Patent Application
20230407243
A computer readable form of the Sequence Listing “6580-P68227US01_SequenceListing.xml” (5,017,998 bytes), submitted via Patent Center and created on Aug. 4, 2023, is herein incorporated by reference.
The present disclosure provides an Escherichia coli strain comprising inactivated genes acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. Also provided is a method for identifying a compound that is an antibacterial agent using an Escherichia coli strain disclosed herein. Further provided is a method for creating an Escherichia coli strain with one active efflux pump.
Gram-negative bacteria represent a serious challenge for antibacterial drug discovery efforts. The outer membrane (OM) is a formidable barrier for the entry of large and hydrophobic compounds, and the inner membrane (IM) reduces the influx of hydrophilic drugs. These two membranes augment the next line of defense, membrane-spanning efflux pumps, which effectively reduce the intracellular and periplasmic concentrations of compounds that have penetrated the cell. Synergy between influx retardation and active efflux contributes considerably to the intrinsic antibiotic resistome of Gram-negative pathogens.
Multidrug-resistance (MDR) efflux pumps, which typically extrude a wide range of structurally unrelated substances, have been particularly well studied. However, bacterial species harbor large networks of additional and often poorly characterized drug efflux pumps. For example, sequence annotation of the Escherichia coli K-12 genome highlighted the presence of 36 known or putative drug efflux pumps, which span five protein families: the ATP-binding cassette (ABC) superfamily, the resistance-nodulation-cell division (RND) superfamily, the major facilitator superfamily (MFS), the small multidrug resistance family (SMR), and the multi-antimicrobial extrusion (MATE) family. E. coli efflux pumps within the RND and ABC superfamilies complex with periplasmic adaptor proteins and the OM channel TolC. These tripartite complexes span the entire cell envelope. Certain MFS pumps, such as EmrB and EmrY, also form tripartite complexes with TolC. However, a majority of MFS members are single component pumps that extrude substrates to the periplasm. Efflux pumps from the MATE and SMR families are also single component efflux pumps. In the case of antibiotics with cytoplasmic targets, synergistic relationships are thought to exist between tripartite systems and single component IM pumps. In this instance, single component pumps extrude substrates to the periplasm, and tripartite assemblies then efflux to the exterior of the cell, which is often referred to as functional ‘interplay’.
Since efflux pumps are a major contributor to antibiotic resistance, delineating the substrate specificities and functions of these membrane-spanning proteins is critical for the development of strategies to compromise and/or circumvent these ancient resistance elements. In addition, while efflux pumps have largely been studied for their ability to extrude most classes of clinically important antibiotics, they are also increasingly associated with physiological functions. Indeed, conservation of the E. coli efflux system further supports such physiological functions. However, important questions remain regarding the physiological roles of these proteins. Overall, a major limitation hindering the study of bacterial efflux pumps has been the lack of a suitable genetic background. It is difficult to delineate the substrate specificities and functions of each pump due to the sheer number encoded within the genome, the differential expression of efflux pump-encoding genes, and functional redundancies. For example, in terms of functional redundancy, E. coli K-12 strains harbor six pumps that efflux tetracycline, which is proposed to provide a more robust and flexible defense mechanism.
To address these limitations, inventors generated and thoroughly characterized an extensively efflux-deficient mutant strain of E. coli. This strain provides a simplified genetic background free of the masking effects and redundancies of promiscuous efflux pumps. While a growing body of literature associates drug efflux pumps with important physiological processes, which suggests their removal could be detrimental or infeasible, inventors successfully inactivated 35 IM efflux pumps comprising the E. coli drug efflux network, generating Efflux KnockOut-35 (EKO-35). Phenotypic profiling of this strain revealed the E. coli drug efflux network is dispensable under optimal growth conditions, with little impact on the cellular proteome in nutrient-rich conditions. Importantly, when EKO-35 is propagated under diverse growth conditions, inventors reveal distinct patterns of dispensability, which opens the way for future studies to investigate the efflux pumps responsible for these conditionally essential phenotypes. To the best of inventors' knowledge, EKO-35 represents the most efflux-deficient bacterial mutant to be reported.
In addition to the important biological insight gained through generation of EKO-35, this strain can also be used as a well-characterized simplified genetic background to study the functions of efflux pumps of interest. To demonstrate the utility of EKO-35, inventors constructed an efflux platform consisting of EKO-35 genomic integrations of genes encoding E. coli efflux pumps forming tripartite complexes with the OM channel TolC. Each strain was profiled against a curated collection of physicochemically diverse compounds, which enabled the inventors to summarize molecular properties contributing to transport in each of these proteins. Inventors also profiled the MexCD pump from Pseudomonas aeruginosa, showing that the platform can be used to study efflux pumps from other organisms. Through the introduction of a large non-selective pore into the OM of EKO-35 (EKO-35-Pore), inventors demonstrate the efflux platform can be additionally utilized to study the specificity of efflux pump inhibitors, and to explore efflux pump interplay. Overall, EKO-35, the developed efflux platform, and the important insight gained into physicochemical substrate specificities and efflux essentiality, will have widespread application for the study of bacterial drug efflux pumps.
Accordingly, provided herein is an Escherichia coli strain comprising at least 20 of inactivated genes acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA.
In some embodiments, the strain comprises at least 25, at least 30 or more of the inactivated genes. In some embodiments, the strain comprises at least 34 of the inactivated genes. In some embodiments, strain comprises all 35 inactivated genes. In some embodiments, the Escherichia coli strain is deposited under International Depositary Authority of Canada (IDAC) accession number 310522-01 deposited on May 31, 2022, or IDAC accession number 070623-01 deposited on Jun. 7, 2023. In some embodiments, the Escherichia coli comprises a nucleic acid having the sequence as shown in SEQ ID NO: 255. In some embodiments, the strain further comprises an open variant of outer membrane ferric siderophore transporter FhuA. In some embodiments, the strain comprises at least one of reactivated acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA, optionally under the control of a constitutive promoter. In some embodiments, one of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA, is reactivated. In some embodiments, the inactivation comprises deletion of the gene, or introducing a mutation into the gene to ablate expression of the gene or eliminate efflux pump activity of the protein expressed by the gene. In some embodiments, acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA are genes encoding for efflux pumps. In some embodiments, the strain comprising at least 20 of inactivated genes acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA is an efflux pump deficient strain. In some embodiments, the strain comprises all 35 inactivated genes is EKO-35.
Also provided is a method for identifying a compound that is an antibacterial agent, comprising
In some embodiments, the decrease in viability of wild-type or the Escherichia coli strain described herein after contacting the compound is at most 10%, 20%, 30%, 40%, 50%, 60%, 70%, or 80%, or at least 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100%.
In some embodiments, the compound is identified as an antibacterial agent if the compound decreases the viability of efflux pump deficient Escherichia coli strain described herein at a faster rate than the decrease in viability of wild-type or the Escherichia coli strain comprising reactivated efflux pump genes.
In some embodiments, the compound decreases the viability of the Escherichia coli strain with one reactivated gene in EKO-35 is less than an efflux deficient strain disclosed herein, optionally Escherichia coli strain EKO-35, thereby identifying specificity of the compound for an efflux pump encoded by the reactivated gene, optionally the compound has a lower minimum inhibitory concentration (MIC) in the Escherichia coli strain of EKO-35 than EKO-35 with a reactivated efflux pump.
In some embodiments, the contacting comprises the Escherichia coli in a suitable culturing media for optimal Escherichia coli growth. In some embodiments, the contacting comprises incubating the Escherichia coli in nutrient-limiting culturing media. In some embodiments, the contacting comprises incubating the Escherichia coli for at least 24 h, 48 h, 72 h, or 96 h. In some embodiments, the culturing media is a media having a pH of about 2, about 3, about 4, or about 5.
