosomal mediated resistance, (i) decreasing penetration of antibiotic (knockout outer membrane protein genes ompK35 and ompK36 in K. pneumoniae ), 7 (ii) increasing the expression of AcrAB-TolC (knockout regulatory gene ramR 8-11 or overexpression of the AdeABC efflux pump (mutation of the adeR or adeS regulatory gen) 41-46 and (iii) antibiotic target gene mutation were performed. 12-15,47-54 For plasmid mediated resistance, different class of plasmid mediated resistant genes including (i) β-lactam resistant genes, (ii) quinolones resistant genes, (iii) aminoglycosides resistant genes, (iv) tetracyclines resistant genes and (v) folate pathway inhibitor resistant genes 16 were individually transferred into susceptible strain. See, Tables 2 and 4.

1.2 Bacterial Strains and Plasmids

1.2.1 K. pneumoniae NVT1001

K. pneumoniae NVT1001, capsular serotype 1, is an isolate from a liver-abscess patient in Taiwan 17 that was fully susceptible to all antibiotics tested, except ampicillin. The NVT1001 strain was used to construct genetically engineered strains with the antibiotic resistance mechanisms that are shown in Table 2, and the revertants of the NVT1001 mutants constructed in this study are shown in Table 3.

1.2.2 A. baumannii KW1

A. baumannii KW1 is a clinical isolate that was fully susceptible to antibiotics tested. The KW1 strain was used to construct genetically engineered strains with the antibiotic resistance mechanisms that are shown in Table 4. The plasmids pYMAb5 and pYMab5Tc are shuttle vectors able to replicate in A. baumannii and Escherichia coli , which carry a resistant determinant of kanamycin or tetracycline respectively. The two shuttle vectors were used to clone the resistance genes of the plasmid-mediated resistance mechanisms in this study (Table 4).

1.3 In-Frame Deletion Mutagenesis

The plasmid pUT-kmy, which consists of an R6K origin of replication, an mobRP4 origin of transfer, and a kanamycin resistance cassette, 18 was ligated with the sacB gene to generate the plasmid pUT-KB, which was then used to construct the mutants. 19 The plasmid pUT-KB is a suicide vector containing the counter-selection marker sacB, which originates from Bacillus subtilis. 20 When this gene is expressed via the integrated pUT-KB, it confers a sucrose-susceptibility phenotype, which enables positive selection with sucrose to detect loss of the vector.

Gene deletions in K. pneumoniae strain NVT1001 were constructed via in-frame deletion mutagenesis. 7 Briefly, two DNA fragments (approximately 1 kb in size) that flanked the regions to be deleted were amplified by PCR using specific primer pairs. The two gel-purified PCR products, containing complementary ends, were then mixed and amplified via overlap PCR. 21,22 The resulting PCR fragment (approximately 2 kb in size) was digested with restriction enzymes and then cloned into pUT-KB that had been similarly digested. For homologous recombination, each of the gene-deletion constructs in pUT-KB was transformed into E. coli S17-1 λpir 23 via electroporation and then mobilized into K. pneumoniae strain NVT1001 via conjugation. Single-crossover strains were selected from BIND (Brilliant green containing Inositol-Nitrate-Deoxycholate) plates supplemented with kanamycin (50 mg/L), as the growth of non- K. pneumoniae strains is effectively suppressed on BIND plates. 24 After the kanamycin-resistant transconjugants were selected, the insertions of the plasmids were verified via PCR using primer pairs that flanked the target genes. After incubation in 20 mL brain-heart infusion (BHI) broth for 6 hours at 37° C. in the absence of kanamycin, the fully grown cultures were spread onto LB agar plates supplemented with 10% sucrose. After the double crossover occurred, the sucrose-resistant and kanamycin-susceptible colonies were selected, and the gene deletions were confirmed via PCR.

1.4 Site-Directed Mutagenesis

DNA fragments of the entire gyrA and parC sequences along with their flanking regions were amplified from K. pneumoniae strain NVT1001 using PCR with specific primer pairs and then cloned into the plasmid pUT-KB. The QuikChange site-directed mutagenesis kit (Stratagene) was used to generate mutations in the gyrA and parC genes in the plasmid using the methods described by the manufacturer. For homologous recombination, plasmids containing mutations in the gyrA or parC gene were transformed into E. coli S17-1 λpir 23 via electroporation and mobilized into K. pneumoniae strain NVT1001 via conjugation. Single-crossover strains were selected from BIND plates supplemented with kanamycin (50 mg/L), as the growth of non- K. pneumoniae contaminants is suppressed on BIND plates. 24 After the kanamycin-resistant transconjugants were selected, the insertion of plasmids was verified via PCR. After incubation in 20 mL BHI broth for 6 hours at 37° C. in the absence of kanamycin, the fully grown cultures were spread onto LB agar plates supplemented with 10% sucrose. After the double crossover occurred, the sucrose-resistant and kanamycin-susceptible colonies were selected, and the gene mutations were confirmed via PCR.