Also provided is a method for creating an Escherichia coli strain producing one or more efflux pump, comprising reactivating one of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA in an efflux deficient strain disclosed herein, optionally Escherichia coli strain EKO-35.
In some embodiments, the reactivation comprises introducing one or more of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA, optionally by reversing the inactivation, optionally by re-introducing the gene back in genome with the same promoter or a different promoter, at the same locus or a different locus, optionally by introducing a knock down-resistant version of the gene.
Other features and advantages of the present disclosure will become apparent from the following detailed description. It should be understood, however, that the detailed description and the specific Examples while indicating preferred embodiments of the disclosure are given by way of illustration only, since various changes and modifications within the spirit and scope of the disclosure will become apparent to those skilled in the art from this detailed description.
Embodiments are described below in relation to the drawings in which:
Unless otherwise indicated, the definitions and embodiments described in this and other sections are intended to be applicable to all embodiments and aspects of the present disclosure herein described for which they are suitable as would be understood by a person skilled in the art.
In understanding the scope of the present disclosure, the term “comprising” and its derivatives, as used herein, are intended to be open ended terms that specify the presence of the stated features, elements, components, groups, integers, and/or steps, but do not exclude the presence of other unstated features, elements, components, groups, integers and/or steps. The foregoing also applies to words having similar meanings such as the terms, “including”, “having” and their derivatives. The term “consisting” and its derivatives, as used herein, are intended to be closed terms that specify the presence of the stated features, elements, components, groups, integers, and/or steps, but exclude the presence of other unstated features, elements, components, groups, integers and/or steps. The term “consisting essentially of”, as used herein, is intended to specify the presence of the stated features, elements, components, groups, integers, and/or steps as well as those that do not materially affect the basic and novel characteristic(s) of features, elements, components, groups, integers, and/or steps.
As used herein, the singular forms “a”, “an” and “the” include plural references unless the content clearly dictates otherwise.
The term “nucleic acid”, “nucleic acid molecule” or its derivatives, as used herein, is intended to include unmodified DNA or RNA or modified DNA or RNA. For example, the nucleic acid molecules of the disclosure can be composed of single- and double-stranded DNA, DNA that is a mixture of single- and double-stranded regions, single- and double-stranded RNA, and RNA that is a mixture of single- and double-stranded regions, hybrid molecules comprising DNA and RNA that may be single-stranded or, more typically double-stranded or a mixture of single- and double-stranded regions. In addition, the nucleic acid molecules can be composed of triple-stranded regions comprising RNA or DNA or both RNA and DNA. The nucleic acid molecules of the disclosure may also contain one or more modified bases or DNA or RNA backbones modified for stability or for other reasons. “Modified” bases include, for example, tritiated bases and unusual bases such as inosine. A variety of modifications can be made to DNA and RNA; thus “nucleic acid molecule” embraces chemically, enzymatically, or metabolically modified forms. The term “polynucleotide” shall have a corresponding meaning.
As used herein, the term “inactivation of a gene”, or a derivative thereof, refers to reduction or elimination in the activity of the protein encoded by a gene due to the reduction or elimination of the gene expression via mutation induced by one or more methods selected from the group consisting of deletion of all or a part of the corresponding gene, substitution of a part of the nucleotide sequence, or deletion or insertion of one or more base pairs into the nucleotide sequence.
As used herein, the term “reactivation”, or a derivative thereof, when relating to a gene, refers to increase or reintroduction in the activity of the protein encoded by a gene due to the increase or reintroduction of the gene expression via mutation induced by one or more methods selected from the group consisting of insertion of all or a part of the corresponding gene, substitution of a part of the nucleotide sequence, or deletion or insertion of one or more base pairs into the nucleotide sequence. Reactivation or reactivated includes restoring the gene as prior to inactivation, with previous promoter or with a different promoter, whether it a constitutive or conditional promoter. Reactivation can include tunable expression to control the level of the restored gene, including overexpression, returning to previous level or under-expressing as compared to previous levels. The reactivation, including reintroduction, can be at the same locus or at a different locus of the bacterial strain's genome. The reactivation, including reintroduction, of gene can include introduction of mutation that affect the function of the gene, for example, efflux pump function, including interaction with compounds or other genes. In some embodiments, reactivation of a gene comprises reintroduction of a gene. In some embodiments, reactivation of a gene occurs at the same locus, or at a different locus in a bacterial strain's genome. In some embodiments, reintroduction of a gene occurs at the same locus, or at a different locus in a bacterial strain's genome.
As used herein, the term “polypeptide” encompasses both peptides and proteins, and fragments thereof of peptides and proteins, unless indicated otherwise. In one embodiment, the therapeutic agent is a polypeptide.
The term “promoter,” as used herein, refers to a nucleotide sequence that directs the transcription of a gene or coding sequence to which it is operably linked.
The term “operably linked”, as used herein, refers to an arrangement of two or more components, wherein the components so described are in a relationship permitting them to function in a coordinated manner. For example, a transcriptional regulatory sequence or a promoter is operably linked to a coding sequence if the transcriptional regulatory sequence or promoter facilitates aspects of the transcription of the coding sequence. The skilled person can readily recognize aspects of the transcription process, which include, but not limited to, initiation, elongation, attenuation and termination. In general, an operably linked transcriptional regulatory sequence is joined in cis with the coding sequence, but it is not necessarily directly adjacent to it.
A “segment” of a nucleotide sequence is a sequence of contiguous nucleotides. A segment can be at least about 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 60, 85, 100, 110, 120, 130, 145, 150, 160, 175, 200, 250, 300, 350, 400, 450, 500 or more contiguous nucleotides.
A “fragment” of an amino acid sequence is a sequence of contiguous amino acids. A segment can be at least about 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 60, 85, 100, 110, 120, 130, 145, 150, 160, 175, 200, 250, 300, 350, 400, 450, 500 or more contiguous amino acids.
The term “antibacterial agent” as used herein refers to a microbial inhibiting agent, including anything that reduces virulence or modifies efflux pump activity, with or without the action of other compounds and adjuvants. For example, an antibacterial agent can be an efflux pump inhibitor.
The term “viability” as used herein refers to measurement known to the skilled person in assessing health of bacteria. Methods known in the art can be used to determine viability. Viability can be determined as a percentage over a control or as a minimal inhibitory concentration (MIC) when it is being affected by a compound, for example, an antibacterial agent such as an efflux pump inhibitor. Viability can also be determined relatively, for example, by comparing the rate of killing or inhibition of growth of the bacterial strain or wild-type bacteria described herein.
The present disclosure provides an Escherichia coli strain that is deficient in efflux pump activity. Accordingly, provided herein is an Escherichia coli strain comprising at least 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, or of inactivated genes acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the strain comprises at least 20 of the inactivated genes. In some embodiments, the strain comprises at least 21 of the inactivated genes. In some embodiments, the strain comprises at least 22 of the inactivated genes. In some embodiments, the strain comprises at least 23 of the inactivated genes. In some embodiments, the strain comprises at least 24 of the inactivated genes. In some embodiments, the strain comprises at least 25 of the inactivated genes. In some embodiments, the strain comprises at least 26 of the inactivated genes. In some embodiments, the strain comprises at least 27 of the inactivated genes. In some embodiments, the strain comprises at least 28 of the inactivated genes. In some embodiments, the strain comprises at least 29 of the inactivated genes. In some embodiments, the strain comprises at least 30 of the inactivated genes. In some embodiments, the strain comprises at least 31 of the inactivated genes. In some embodiments, the strain comprises at least 32 of the inactivated genes. In some embodiments, the strain comprises at least 33 of the inactivated genes. In some embodiments, the strain comprises at least 34 of the inactivated genes. In some embodiments, strain comprises all 35 inactivated genes. EKO-35v1 and EKO-35v2 are examples of EKO-35. In some embodiments, the Escherichia coli strain is EKO-35v1. In some embodiments, the Escherichia coli strain is EKO-35v2. In some embodiments, the Escherichia coli strain is deposited under IDAC accession number 310522-01 or IDAC accession number 070623-01. In some embodiments, the Escherichia coli strain is deposited under IDAC accession number 310522-01. In some embodiments, the Escherichia coli strain is deposited under IDAC accession number 070623-01. In some embodiments, the strain further comprises an open variant of outer membrane ferric siderophore transporter FhuA. In some embodiments, the strain further comprises deletion of tolC gene. In some embodiments, the strain comprises at least one of reactivated acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA, optionally under the control of a constitutive promoter. In some embodiments, one of reactivated acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the inactivation comprises deletion of the gene, or introducing a mutation into the gene to ablate expression of the gene or eliminate efflux pump activity of the protein expressed by the gene. In some embodiments, acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA are genes encoding for efflux pumps. In some embodiments, the strain comprising at least 20 of inactivated genes acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA is an efflux pump deficient strain. In some embodiments, the strain comprises all 35 inactivated genes is EKO-35. In some embodiments, the strain comprises a nucleic acid comprising at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.1%, 99.2%, 99.3%, 99.4%, 99.5%, 99.6%, 99.7%, 99.8%, 99.9%, 99.91%, 99.92%, 99.93%, 99.94%, 99.95%, 99.96%, 99.97%, 99.98%, 99.99%, 99.999%, or 99.9999% sequence identity as any nucleotide sequence described herein. In some embodiments, the strain comprises a nucleic acid comprising 100% sequence identity as any nucleotide sequence described herein. In some embodiments, the strain comprises a nucleic acid comprising at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.1%, 99.2%, 99.3%, 99.4%, 99.5%, 99.6%, 99.7%, 99.8%, 99.9%, 99.91%, 99.92%, 99.93%, 99.94%, 99.95%, 99.96%, 99.97%, 99.98%, 99.99%, 99.999%, or 99.9999% sequence identity as the nucleotide sequence as shown in SEQ ID NO: 255. In some embodiments, the strain comprises a nucleic acid comprising at least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.1%, 99.2%, 99.3%, 99.4%, 99.5%, 99.6%, 99.7%, 99.8%, 99.9%, 99.91%, 99.92%, 99.93%, 99.94%, 99.95%, 99.96%, 99.97%, 99.98%, 99.99%, 99.999%, or 99.9999% sequence identity as the nucleotide sequence as shown in SEQ ID NO: 255, and having least 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, or 35 of inactivated genes acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the strain comprises a nucleic acid comprising 100% sequence identity as the nucleotide sequence as shown in SEQ ID NO: 255.
The molecular tools for inactivating and reactivating genes are known in the art. In some embodiments, the reactivation of gene comprises reintroducing gene under control of a constitutive promoter.
Also provided is a method for identifying a compound that is an antibacterial agent, comprising
In some embodiments, the antibacterial agent is a microbial inhibiting agent that reduces virulence or modifies efflux pump activity, with or without the action of other compounds and adjuvants. In some embodiments, the antibacterial agent is an efflux pump inhibitor.
In some embodiments, the decrease in viability of wild-type or the Escherichia coli strain described herein after contacting the compound is at most 10%, 20%, 30%, 40%, 50%, 60%, 70%, or 80%, or at least 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100%.
In some embodiments, the compound is identified as an antibacterial agent if the compound decreases the viability of efflux pump deficient Escherichia coli strain described herein at a faster rate than the decrease in viability of wild-type or the Escherichia coli strain comprising reactivated efflux pump genes.
In some embodiments, the compound decreases the viability of the Escherichia coli strain with one reactivated gene in an efflux deficient strain disclosed herein, optionally Escherichia coli strain EKO-35, is less than the efflux deficient strain disclosed herein, optionally Escherichia coli strain EKO-35, identifying specificity of the compound for an efflux pump encoded by the reactivated gene, optionally the compound has a lower minimum inhibitory concentration (MIC) in the Escherichia coli strain of EKO-35 than EKO-35 with a reactivated efflux pump.
The skilled person recognizes optimal conditions to carry out methods for identifying antibacterial agent or growth condition for the E. coli strain described herein. For example, optimal aeration can include broth cultures grown with aeration at 220 rpm. For example, for growth profiling, microtiter plates can be incubated at 37° C. or 25° C. with continuous linear shaking at 600 rpm. Other optimal conditions, as well as nutrient-limited conditions, are described in the Example.
In some embodiments, the contacting comprises the Escherichia coli in a suitable culturing media for optimal Escherichia coli growth. In some embodiments, the contacting comprises incubating the Escherichia coli in nutrient-limiting culturing media. In some embodiments, the contacting comprises incubating the Escherichia coli for at least 24 h, 48 h, 72 h, or 96 h. In some embodiments, the culturing media is a media having a pH of about 2, about 3, about 4, or about 5.
Also provided is a method for identifying a compound that is an antibacterial agent, comprising
Also provided is a method for creating an Escherichia coli strain producing one or more efflux pump, comprising reactivating one of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA in the Escherichia coli strain EKO-35.
In some embodiments, the reactivation comprises introducing one or more of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA, optionally by reversing the inactivation, optionally by re-introducing the gene back in genome with the same promoter or a different promoter, at the same locus or a different locus, optionally by introducing a knock down-resistant version of the gene.
Also provided is a method for creating an efflux pump deficiency E. coli strain, the method comprises inactivating at least 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, or 35 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating at least 20 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating at least 21 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating at least 22 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating at least 23 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating at least 24 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating at least 25 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating at least 26 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating at least 27 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating at least 28 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating at least 29 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating at least 30 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating at least 31 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating at least 32 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating at least 33 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating at least 34 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating all 35 genes of acrB, acrD, acrF, mdtF, macB, emrB, mdtL, mdtK, bcr, ydeA, mdtM, yddA, yebQ, emrE, mdtD, sugE, ynfM, emrD, ydeF, mdtJ, ydiM, mdtB, mdlA, emrY, mdfA, fsr, mdtG, mdtH, yieO, mdlB, mdtO, yojI, yajR, ydhC, and cusA. In some embodiments, the method comprises inactivating genes in the following order: ΔyajR; mdtO; ydhC; emrE; yojI; mdtD, sugE; ynfM, emrD, ydeF; mdlA, emrY; mdtK, bcr; mdtG; mdtH, mdlB, macB, yddA; fsr; ydiM; yieO; mdfA; mdtM; mdtJ; emrB, mdtB, mdtL; yebQ, cusA; mdtF; ydeA; acrF; acrD, and acrB.