Plasmid pUT-kmy, which consists of an R6K origin of replication, a mobRP4 origin of transfer, and a kanamycin resistant determinant, 56 was ligated with a sacB gene to generate plasmid pUT-KB for constructing mutants. 57 Plasmid pUT-KB is a suicide vector containing a counter-selection marker, sacB, which originates from Bacillus subtilis. 58 When this gene is expressed on the integrated pUT-KB, it confers a sucrose-sensitivity phenotype, which enables positive selection with sucrose to detect the loss of the vector.

DNA fragments of the entire adeR, adeS, gyrA and parC along with their flanking regions were amplified from A. baumannii KW1 using PCR with specific primer pairs and then cloned into plasmid pUT-KB. The QuikChange site-directed mutagenesis kit (Stratagene) was used to generate mutations in the adeR, adeS, gyrA and parC genes in the plasmids using the methods described by the manufacturer. For homologous recombination, plasmids containing mutations in the adeR, adeS, gyrA or parC gene were transformed into E. coli S17-1 λpir 23 via electroporation and mobilized into A. baumannii KW1 via conjugation. Single-crossover strains were selected from Luria-Bertani (LB) plates supplemented with cefotaxime (1 μg/ml) and kanamycin (50 m/ml), as the growth of non- A. baumannii contaminants is suppressed by cefotaxime (1 μg/ml). The kanamycin-resistant transconjugants were selected and the insertion of plasmids was verified via PCR. After incubation in 20 ml brain heart infusion (BHI) broth for 6 hours in the absence of kanamycin at 37° C., the fully grown cultures were spread onto LB plates supplemented with 10% sucrose. After the double crossover occurred, the sucrose-resistant and kanamycin-susceptible colonies were selected, and the gene mutations were confirmed via PCR.

1.5 Construction of Revertants

The allelic-exchange method was used to restore the wild-type gene in the K. pneumoniae NVT1001 mutants with constructed chromosome-mediated resistance mechanisms, as described by Tsai et al. 19 Briefly, a DNA fragment of the entire wild-type gene sequence along with flanking regions was amplified from K. pneumoniae NVT1001 using PCR with specific primers. The resulting PCR fragment was digested and then cloned into pUT-KB. For homologous recombination, this plasmid was transformed into E. coli S17-1 λpir 23 via electroporation and mobilized into the K. pneumoniae NVT1001 mutant via conjugation. Single-crossover strains were selected from BIND plates supplemented with kanamycin (50 mg/L), as the growth of non- K. pneumoniae strains is effectively suppressed on the BIND plates. 24 After the kanamycin-resistant transconjugants were selected, the insertion of the plasmid was verified via PCR. After incubation in 20 mL BHI broth for 6 hours at 37° C. in the absence of kanamycin, the fully grown cultures were spread onto LB agar plates supplemented with 10% sucrose. After the double crossover occurred, the sucrose-resistant and kanamycin-susceptible colonies were selected, and the restoration of the wild-type gene was confirmed via DNA sequencing.

1.6 Plasmid Construction and Transformation

DNA fragments of the resistance genes along with their flanking regions were amplified from clinical plasmids via PCR with specific primer pairs. The resulting PCR fragments were digested and then cloned into the plasmid pACYC177 or pACYC184. The resulting plasmids were then transformed into K. pneumoniae strain NVT1001 via electroporation. The recombinant bacteria were plated onto LB agar plates containing kanamycin (50 mg/L) or chloramphenicol (50 mg/L), and the presence of the cloned gene was confirmed via PCR and DNA sequencing.

DNA fragments of the resistance genes along with their flanking regions were amplified from clinical plasmids via PCR with specific primer pairs. The resulting PCR fragments were digested and then cloned into the plasmid pYMAb5 or pYMAb5Tc. The resulting plasmids were then transformed into A. baumannii KW1 via electroporation. The recombinant bacteria were plated onto LB agar plates containing kanamycin (50 μg/ml) or tetracycline (50 μg/ml), and the presence of the cloned gene was confirmed via PCR and DNA sequencing.