The following non-limiting Examples are illustrative of the present disclosure:
Bacterial strains and plasmids used in this disclosure are provided in Table 2. E. coli K-12 str. BW25113, the parental strain of the Keio Collection (Baba, T. et al., 2006) was used as the background for generation of EKO-35. Specifically, an ΔacrB mutant from the Keio Collection was used as the first deletion mutant. E. coli TOP10 or E. coli DH5a strains were used as routine cloning hosts. E. coli strains for resistance cassette amplification were obtained from the Keio Collection, P. aeruginosa PAO1 was provided by Dr. Cezar Khursigara (University of Guelph). An E. coli K-12 str. BW25113 harboring the fhuA ΔC/Δ4L gene under the control of the constitutive synthetic promoter BBa_J23104 was used as a source for the ‘Pore’ (Johnson, J. W. et al., 2022). Plasmids for CRISPR-Cas9 mediated counterselection, pCas and pTargetF, were purchased from Addgene (Jiang, Y. et al., 2015). Plasmids for the λ-Red recombinase system, pKD46 and pCP20 (Datsenko, K. A. & Wanner, B. L, 2000), and expression of efflux genes, pINT2 and pGDP2 were used (Cox, G. et al., 2017). Strains were routinely grown in Lysogeny broth (LB) (Bioshop) at 37° C. or For optimal aeration, broth cultures were grown with aeration at 220 rpm. For growth profiling, microtiter plates were incubated at 37° C. or 25° C. with continuous linear shaking at 600 rpm. For susceptibility testing, microtiter plates were grown at 37° C. with continuous linear shaking at 900 rpm. Ampicillin (100 μg/mL) (Bioshop), kanamycin (50 μg/mL) (Sigma-Aldrich), spectinomycin (50 μg/mL) (Bioshop), and gentamicin (10 μg/mL) (BioBasic) were used at the listed concentrations for selection of resistance markers. For nutrient-limited conditions, strains were grown in M9 (Bioshop) supplemented with 2 mM MgSO4 (Bioshop), 0.1 mM CaCl2) (Bioshop) and glucose (w/v) (Bioshop). To profile growth in amino acid-limited medium supplemented with iron, MOPS medium (TEKNOVA) was utilized. Buffered “low-salt” medium was prepared (Lewinson, O. et al., 2014) with 100 mM of the appropriate buffer [pH 5.0, homopiperazine-N,N=-bis-2-(ethanesulfonic acid) (HOMOPIPES), pH 5.5 to 6.5, 2-(N-morpholino)ethanesulfonic acid (MES); pH 7.0 to 7.5, 3-(N-morpholino)propanesulfonic acid (MOPS)]. For profiling growth at pH 8.0 to 9.0, 1,3-bis(tris(hydroxymethyl)methylamino)propane (Bis-tris propane) was used at 50 mM due to toxicity at higher concentrations. Susceptibility testing was conducted in cation-adjusted Mueller Hinton II Broth (MHB II) (BD Difco).
aP-value
aP-value
aP-value for EKO-35 compared to EKO-35 araC::MdtEF
E. coli K-12 str.
E. coli TOP10
E. coli DH5a
E. coli EKO-35
E. coli δtolC
E. coli pore
P. aeruginosa PAO1
Generation of EKO-35 was achieved using a combination of the λ-Red recombinase system (Datsenko, K. A. & Wanner, B. L., 2000) and CRISPR-Cas9 counter-selection (Jiang, Y. et al., 2015). The efflux genes were inactivated in the order denoted in Table 3. All PCR reactions and restriction enzyme digests were prepared according to manufacturers' guidelines. Amplicons were purified using a GeneJET PCR purification kit (Thermo Fisher Scientific) according to manufacturer's guidelines. The 2×GB-AMP™ high-fidelity PaCeR™ polymerase Master Mix (GeneBio Systems Inc) and Taq 2× polymerase Master Mix (FroggaBio) were used according to the manufacturer's suggested guidelines.
For λ-Red recombineering, electrocompetent cells of the mutant strain of interest were transformed with 50 ng of pKD46 (Datsenko, K. A. & Wanner, B. L., 2003). A broth culture was grown to the mid exponential phase (OD600 nm˜0.5) in the presence of 2 mM arabinose and ampicillin to induce recombinase expression. Utilizing the PaCeR™ high-fidelity polymerase, and primers annealing 50 base pairs (bp) upstream and downstream of the gene of interest (Table 4), kanamycin resistance cassettes were amplified from the respective Keio strain. Amplicon size (1500 bp) was verified via gel electrophoresis and the remaining PCR product was purified. Recombinase induced electrocompetent cells were transformed with 250 ng of each amplicon (Bio-Rad MicroPulser, Ec1 setting, 1 mm electroporation cuvette (Fisher)). The cells were recovered and grown overnight on selective agar (LB with kanamycin) to identify gene disruptions. Successful gene knockouts were transformed with pCP20 (Table 2 “Strains and Plasmid)). A single colony was then inoculated into 3 mL of LB containing ampicillin and incubated at 30° C. to induce removal of the resistance cassette. Cassette removal was confirmed using PCR and primers annealing 250 bp upstream and downstream of the gene of interest (Table 4). Amplicon size (500 bp) for successful cassette removal was verified via gel electrophoresis. Efflux genes inactivated using the λ-Red recombinase system are indicated in Table 3.
For CRISPR-Cas9-mediated counterselection, the methodology described by Jiang et al. was modified for high-throughput screening of mutants (Jiang, Y. et al., 2015). CRISPR guide software (Benchling) was employed for selection of appropriate N20 sequences. pTargetF was modified via PCR to introduce an N20 for the gene of interest (Table 4). Amplicon size (2100 bp) was verified via gel electrophoresis, and the remaining PCR product was purified. To enable rapid screening of positive mutants and to disrupt the target gene, ssDNA repair oligos (˜100 bp in length) were designed to contain an AseI restriction site and three tandem stop codons (Table 4). All ssDNA repair oligos were purchased through Integrated DNA Technologies (IDT). Electrocompetent cells of the mutant strain of interest were transformed with 50 ng of pCas. A broth culture of each strain was grown to the mid exponential phase (OD600 nm˜0.5) in the presence of kanamycin and 10 mM arabinose to induce recombinase expression. To recombinase induced electrocompetent cells, 100 ng of pTargetF that was modified to contain the desired N20 sequence, and 2000 ng of repair ssDNA targeting the gene of interest were electroporated (Bio-Rad MicroPulser, Ec1 setting, 1 mm electroporation cuvette (Fisher)). The cultures were recovered in LB at 30° C. and propagated on selective agar (LB with kanamycin and spectinomycin) to identify successful gene disruptions. For high-throughput screening of colonies, Taq polymerase was used with primers annealing to the target region of each gene (Table 4). The amplicons were digested with AseI and successfully inactivated genes were identified via gel electrophoresis by digestion relative to a wild-type negative control. Insertion of the three tandem stop codons into the gene of interest was verified using Sanger sequencing at the Advanced Analysis Centre (AAC) (University of Guelph). Genes disrupted using CRISPR-Cas9-mediated counter selection are indicated in Table 3.