1.7 Antimicrobial Susceptibility Testing

The MICs of antibiotics were determined using the Etest (Biodisk AB, Sweden) according to the manufacturer's instructions. The MICs from the Etest, which fall between standard two-fold dilutions, were rounded up to the next highest two-fold value. The MICs of ceftazidime and cefotaxime against K. pneumoniae NVT1001 harbouring the pACYC177 plasmid with bla OXA-48 were further determined using a broth microdilution test according to the recommendations of the CLSI, 25 and the results were interpreted according to the breakpoints established by the CLSI in 2017. 25 Meanwhile, due to the lack of CLSI breakpoints in 2016 and 2017, the results for cephalothin were interpreted according to the breakpoints established by the CLSI in 2015, 26 and the results for moxifloxacin were interpreted according to the breakpoints established by the EUCAST in 2017. 27

1.8 Mouse Infection Model

Pathogen-free 6- to 8-week-old male BALB/c mice were obtained from the National Laboratory Animal Center (Taiwan) and maintained in the pathogen-free vivarium of the Laboratory Animal Center of National Defense Medical Center (Taiwan). K. pneumoniae NVT1001 harbouring the pACYC177 plasmid with bla OXA-48 was cultured overnight at 37° C. in BHI broth and then diluted (1:100) in fresh BHI broth. The culture was incubated until the mid-exponential growth phase, and the cells were then washed once, resuspended in 0.85% sterile saline, and adjusted to the desired concentrations according to the OD 600 value. The concentrations were verified by plating the cells to determine viable counts. Six mice from each group were then injected intraperitoneally with 0.1 mL of the cell suspension containing twice the 90-100% lethal dose (LD90-100); this inoculum of 4×10 3 cfu/mouse was known to be 100% lethal within 7 days (data not shown). Meanwhile, antibiotic doses were prepared in 0.85% sterile saline. At 1 hour post-inoculation, antibiotic or 0.85% sterile saline alone was administered as a single subcutaneous injection with a volume of 0.3 mL per dose. The mice were then monitored daily for 7 days to measure survival. The ED50 was defined as the single dose giving protection to 50% of the test mice.

1.9 Quantitative real-time PCR (qRT-PCR)

K. pneumoniae strains were cultured in Mueller-Hinton broth (MHB) at 37° C. overnight, diluted (1:100) in fresh MHB, and then incubated at 37° C. until an OD600 of 0.8 (the mid-exponential growth phase) is reached. RNA was extracted using the RNeasy kit and treated with RNase-free DNase I according to the manufacturer's instructions (Qiagen). RNA yield and quality was measured using the NanoVue spectrophotometer (GE Healthcare Life Sciences). cDNAs were synthesized from 1 μg of RNA template using the Omniscript Reverse Transcriptase according to the manufacturer's instructions (Qiagen). Relative quantification of gene expression was performed using ABI PRISM 7900HT real-time PCR System (Applied Biosystems) with the Maxima SYBR Green qPCR Master Mix (Thermo Scientific) and primers listed in Table 5. Fold-change values were calculated after normalization of each gene to the 23S rRNA internal control, 3 using the comparative threshold method with the NVT1001 strain as the reference strain.

2. Results

2.1 K. pneumoniae NVT1001

2.1.1 Porin Loss and Efflux-Pump Overexpression in Antibiotic Resistance

The MICs of antibiotics against chromosome-mediated resistance mechanisms, namely, OmpK35/36 loss and AcrAB-TolC overexpression (ΔramR), are shown in Table 6. These results demonstrate that OmpK35/36 loss is related to increased resistance to β-lactams, folate pathway inhibitors, fosfomycins and nitrofurans, whereas AcrAB-To1C overexpression (ΔramR) is associated with increased resistance to β-lactams, quinolones, tetracyclines, folate pathway inhibitors, phenicols and nitrofurans. The two resistance mechanisms alone or in combination had no significant (≥4-fold) effects on the MIC(s) of imipenem, aminoglycosides or polymyxins. The results shown above were further validated by testing the antimicrobial susceptibility of the revertant strains, and similar results were found when comparing the MIC of the wild-type strain NVT1001 or its mutant to the MIC of the revertant with the same genotype (Table 7).

2.1.2 β-Lactams

The MICs of β-lactams against associated resistance mechanisms are shown in Tables 8 and 9. In particular, the MICs for extended-spectrum β-lactamases (ESBLs) and AmpC β-lactamases are shown in Table 8, while those for carbapenemases are shown in Table 9. The production of ESBLs when tested alone showed no significant (≥4-fold) effects on the MIC of piperacillin/tazobactam, cefoxitin, imipenem or doripenem, and the production of AmpC β-lactamases when tested alone showed no significant (≥4-fold) effects on the MIC of imipenem (Table 8). With or without ESBLs or AmpC β-lactamases, the strains with AcrAB-TolC overexpression (ΔramR) all showed no significant (≥4-fold) effects on the MIC of imipenem or doripenem (Table 8). The production of KPC-2 or KPC-3 alone could confer intermediate resistance to cefoxitin as well as resistance to the other tested β-lactams (Table 9). In addition, with or without OmpK35/36 loss and/or AcrAB-TolC overexpression (ΔramR), the IMP-8, NDM-1, VIM-1 and OXA-48 strains were susceptible to aztreonam (Table 9).

2.1.3 Quinolones

The MICs of quinolones against chromosome-mediated resistance mechanisms are shown in Table 10. Strains with a single mutation at GyrA (S83I, S83L, S83F, S83Y or D87N) or with AcrAB-TolC overexpression (ΔramR) showed significant (≥4-fold) increases in the MICs of all quinolones tested (Table 5). In contrast, the single mutation at ParC (S80I) showed no significant (≥4-fold) effects on the MIC of any of the quinolones tested, whereas significant (≥4-fold) increases in the MICs of quinolones were observed for the S83I, S83L, S83F, S83Y, S83I/D87N, S83L/D87N, S83F/D87N, S83Y/D87N and ΔramR/S83I mutants (Table 10). The results shown above were further validated by testing the antimicrobial susceptibility of the revertant strains, and similar results were found when comparing the MIC of the wild-type strain NVT1001 or its mutant to the MIC of the revertant with the same genotype (Table 11).

The MICs of quinolones against plasmid-mediated resistance mechanisms with or without combination with chromosome-mediated resistance mechanisms are shown in Table 6. In contrast to QnrB or QnrS, strains with AAC(6′)-Ib-cr exhibited a significant (≥4-fold) increase in the MIC of only ciprofloxacin or norfloxacin (Table 12). Even without plasmid-mediated resistance mechanisms, strains with the GyrA (S83I) mutation and AcrAB-TolC overexpression (ΔramR) could resist all tested quinolones (Table 12).

2.1.4 Aminoglycosides

The MICs of aminoglycosides for each resistance mechanism are shown in Table 13. Several different resistance profiles were found among the strains with different aminoglycoside-modifying enzymes, though all were susceptible to amikacin (Table 13). The production of 16S rRNA methylase, ArmA and RmtB was not associated with observable effects on the MIC of spectinomycin or streptomycin, whereas both conferred strong resistance to all other aminoglycosides tested (Table 13).

2.1.5 Tetracyclines

The MICs of tetracyclines for various resistance mechanisms are shown in Table 14. With or without AcrAB-To1C overexpression (ΔramR), the production of Tet(B), Tet(C), Tet(D), or Tet(M) showed no observable effects on the MIC of tigecycline (Table 8). AcrAB-TolC overexpression (ΔramR) and Tet(A) production both conferred an 8-fold increase in the MIC of tigecycline, and a 32-fold increase was found when the two mechanisms were combined (Table 14).

2.1.6 Folate Pathway Inhibitors

The MICs of folate pathway inhibitors against associated resistance mechanisms are shown in Table15. Strains with single resistance mechanisms were all susceptible (≤2/38 mg/L) to trimethoprim/sulfamethoxazole, and strains with certain combinations of mechanisms became resistant to trimethoprim/sulfamethoxazole (Table 15). Strains with porin loss and/or Sul production all showed no significant (≥4-fold) effects on the MIC of trimethoprim, whereas most of the other strains with combined mechanisms were associated with significant (≥4-fold) increases in the MICs of all folate pathway inhibitors tested (Table 15).

2.1.7 In Vivo Investigation

To further demonstrate that this system can also be used to test the efficacy of antibiotics in vivo, a mouse infection model was used. The 50% effective doses (ED50 values) of ceftazidime and cefotaxime were specifically determined using K. pneumoniae NVT1001 harbouring the pACYC177 plasmid with bla OXA-48 ( FIG. 1 ). The in vivo efficacy of ceftazidime was higher than that of cefotaxime, which had an estimated ED50 of 30 mg/kg, whereas no mice survived treatment with 30 mg/kg cefotaxime ( FIG. 1 ). The higher efficacy of ceftazidime was also observed in the in vitro MIC assays, in which the MICs of ceftazidime and cefotaxime were 0.5 mg/L and 2 mg/L, respectively, using the broth microdilution test.

2.2 A. baumannii KW1

2.2.1 Efflux Pump Overexpression

The single mutation in the regulator genes of AdeABC efflux pump could significantly (≥4-fold) increase the MICs of ceftazidime, gentamicin, tetracycline and/or tigecycline, including D20N, A91V or P116L mutation in AdeR and G30D, A94V, R152K or T153M mutation in AdeS (Table 16). These mutations could also confer a 2 or 3-fold increase in the MIC of ciprofloxacin (Table 16).

2.2.2 Quinolones Target Site Mutation

The single mutation S83L in GyrA could significantly (4-fold) increase the MICs of all quinolones tested (Table 17). With the further mutation in ParC, including G78C, S80L,

S80W, S80Y or E84K mutation, could increase these MICs to high-level resistance (Table 17), whereas the single mutation of all these ParC mutations showed no significant (≥4-fold) effects on the MICs of any of the quinolones tested (Table 17).