AG
GAG
GAG
AG
GAG
TTA ATT ATA TTT CAG GCC AAC GGC CCA TAC
AG
GAG
TTA ATT GAC GCG CAT AAA TTG CGG CGT ACC
AG
GAG
AG
GAG
TTA ATT AAG AAG GGC GCT GGT AGA GCG TCG
AG
GAG
AG
GAG
TTA ATT ATT AGT CTG CCA AAG TGT CTC CAG
AG
GAG
TTA TTA ATT CGG ATA GTC CAC TTC CGG CAG
AG
AGT
ATT GAC AGC CCC GAG ATA GCG ACG CCA TTC
AG
GAG
AG
GAG
AG
GAG
AG
GAG
AG
GAG
AG
GAG
AG
GAG
ATT TTT ACG CCA CGG CGA ACC GTA CGC TAA
AG
GAG
AG
GAG
AG
TGAG
AG
GAG
AG
GAG
AG
GAG
Genomic DNA was extracted using the One-4-All Genomic DNA Miniprep Kit (BioBasic), according to the manufacturer's guidelines. Quality of the extracted gDNA was assessed using gel electrophoresis. Illumina DNA library preparation was performed using an Illumina Nextera kit by the Microbial Genome Sequencing Center (Pennsylvania, USA), which was followed by Illumina sequencing on a NextSeq 2000 platform. Analysis of the raw reads was performed using Geneious Prime 2021.0.2 (Kearse, M. et al., 2012). Low quality reads were trimmed using an in-suite BBDuk plug-in. Raw wild-type reads were assembled to an NCBI reference genome (Accession No. CP009273.1) with bowtie2. The resulting assembly was used as a reference to assemble the EKO-35 mutant reads. Differences between the wild-type BW25113 and EKO-35 strains were identified by searching for single nucleotide polymorphisms (SNPs) and deletions using the following thresholds: minimum variant frequency of 0.75, maximum variant P-value of 10−6, and minimum variant P-value of 10−5. 11 mutations were identified, including a mutation in hdfR (T806C, L269P). CRISPR-Cas9-mediated counter selection was utilized to repair the mutation in hdfR, which introduced two intentional silent mutations (hdfR C722G and A838G) to remove the adjacent PAM site and AseI-guided screening purposes as described above. For the EKO-35-Pore strain, genomic DNA extraction, sequencing, analysis, genome assembly, and mutation identification was performed as described above. Whole genome sequencing was deposited in the GenBank database (BioProject ID PRJNA838981).
Construction of the pINT2 Efflux Gene Library
Inventors first attempted to generate marker less integrations of efflux-encoding genes into araC using the pINT1 plasmid (Cox, G. et al., 2022), under the constitutive control of the strong PBIa promoter. However, inventors observed numerous deleterious mutations following ligation into this vector and genomic integration, indicating high expression levels were not feasible. Consequently, the pINT2 plasmid was selected for the expression of efflux pump encoding genes from the same constitutive PLacI promoter. This plasmid enables single-copy genomic integration of a selected gene through integration into the nonessential araC gene within the arabinose operon (Cox, G. et al., 2022). All E. coli genes were amplified from E. coli BW25113 genomic DNA, and mexCD was amplified from P. aeruginosa PAO1, using a high-fidelity polymerase, followed by ligation into pINT2. Successful clones were verified using PCR (Primers: pLac-Fwd/pINT-Rev, Table 4) and confirmed via Sanger sequencing at The Centre for Applied Genomics (TCAG) (The Hospital for Sick Children) and the AAC (University of Guelph). For the interplay studies, this process was repeated using the pGDP2 plasmid (Cox, G. et al., 2022).
For genomic integration, genes of interest and the adjacent kanamycin resistance cassette were amplified from pINT2 using a high-fidelity polymerase and the F1B-Fwd/F4B-Rev or F1C-Fwd/F4C-Rev primers (Table 4). For the integration of efflux pump-encoding genes, electrocompetent EKO-35 cells were transformed with ng of pKD46. Electrocompetent recombinase-induced cells were transformed with 500 ng of each amplicon (BioRad MicroPulser, Ec1 setting, 1 mm electroporation cuvette (Fisher). Genomic integrations were verified via PCR (Primers: PolB-Fwd/pINT-Rev, Table 4). Successful integrations were transformed with pCP20 to remove the kanamycin resistance cassette. Prior to phenotypic profiling, all integrated genes, including their promoters, were verified using Sanger sequencing (TCAG, The Hospital for Sick Children).
For genomic integration of the pore, the fhuA ΔC/Δ4L gene and the adjacent gentamicin resistance cassette were amplified using a high-fidelity polymerase and the fhuA_40_Up/gImS_3680_Low primers (Table 4). Integration of the pore gene was performed as described above. The pore gene was introduced into the intergenic region between the glmS and pstS genes, with gene expression under the control of a constitutive promoter (Kearse, M. et al., 2012). Genomic integrations were verified using PCR (Primers: pstS_520_Up/glmS_3680_Low, Table 4). Successful integrations were transformed with pCP20 to remove the gentamicin resistance cassette as previously described. Activity of the pore was confirmed through susceptibility testing with the large antibiotic vancomycin prior to phenotypic profiling and further susceptibility testing.
To enable comparison between efflux production in different genetic backgrounds, each efflux pump-encoding gene was also integrated into the genome of the wild-type BW25113 strain, the single gene deletion mutant from the Keio Collection, and EKO-35. Strains of interest were propagated on LB agar at 37° C. Cell inoculum was prepared using the colony resuspension method and applied to a 96-well well plate (VWR) in technical duplicate, and the minimum inhibitory concentrations were determined according to Clinical & Laboratory Standards Institute (CLSI) protocols in MHB II (Patel J. B. et al, 2015). The plates were incubated for 18 h (Multitron Shaker, Infors HT) at 37° C. with aeration at 900 rpm. The plates were equilibrated to room temperature before the OD600 nm was measured using a BioTek Synergy H1 microplate reader. For interplay profiling, blank subtracted OD600 nm values >0.1 were considered to represent growth.
For growth profiling in nutrient-rich conditions, strains were propagated on LB agar for 18 h at 37° C. Single colonies were inoculated into LB and grown at 37° C. until the mid-exponential phase (OD600 nm˜0.6) was reached. All strains were assessed with at least three biological replicates. The cultures were standardized to an OD600 nm˜0.1 in sterile 0.85% saline (w/v). Standardized cultures were diluted 1/200 into LB and 100 μL of the resulting dilution were applied to round-bottom 96-well microtiter plates (VWR). To prevent evaporation, the microtiter plates were sealed (labeling tape, Fisher Scientific). The OD600 nm was measured every 15 minutes over the course of 24 h using a BioTek Synergy H1 microplate reader. Growth was assessed at both 37° C. and 25° C.
For growth profiling in nutrient-limited conditions, strains were propagated in biological triplicate on LB agar at 37° C. Multiple single colonies were suspended, using three separately prepared inoculums per strain, in sterile 0.85% saline (w/v) to an OD600 nm˜0.1. Standardized cultures were diluted 1/200 in fresh M9 and 100 μL of the resulting dilution was applied to round-bottom 96-well microtiter plates (VWR). The plate was sealed (labeling tape, Fisher Scientific) to prevent evaporation. Growth was assessed as described above, at both 37° C. and 25° C., for 48 h.
For physiological profiling in low-oxygen conditions, strains were prepared in LB or M9 as described above. To a round-bottom 96-well microtiter plate (VWR, tissue culture treated), 100 μL of the diluted cultures were applied in triplicate. Prior to incubation, the sample plate was placed into an anaerobic jar (Oxoid™ AnaeroJar, Thermo Fisher Scientific) with an anaerobic gas generator (AnaeroPack™ Thermo Fisher Scientific) for 30 min. The plates were incubated in a BioTek Synergy H1 plate reader, equipped with an 02 gas controller. Using nitrogen gas, the oxygen levels in the plate reader were maintained at 1% (LB) and 5% (M9). Growth was assessed as described above.
To assess fitness of EKO-35 expressing wild-type copies of the genes identified to harbor nonsynonymous mutations, Inventors obtained clones from the E. coli ASKA library (Kitagawa, M. et al., 2005) with the rspA, tufA, pitA, wcaC, gyrB, and yjfC genes carried on plasmids, with gene expression under the control of an IPTG-inducible promoter. The ASKA plasmids were verified using Sanger sequencing and gene expression was induced with 0.1 mM IPTG. In the wild-type E. coli K-12 strain, the plasmids were maintained with 25 μg/mL chloramphenicol. Due to the changes in susceptibility of the efflux-deficient strains, in EKO-35 the plasmids were maintained with 4 μg/mL and 1 μg/mL chloramphenicol in Lysogeny broth and M9 minimal glucose medium, respectively.
Overnight cultures were propagated in LB at 37° C. for 18 h. Saturated cultures were diluted 1/100 into LB or M9. To a flat-bottom 96-well microtiter plate (CoStar, untreated polystyrene), 150 uL of diluted culture was applied in triplicate. The plates were sealed to reduce evaporation and incubated statically at 37° C. for 24 h (nutrient-rich media) or 48 h (nutrient-limited media). To assess the effect of the nonsynonymous mutations on biofilm formation, this procedure was repeated with the addition of 0.1 mM IPTG for plasmid induction and chloramphenicol for plasmid maintenance at the concentrations specified above.
Following incubation, the optical density (OD600 nm) of the samples were measured using a BioTek Synergy H1 microplate reader. The cultures were aspirated, and the plates were washed in triplicate with 300 uL of deionized water using a BioTek ELx405 microtiter plate washer. Each well was stained with 175 uL of (w/v) crystal violet (CV), followed by static incubation at room temperature for 20 min. The CV was removed, and the plates were washed with deionized water until no excess stain was present. The plates were allowed to air dry before 175 uL of 0.7% acetic acid was applied to each well, followed by incubation at room temperature for min to solubilize the CV. The absorbance of each sample was measured at 595 nm using a BioTek Synergy H1 microplate reader.
EKO-35 strains containing the chromosomally integrated genes of interest were propagated on LB agar at 37° C. for 18 h. Single colonies were used to inoculate 3 mL of LB followed by incubation at 37° C. for 18 h with aeration at 220 rpm. The bacterial cells were harvested by centrifugation and resuspended in phosphate buffered saline (PBS) ([pH 7.4], 137 mM NaCl (Bioshop), 2.7 mM KCl (Bioshop), 10 mM Na2PO4 (Bioshop) and 1.8 mM K2PO4) to an OD600 nm˜1.0, followed by 1/2000 dilution into MHB II.
To a 384-well microtiter plate (Corning, untreated polystyrene), 500 nL of compound was dispensed, in duplicate, and serially titrated 2-fold using an Echo 550 acoustic liquid dispenser (Labcyte Inc.). For compounds dissolved in DMSO, the concentration of DMSO did not exceed 1% of the final well volume. As a solvent control, 500 nL of DMSO was applied to wells that contained no compound. Using a multichannel pipette, 50 μL of the prepared bacterial inoculum was applied to each well. To enable background correction post-incubation, the OD600 nm was measured (Biotek Synergy Neo2 microplate reader) prior to incubation. The plates were incubated at 37° C. with aeration at 900 rpm for 18 h. The plates were equilibrated to room temperature and the OD600 nm measured using a BioTek Synergy Neo2 microplate reader. Data was analyzed in both Prism (GraphPad, Version 9.2.0) and Microsoft Excel (Version 16.53). Raw data were input into Microsoft Excel and the pre-incubation OD600 nm measurements were subtracted from the post-incubation OD600 nm measurements. These corrected measurements were input into Prism 9 as a grouped analysis. MIC values for each compound and the corresponding strain were normalized using Prism 9, where the highest MIC value per compound represented 100%. All other MIC values were adjusted accordingly and visualized using the single-color scale grouped heat map function. Identification of compound chemical properties was achieved using DataWarrior (Version 5.5.0). Using PubChem, the SMILES chemical notation for each compound was obtained and compiled in Microsoft Excel. The resulting spreadsheet was input into DataWarrior and properties were calculated from the compound chemical structures.
Synergy measurement using checkerboard analysis was performed in 96-well microtiter plates using the microdilution broth method, according to CLSI guidelines (Patel J. B., 2015). The cell inocula were prepared using the colony resuspension method. The minimum inhibitory concentrations (MICs) were defined as the lowest concentration that provided no growth, as determined by measurement of the OD600nm using a Synergy H1 microplate reader (BioTek). Susceptibility testing was performed in MHB II with a final volume of 100 μL. PAI3N (Bachem Americas) was solubilized in MHB II. When assessing PAI3N synergy, linezolid, oxacillin, fusidic acid, and erythromycin were solubilized in 100% DMSO. Ciprofloxacin and novobiocin were solubilized in distilled water. NMP, linezolid, oxacillin, fusidic acid were solubilized in 50% DMSO (v/v), and erythromycin in 50% ethanol (v/v). Ethidium bromide and ciprofloxacin were solubilized in distilled water. Solvent controls were included in each plate.
The Fractional Inhibitory Concentration Index (FICI) was used to assess the synergy of PAI3N in combination with different antibiotics. The FICI represents the ΣFIC of each drug. The FICI for each drug was calculated using the following formula: FICI=FICA+FICB=(CA/MICA)+(CB/MICB). Where MICA and MICB are the MICs of drugs A (PAI3N/NMP) and B (antibiotic) alone, and CA and CB are the MICs of the drugs in combination. The effects of PAβN or NMP in combination with antibiotics were classified as: synergistic (FICI<0.5), additive (FICI>0.5-1.0), indifferent (FICI>1.0-2.0), and antagonistic (FICI>2.0). Fold increases in resistance provided by efflux pumps were calculated by dividing the MIC values of strains expressing efflux pumps by the MIC value of EKO-35.
Overnight cultures of EKO-35, EKO-35 araC::acrB, and EKO-35 araC::acrBD408A were subcultured (1:100) into 200 mL LB and incubated at 37° C. with 220 rpm shaking until an optical density (OD600 nm) of 0.6 was reached. The cells were harvested by centrifugation (5,000×g) and resuspended in 100 mM Tris HCl [pH 7.0] with 150 mM NaCl. The cells were lysed by sonication and the sonicate was cleared at 4,000×g for 10 min, followed by removal of occlusion bodies at 20,000×g for 20 min. Crude membranes were isolated by ultracentrifugation at 100,000×g for 1 h. Membrane fractions were resuspended in 100 mM Tris HCl [pH 7.0] with 150 mM NaCl and 2% sodium dodecyl sulphate (SDS). Fractions were quantified using a BCA assay (Thermo Fisher). 10 μg of each sample was resolved by SDS-PAGE and transferred to Amersham™ Protran™ 0.45 μM nitrocellulose membrane (GE) for Western blot analysis using a polyclonal rabbit anti-AcrB (Hazel, A. J. et al., 2019) and an anti-Rabbit IgG(H+L) horseradish peroxidase (HRP) conjugated secondary antibody (Invitrogen). Visualization was performed using the Luminata Crescendo Western chemiluminescent HRP substrate (Millipore) and a Bio-Rad ChemiDocXRS+ system. Total protein normalization was achieved using Bio-Rad stain-free gel imaging and ImageLab (version 6.1).