2.2.3 Plasmid-Mediated Resistance Mechanism of β-Lactams

The MICs of β-lactams against associated resistance mechanisms are shown in Table 18. The production of CTX-M-15 or ADC-30 could significantly (≥4-fold) increase the MICs of all β-lactams tested. The production of each carbapenemase tested all showed no increase effects on the MIC of aztreonam, and no increase effects on the MIC of ceftazidime could also be found on the production of OXA-type carbapenemases tested in this study.

2.2.4 Plasmid-Mediated Resistance Mechanism of Aminoglycosides

The MICs of aminoglycosides for each resistance mechanism are shown in Table 19. Different resistance effects could be found on the production of these aminoglycoside-modifying enzymes, whereas all showed no increase effects on the MICs of amikacin, netilmicin and streptomycin. The production of 16S rRNA methylase, ArmA, was not associated with increase effects on the MICs of spectinomycin and streptomycin, whereas conferred strong resistance to all other aminoglycosides tested.

2.2.5 Plasmid-Mediated Resistance Mechanism of Tetracyclines

The MICs of tetracyclines for its resistance mechanisms are shown in Table 20. The production of Tet(A), Tet(B) or Tet(C) could significantly (4-fold) increase the MICs of tetracycline, doxycycline and minocycline. In the MICs of tigecycline, the production of Tet(B) or Tet(M) showed no observable effects, whereas the production of Tet(A) could confer a 2 to 3-fold increase.

3. Discussion

In this study, 193 genetically engineered strains with different resistance mechanisms were constructed from K. pneumoniae NVT1001, a fully susceptible clinical isolate, including 29 strains with chromosome-mediated resistance, 33 strains with plasmid-mediated resistance and 131 strains with a combination of the two resistance mechanisms. In addition, 37 genetically engineered strains with different resistance mechanisms were constructed from A. baumannii KW1, a fully susceptible clinical isolate, including 18 strains with chromosome-mediated resistance and 19 strains with plasmid-mediated resistance. Because the resistance mechanisms of these strains were constructed, all are known to be resistant to specific antibiotics. In addition, the plasmid-mediated resistance mechanisms were constructed using non-conjugative plasmids without transposons, so these mechanisms are not easy to spread. These features suggest that using these strains is safer than using drug-resistant clinical isolates.

Also in contrast to clinical drug-resistant isolates, these genetically engineered strains have clear and simple antibiotic resistance mechanisms. In in vitro MIC assays, the resistance profiles of these strains were confirmed by testing the MICs against several well-known antibiotics. Investigators can estimate the effectiveness of their antibiotics by comparing their MIC results to those of well-known antibiotics. The MIC assay is usually the starting point for assessing antibiotics, and our results suggest that these strains were ready for testing antibiotic activities in vitro.

In vivo mouse infection models are commonly used to demonstrate the efficacy of antibiotics in protecting against lethal infection. 28-30 Using a clinical K. pneumoniae isolate producing the carbapenemase OXA-48, a previous study found that the MIC of ceftazidime was 4-fold lower than that of cefotaxime and that the ED50 of ceftazidime was lower than that of cefotaxime. 28 A similar result was obtained in the present study using K. pneumoniae NVT1001 harbouring the pACYC177 plasmid with bla OXA-48 . In contrast to many other clinical isolates, this genetically engineered strain can effectively infect BALB/c mice without weakening their immune system, and its 100% lethal dose (LD100) is 4×10 3 cfu in the mouse infection model, similar to its parental strain NVT1001 (data not shown). Obtained without changing the immune system of mice, the in vivo results should be closer to the true antibiotic efficacy during clinical use. The results thus suggest that these genetically engineered strains are adequate for testing antibiotic activities in vivo.

For the development of antibiotics in the face of multidrug resistance, in vitro selection and earlier evaluation are important steps before in vivo studies and further clinical evaluation. In contrast to the use of clinical isolates, the use of the platform technology described here to evaluate antibiotics may elucidate the exact reason for and the level of resistance both in vitro and in vivo. This information may in turn help in estimation of the efficacy of and potential resistance to antibiotics. The modification of antibiotics' chemical structure and discontinuities in development can lower the costs associated with the development process. Furthermore, the global prevalence and distribution of antibiotic resistance genes are already available from various studies and monitoring groups. 31-36 In combination with these data, the data obtained from this platform technology may also help in estimation of the rates of resistance to newly developed drugs across regions. Additionally, antibiotics should be used more carefully in regions where specific resistance mechanisms are epidemic.