Saturated LB overnight cultures of the wild-type and EKO-35 strains were inoculated (1/100 dilution) into 50 mL of LB until the mid-exponential phase of growth was reached (OD600 nm=0.5). 3×10 9 cells were harvested by centrifugation (4000×g), washed twice in phosphate-buffered saline (PBS), and the cell pellet was flash frozen in liquid nitrogen. Whole-cell proteome samples were generated using a modified total proteome extraction protocol (Rappsilber, J. et al., 2007). The cell pellets were stored at −80° C. until the experiment was performed. Briefly, the cell pellet was resuspended in ice cold 100 mM Tris-HCl [pH 8.5]. The cells were lysed using a probe sonicator (three cycles of 30 sec on/off, 30% amplitude) and the samples were treated with 2% SDS and 10 mM dithiothreitol (DTT). To enrich membrane proteins, bacterial membranes were isolated as escribed above. To minimize contamination with cytoplasmic proteins, the membrane pellet was washed in ice cold 100 mM Tris-HCl [pH 8.5], followed by ultracentrifugation. The membrane pellet was resuspended in 100 mM Tris-HCl [pH 8.5] with 2% SDS and 10 mM dithiothreitol (DTT). Both the whole-cell and membrane samples were heated to 95° C. prior to treatment with 55 mM iodoacetamide (IAA). The samples were then incubated with 100% acetone overnight at −20° C. Precipitated protein from both the whole-cell and membrane fractions were combined by centrifugation at 10,000×g at 4° C. and washed twice with 80% acetone. The protein pellets were then solubilized in 40 mM HEPES [pH 5.0] with 8 M urea and quantified using a bovine serum albumin (BSA) tryptophan assay. The samples were digested overnight at room temperature with LysC and trypsin proteases (Promega, protein/enzyme ratio 50:1). 10% v/v trifluoroacetic acid (TFA) was then added and 50 μm acidified peptides were purified and desalted using a STop And Go Extraction (STAGE) tip with 3 layers of C18 resin using the described protocol (Rappsilber, J. et al., 2007).
Dried peptides were suspended in Buffer A (2% (v/v) acetonitrile, 0.1% (v/v) trifluoroacetic acid, 0.5% (v/v) acetic acid) and 25 ng of peptides were analyzed on a Q Exactive™ HF-X hybrid quadrupole-orbitrap mass spectrometer (ThermoFisher Scientific) coupled to an EasynLC™ 1200 High-Performance Liquid Chromatography (ThermoFisher Scientific). The samples were loaded onto an in-line 75 μm×50 cm PepMap RSLC EASY-Spray column filled with 2 μm C18 reverse-phase silica beads (ThermoFisher Scientific). Peptides were separated and directly electrosprayed into the mass spectrometer using a linear gradient from 3 to 20% Buffer B (80% (v/v) acetonitrile, 0.5% (v/v) acetic acid) over 18 min, from 20 to 35% Buffer B over 31 mins, followed by a steep 2 min ramp to 100% Buffer B for 9 min in 0.1% formic acid at a constant flow of 250 nL/min. The mass spectrometer was operated in data-dependent mode, switching automatically between one fill scan and subsequent MS/MS scans of 30 most abundant peaks, with full-scans (m/z 400-1600) acquired in the Orbitrap analyzer with a resolution of 60,000 at 400 m/z.
Raw mass spectrometry files were analyzed using MaxQuant (ver. 1.6.14.0 (Cox, J. & Mann, M., 2008). The spectra were searched using the Andromeda search engine with the E. coli K-12 proteome as reference (Accession No. P000000625, accessed June 2021 with 4391 sequences). A minimum of two distinct peptides were required for protein identification and the FDR was set to 1%. The ‘match between runs’ feature of MaxQuant was enabled. Quantification was performed by label-free quantification (LFQ) using the MaxLFQ algorithm (Cox, J. et al., 2014). The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD033975. All mass spectrometry experiments were performed in biological quadruplicate.
In nutrient-rich conditions, a total of 1,979 proteins were identified from whole cell extracts of the wild-type strain, which represented 45% of the predicted proteome. Principal component analysis (PCA) separated EKO-35 from the wild-type strain (Component 1, 40.6%), with slight variation observed between the biological replicates (Component 2, 17.2%) (
Statistical analysis and data visualization was performed using Perseus (version 1.6.2.2) (Cox, J. & Mann, M. 2012). Further analysis was only performed on proteins present in at least three out of four biological replicates within each strain. Upon filtering, missing values were imputed from a normal distribution. Significant changes in abundance of proteins between the two proteomes were defined with a false discovery rate (FDR)-correct Student's t-test (p-value 0.05) and Benjamini-Hochberg multiple hypothesis correction testing (FDR=0.05) was applied. 1D-annotation enrichment analysis of UniProt keywords enabled global assessment of changes in protein abundance (Tyanova, S. et al., 2016) using two-sample t-tests with P-value 0.05, FDR=0.05 and score <−0.5, <0.5. Data visualization was performed using GraphPad (version 8).
Generation of EKO-35: Inactivation of E. coli Drug Efflux Pumps
Inventors' first goal was to generate a simplified genetic background to overcome the challenges associated with the complexities of intact drug efflux networks. Using a combination of λ-Red recombineering and CRISPR Cas9-mediated counter selection. Inventors sequentially inactivated 36 genes encoding IM pumps from the genome of E. coli K-12 strain BW25113. Inventors started with an ΔacrB mutant from the Keio Collection, and then removed a further 12 pumps using the λ-Red system. One limitation of this approach is the possibility of unintended genomic deletions due to the introduction of adjacent Flp recognition target sites following the removal of numerous genes. Thus, the remaining genes were inactivated using CRISPR Cas9-mediated counter selection, and the introduction of three tandem stop codons into the beginning of each gene (Table 3).
The efflux genes were inactivated in the following order: ΔacrB; acrD; acrF; mdtF; macB; emrB; mdtL; mdtK; bcr; ydeA; mdtM; yddA; yebQ; emrE; mdtD; sugE; ynfM; emrD; ydeF; mdtJ; ydiM; mdtB, mdlA; emrY; mdfA; fsr; mdtG; mdtH; yieO; mdlB, mdtO, yojH, yojI, yajR, ydhC; cusA. While generating the strain, inventors identified a missense mutation (L269P) within the gene encoding the Hns-dependent flhDC regulator (HdfR), which inventors deduced was present in the acrB mutant used as the starting background of EKO-35. To ascertain whether the hdfR mutation occurred in response to loss of acrB, inventors removed acrB from the genome of the wild-type E. coli BW25113 strain using the λ-Red system. Sequencing of the hdfR gene revealed the mutation was not present (data not shown). Since HdfR is a transcriptional regulator of important physiological processes, and the mutation did not appear to be induced by loss of acrB, inventors repaired this mutation using CRISPR Cas9-mediated counter selection.
Since many of the efflux pump-encoding genes have predicted start codons and are poorly characterized, it is possible that alternative start codons located downstream from the inserted tandem stop codons (Table 3) could be utilized for a subset of the CRISPR-inactivated pumps. To investigate, inventors profiled the genome of the efflux-deficient strain, which included the inserted stop codons, using the Prodigal prokaryotic gene recognition and translation initiation site identifier. Overall, the program did not predict production of any potentially functional efflux pumps, which supports the notion that the tandem stop codons were sufficient to prematurely terminate translation, and that alternate start codons would not be utilized (Tables 26A and 26B). Indeed, Prodigal analysis of the wild-type strain's genome predicted production of all pumps (Tables 26A and 26B).