The set of resistant strains included in this platform technology can be expanded. Specifically, through genetic construction, other resistance mechanisms or different combinations of mechanisms can also be constructed in the same parental strain, including newly identified mechanisms. Given that the genetic background of several important bacterial pathogens with high resistance rates is quite different from that of K. pneumoniae , specific resistance genes can be found in these pathogens. 37,38 For example, the overexpression of the AdeABC efflux pump can confer antibiotic resistance in A. baumannii , whereas this efflux pump does not exist in K. pneumoniae. 37 To completely evaluate the activities of antibiotics, these specific resistance mechanisms should also be constructed in their original pathogens; in fact, several of them have been successfully constructed by our group (data not shown).

In summary, various chromosomal and plasmid resistance mechanisms were constructed in this study, and especially mechanisms conferring resistance to β-lactams, quinolones, aminoglycosides, tetracyclines or folate pathway inhibitors, in fully susceptible bacterial strains. Our results demonstrate that this platform technology can be used to efficiently and effectively evaluate antibiotics targeting specific resistance mechanisms both in vitro and in vivo. The system can also be used to screen new antibiotics against multidrug-resistant bacteria.

1 O'Neill J, chair. Tackling drug-resistant infections globally: final report and recommendations. Review on Antimicrobial Resistance , London, United Kingdom 2016.

2 Boucher H W, Talbot G H, Benjamin D K et al. 10×′20 Progress—development of new drugs active against gram-negative bacilli: an update from the Infectious Diseases Society of America. Clin Infect Dis 2013; 56: 1685-94.

3 WHO. Global priority list of antibiotic-resistant bacteria to guide research, discovery, and development of new antibiotics. World Health Organization 2017. http://www.who.int/medicines/publications/global-priority-list-antibiotic-resistant-bacteriden/.

4 WHO. Antimicrobial resistance: global report on surveillance. World Health Organization 2014. http://www.who.int/drugresistance/documents/surveillancereport/en/.

5 Iredell J, Brown J, Tagg K. Antibiotic resistance in Enterobacteriaceae: mechanisms and clinical implications. BMJ 2016; 352.

6 Mendes R E, Castanheira M, Gasink L et al. β-lactamase characterization of gram-negative pathogens recovered from patients enrolled in the phase 2 trials for ceftazidime-avibactam: clinical efficacies analyzed against subsets of molecularly characterized isolates. Antimicrob Agents Chemother 2015; 60: 1328-35.

7 Tsai Y K, Fung C P, Lin J C et al. Klebsiella pneumoniae outer membrane porins OmpK35 and OmpK36 play roles in both antimicrobial resistance and virulence. Antimicrob Agents Chemother 2011; 55: 1485-93.

8 Wang X, Chen H, Zhang Y et al. Genetic characterisation of clinical Klebsiella pneumoniae isolates with reduced susceptibility to tigecycline: Role of the global regulator RamA and its local repressor RamR. Int J Antimicrob Agents 2015; 45: 635-40.

9 Bailey A M, Ivens A, Kingsley R et al. RamA, a member of the AraC/XylS family, influences both virulence and efflux in Salmonella enterica serovar Typhimurium. J Bacteriol 2010; 192: 1607-16.

10 Bialek-Davenet S, Leflon-Guibout V, Tran Minh O et al. Complete deletion of the ramR gene in an in vitro-selected mutant of Klebsiella pneumoniae overexpressing the AcrAB efflux pump. Antimicrob Agents Chemother 2013; 57: 672-3.

11 Blanco P, Hernando-Amado S, Reales-Calderon J A et al. Bacterial multidrug efflux pumps: much more than antibiotic resistance determinants. Microorganisms 2016; 4.

12 Stoesser N, Batty E M, Eyre D W et al. Predicting antimicrobial susceptibilities for Escherichia coli and Klebsiella pneumoniae isolates using whole genomic sequence data. J Antimicrob Chemother 2013; 68: 2234-44.

13 Ruiz E, Saenz Y, Zarazaga M et al. qnr, aac(6′)-Ib-cr and qepA genes in Escherichia coli and Klebsiella spp.: genetic environments and plasmid and chromosomal location. J Antimicrob Chemother 2012; 67: 886-97.

14 Rodriguez-Martinez J M, Diaz de Alba P, Briales A et al. Contribution of OqxAB efflux pumps to quinolone resistance in extended-spectrum-β-lactamase-producing Klebsiella pneumoniae. J Antimicrob Chemother 2013; 68: 68-73.

15 Fu Y, Guo L, Xu Y et al. Alteration of GyrA amino acid required for ciprofloxacin resistance in Klebsiella pneumoniae isolates in China. Antimicrob Agents Chemother 2008; 52: 2980-3.

16 Roberts M C, Schwarz S, Aarts H J. Commentary on “Acquired antibiotic resistance genes: an overview”. Front Microbiol 2012; 3.

17 Yeh K M, Kurup A, Siu L K et al. Capsular serotype K1 or K2, rather than magA and rmpA, is a major virulence determinant for Klebsiella pneumoniae liver abscess in Singapore and Taiwan. J Clin Microbiol 2007; 45: 466-71.

18 Yeh K M, Lin J C, Yin F Y et al. Revisiting the importance of virulence determinant magA and its surrounding genes in Klebsiella pneumoniae causing pyogenic liver abscesses: exact role in serotype K1 capsule formation. J Infect Dis 2010; 201: 1259-67.

19 Tsai Y K, Liou C H, Lin J C et al. A suitable streptomycin-resistant mutant for constructing unmarked in-frame gene deletions using rpsL as a counter-selection marker. PloS one 2014; 9: e109258.

20 Reyrat J-M, Pelicic V, Gicquel B et al. Counterselectable markers: untapped tools for bacterial genetics and pathogenesis. Infect Immun 1998; 66: 4011-7.

21 Ho S N, Hunt H D, Horton R M et al. Site-directed mutagenesis by overlap extension using the polymerase chain reaction. Gene 1989; 77: 51-9.

22 Horton R M, Hunt H D, Ho S N et al. Engineering hybrid genes without the use of restriction enzymes: gene splicing by overlap extension. Gene 1989; 77: 61-8.

23 Skorupski K, Taylor R K. Positive selection vectors for allelic exchange. Gene 1996; 169: 47-52.

24 Ohtomo R, Saito M. A new selective medium for detection of Klebsiella from dairy environments. Microbes Environ 2003; 18: 138-44.

25 Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing: Twenty - seventh Informational Supplement M 100-S27. CLSI, Wayne, Pa., USA, 2017.

26 Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing: Twenty - fifth Informational Supplement M 100-S25. CLSI, Wayne, Pa., USA, 2015.

27 EUCAST. The European Committee on Antimicrobial Susceptibility Testing. Breakpoint tables for interpretation of MICs and zone diameters. Version 7.1, 2017.

28 Mimoz O, Grégoire N, Poirel L et al. Broad-spectrum β-lactam antibiotics for treating experimental peritonitis in mice due to Klebsiella pneumoniae producing the carbapenemase OXA-48. Antimicrob Agents Chemother 2012; 56: 2759-60.

29 Reyes N, Aggen J B, Kostrub C F. In vivo efficacy of the novel aminoglycoside ACHN-490 in murine infection models. Antimicrob Agents Chemother 2011; 55: 1728-33.

30 Knudsen J D, Fuursted K, Frimodt-Møller N et al. Comparison of the effect of cefepime with four cephalosporins against pneumococci with various susceptibilities to penicillin, in vitro and in the mouse peritonitis model. J Antimicrob Chemother 1997; 40: 679-86.

31 van Duin D, Doi Y. The global epidemiology of carbapenemase-producing Enterobacteriaceae. Virulence 2016: 1-10.

32 Tangden T, Giske C G. Global dissemination of extensively drug-resistant carbapenemase-producing Enterobacteriaceae: clinical perspectives on detection, treatment and infection control. J Intern Med 2015; 277: 501-12.

33 Wachino J, Arakawa Y. Exogenously acquired 16S rRNA methyltransferases found in aminoglycoside-resistant pathogenic Gram-negative bacteria: an update. Drug Resist Updat 2012; 15: 133-48.

34 Nordmann P, Naas T, Poirel L. Global spread of carbapenemase-producing Enterobacteriaceae. Emerg Infect Dis 2011; 17: 1791-8.

35 Gupta N, Limbago B M, Patel J B et al. Carbapenem-resistant Enterobacteriaceae: epidemiology and prevention. Clin Infect Dis 2011; 53: 60-7.

36 Hawkey P M, Jones A M. The changing epidemiology of resistance. J Antimicrob Chemother 2009; 64: i3-i10.

37 Santajit S, Indrawattana N. Mechanisms of antimicrobial resistance in ESKAPE pathogens. Biomed Res Int 2016; 2016: 2475067.

38 Rice L B. Federal funding for the study of antimicrobial resistance in nosocomial pathogens: no ESKAPE. J Infect Dis 2008; 197: 1079-81.

39. WHO. 2017. Global priority list of antibiotic-resistant bacteria to guide research, discovery, and development of new antibiotics. World Health Organization.

40. Lin C T, Wu C C, Chen Y S, Lai Y C, Chi C, Lin J C, Chen Y, Peng H L. 2011. Fur regulation of the capsular polysaccharide biosynthesis and iron-acquisition systems in Klebsiella pneumoniae CG43. Microbiology 157:419-429.

41. Yoon E J, Courvalin P, Grillot-Courvalin C. 2013. RND-type efflux pumps in multidrug-resistant clinical isolates of Acinetobacter baumannii : major role for AdeABC overexpression and AdeRS mutations. Antimicrob Agents Chemother 57:2989-95.

42. Hornsey M, Loman N, Wareham D W, Ellington M J, Pallen M J, Turton J F, Underwood A, Gaulton T, Thomas C P, Doumith M, Livermore D M, Woodford N. 2011. Whole-genome comparison of two Acinetobacter baumannii isolates from a single patient, where resistance developed during tigecycline therapy. J Antimicrob Chemother 66:1499-503.

43. Hornsey M, Ellington M J, Doumith M, Thomas C P, Gordon N C, Wareham D W, Quinn J, Lolans K, Livermore D M, Woodford N. 2010. AdeABC-mediated efflux and tigecycline MICs for epidemic clones of Acinetobacter baumannii . Journal of Antimicrobial Chemotherapy 65:1589-1593.

44. Coyne S, Guigon G, Courvalin P, Perichon B. 2010. Screening and quantification of the expression of antibiotic resistance genes in Acinetobacter baumannii with a microarray. Antimicrob Agents Chemother 54:333-40.

45. Higgins P G, Schneiders T, Hamprecht A, Seifert H. 2010. In vivo selection of a missense mutation in adeR and conversion of the novel bla OXA-164 gene into bla OXA-58 in carbapenem-resistant Acinetobacter baumannii isolates from a hospitalized patient. Antimicrobial Agents and Chemotherapy 54:5021-5027.

46. Marchand I, Damier-Piolle L, Courvalin P, Lambert T. 2004. Expression of the RND-type efflux pump AdeABC in Acinetobacter baumannii is regulated by the AdeRS two-component system. Antimicrobial Agents and Chemotherapy 48:3298-3304.

47. Ardebili A, Lari A R, Beheshti M, Lari E R. 2015. Association between mutations in gyrA and parC genes of Acinetobacter baumannii clinical isolates and ciprofloxacin resistance. Iran J Basic Med Sci 18:623-6.

48. Lopes B S, Amyes S G. 2013. Insertion sequence disruption of adeR and ciprofloxacin resistance caused by efflux pumps and gyrA and parC mutations in Acinetobacter baumannii . Int J Antimicrob Agents 41:117-21.

49. Liu Y H, Kuo S C, Lee Y T, Chang I C, Yang S P, Chen T L, Fung C P. 2012. Amino acid substitutions of quinolone resistance determining regions in GyrA and ParC associated with quinolone resistance in Acinetobacter baumannii and Acinetobacter genomic species 13TU. J Microbiol Immunol Infect 45:108-12.

50. Park S, Lee K M, Yoo Y S, Yoo J S, Yoo J I, Kim H S, Lee Y S, Chung G T. 2011. Alterations of gyrA, gyrB, and parC and activity of efflux pump in fluoroquinolone-resistant Acinetobacter baumannii . Osong Public Health and Research Perspectives 2:164-170.

51. Mak J K, Kim M J, Pham J, Tapsall J, White P A. 2009. Antibiotic resistance determinants in nosocomial strains of multidrug-resistant Acinetobacter baumannii . J Antimicrob Chemother 63:47-54.

52. Hamouda A, Amyes S G. 2004. Novel gyrA and parC point mutations in two strains of Acinetobacter baumannii resistant to ciprofloxacin. J Antimicrob Chemother 54:695-6.

53. Vila J, Ruiz J, Goni P, Jimenez de Anta T. 1997. Quinolone-resistance mutations in the topoisomerase IV parC gene of Acinetobacter baumannii . J Antimicrob Chemother 39:757-62.

54. Karah N, Haldorsen B, Hegstad K, Simonsen G S, Sundsfjord A, Samuelsen Ø. 2011. Species identification and molecular characterization of Acinetobacter spp. blood culture isolates from Norway. Journal of Antimicrobial Chemotherapy 66:738-744.

55. Roberts M C, Schwarz S, Aarts H J. 2012. Commentary on “Acquired antibiotic resistance genes: an overview”. Frontiers in Microbiology 3.

56. Yeh K M, Lin J C, Yin F Y, Fung C P, Hung H C, Siu L K, Chang F Y. 2010. Revisiting the importance of virulence determinant magA and its surrounding genes in Klebsiella pneumoniae causing pyogenic liver abscesses: exact role in serotype K1 capsule formation. J Infect Dis 201:1259-67.

57. Tsai Y K, Liou C H, Lin J C, Ma L, Fung C P, Chang F Y, Siu L K. 2014. A suitable streptomycin-resistant mutant for constructing unmarked in-frame gene deletions using rpsL as a counter-selection marker. PLoS One 9:e109258.

58. Reyrat J-M, Pelicic V, Gicquel B, Rappuoli R. 1998. Counterselectable markers: untapped tools for bacterial genetics and pathogenesis. Infect Immun 66:4011-4017.