In addition, inventors also took advantage of new developments in protein structure predictions, and carefully analyzed AlphaFold-generated models of each efflux pump to ascertain whether the predicted start codons were correct based on the structural features of the proteins. Such an analysis indicated the inserted stop codons were sufficient to inactivate the different genes; however, the modeled structure of a predicted ABC efflux pump (YojH) indicated that the yojH gene does not encode an efflux pump. While yojH is located adjacent to another ABC efflux pump, YojI, which does structurally resemble an ABC transporter, the predicted YojH structure lacks the structural features of an ABC-type efflux pump, including transmembrane helices. Investigating further, inventors observed that YojH is described as malate:quinone oxidoreductase, a membrane-associated enzyme involved in the citric acid cycle/glyoxylate cycle. However, the function of this protein in E. coli is poorly described and the enzyme has not been shown to catalyze oxidation of malate in E. coli. Since the gene does not encode an efflux pump, inventors repaired the inactivated yojH gene by removing the inserted stop codons using CRISPR Cas9-mediated counter selection. This strain was subsequently designated Efflux KnockOut-35 (EKO-35), and was used for phenotypic characterization and construction of an efflux platform, as described below. The EKO-35 genome sequence confirmed successful disruption of the 35 efflux-encoding genes, including successful repair of hdfR and yojH, and also revealed ten additional secondary mutations (
With the intent of using EKO-35 as a simplified genetic background to study the functions of drug efflux pumps, inventors' second goal was to fully characterize EKO-35 prior to further investigations. Other efflux-deficient backgrounds (e.g., tolC mutants) display pleiotropic phenotypes, including severe growth defects in minimal M9 medium with glucose supplied as the carbon source. Indeed, tolC inactivation causes periplasmic accumulation of enterobactin in this low-iron growth medium, which underlies this conditionally essential phenotype. In addition, tolC mutants exhibit membrane stress and depletion of essential metabolites, resulting in ‘metabolic shutdown’ in minimal glucose medium. Thus, inventors were curious to ascertain whether loss of 35 IM efflux pumps also affected the fitness of E. coli.
First, inventors generated a control strain that was used as a comparator during phenotypic profiling of EKO-35. RND efflux pumps, such as AcrB, are trimeric complexes spanning the IM up to 36 times. It has been suggested that caution should be taken when interpreting phenotypes of gene deletion mutants since such phenotypes could be due to loss of integral membrane proteins, rather than loss of efflux function. Based on previous studies, inventors introduced a D408A substitution into AcrB to enable the production of an efflux inactive variant in EKO-35. Using the pINT2 plasmid, the mutated acrB gene was integrated into the arabinose operon of EKO-35 (EKO-35 araC::acrBD408A), with gene expression under the control of the constitutive PLacI promoter. Inventors confirmed the AcrBD408A protein was inactive and present at comparable levels to the wild-type protein within the membrane of EKO-35 (
Next, while the successful generation of EKO-35 showed the E. coli drug efflux system is dispensable for growth, inventors measured the growth kinetics and assessed the cellular morphology of EKO-35 under commonly used laboratory conditions: nutrient-rich (Lysogeny broth) and nutrient-limited (M9 defined minimal medium with glucose) medium at 37° C. (
Next, to explore whether the nonsynonymous EKO-35 mutations (Table 5 and Table 6) were compensatory, and to ascertain if they impacted any of the observed phenotypes detailed in this disclosure, inventors obtained clones from the E. coli ASKA library harboring the rspA, tufA, pitA, wcaC, gyrB, and yjfC genes on plasmids with gene expression under the control of an IPTG-inducible promoter. Growth profiling revealed pitA overexpression conferred a lethal phenotype, and the overexpression of the remaining genes induced a fitness cost in EKO-35 under nutrient-rich optimal growth conditions, indicating that the mutations could be compensatory in nature (
With the intent of using EKO-35 as a simplified genetic background to study the functions of drug efflux pumps, inventors' second goal was to fully characterize EKO-35 prior to further investigations. Other efflux-deficient backgrounds (e.g., tolC mutants) display pleiotropic phenotypes, including severe growth defects in minimal M9 medium with glucose supplied as the carbon source. Indeed, tolC inactivation causes periplasmic accumulation of enterobactin in this low-iron growth medium, which underlies this conditionally essential phenotype. In addition, tolC mutants exhibit membrane stress and depletion of essential metabolites, resulting in ‘metabolic shutdown’ in minimal glucose medium. Thus, inventors were curious to ascertain whether loss of 35 IM efflux pumps also affected the fitness of E. coli.
First, inventors generated a control strain that was used as a comparator during phenotypic profiling of EKO-35. RND efflux pumps, such as AcrB, are trimeric complexes spanning the IM up to 36 times. It has been suggested that caution should be taken when interpreting phenotypes of gene deletion mutants since such phenotypes could be due to loss of integral membrane proteins, rather than loss of efflux function. Based on previous studies, inventors introduced a D408A substitution into AcrB to enable the production of an efflux inactive variant in EKO-35. Using the pINT2 plasmid, the mutated acrB gene was integrated into the arabinose operon of EKO-35 (EKO-35 araC::acrBD408A), with gene expression under the control of the constitutive PLacI promoter. Inventors confirmed the AcrBD408A protein was inactive and present at comparable levels to the wild-type protein within the membrane of EKO-35 (
Next, while the successful generation of EKO-35 shows the E. coli drug efflux system is dispensable for growth, inventors measured the growth kinetics and assessed the cellular morphology of EKO-35 under commonly used laboratory conditions: nutrient-rich (Lysogeny broth) and nutrient-limited (M9 defined minimal medium with glucose) medium at 37° C. (
Next, to explore whether the nonsynonymous EKO-35 mutations (Table and Table 6) were compensatory, and to ascertain if they impacted any of the observed phenotypes detailed in this disclosure, inventors obtained clones from the E. coli ASKA library harboring the rspA, tufA, pitA, wcaC, gyrB, and yjfC genes on plasmids with gene expression under the control of an IPTG-inducible promoter. Growth profiling revealed pitA overexpression conferred a lethal phenotype, and the overexpression of the remaining genes induced a fitness cost in EKO-35 under nutrient-rich optimal growth conditions, indicating the mutations could be compensatory in nature (
To further characterize EKO-35, and to gain fundamental insight into how E. coli responds to loss-of-efflux, inventors assessed fluctuations in the membrane-enriched cellular proteome of cells sampled at the mid-exponential phase of growth (
Comparative proteomics showed significant changes in the abundance of 111 proteins in response to loss of the E. coli efflux system: 38 proteins were significantly increased in the wild-type proteome, and 73 proteins were significantly increased in the proteome of EKO-35 (
Using a false discovery rate (FDR) of 2%, annotation enrichment 42 of the EKO-35 cellular proteome revealed few differences compared to the wild-type proteome. Increasing the FDR to 5% identified three categories enriched in the wild-type strain, including iron transport. These categories correlated to the reduced abundance of proteins with annotated functions associated with iron-transport in EKO-Drug efflux pumps are well-characterized for their role in siderophore extrusion. Therefore, EKO-35 can downregulate iron acquisition systems to alleviate the fitness cost associated with accumulation of these molecules. However, it is important to note that these experiments were performed in relatively nutrient-rich conditions. In addition, five categories were enriched in EKO-35, including bacterial flagellum, phosphate transport, nitrate assimilation, chemotaxis, and electron transport (
Next, comparative proteomics was applied to assess fluctuations in the proteome of EKO-35 in nutrient-limited conditions. Inventors identified a total of 2,019 proteins from whole cell extracts of the wild-type strain, which represented 46% of the predicted proteome. PCA defined significant separation between EKO-35 and the wild-type strain (Component 1, 53.2%), and biological variation (Component 2, 14%) (
Similar to nutrient-rich conditions, annotation enrichment revealed few differences between the wild-type and EKO-35 proteomes using an FDR of 2%. When the FDR was increased to 5%, the wild-type proteome was enriched with categories including acetylation, tricarboxylic acid cycle, sugar transport, ubiquinone, and quinones (
Finally, two proteins associated with the EKO-35 nonsynonymous mutations were detected in the proteomes of wild-type E. coli and EKO-35; PitA was significantly increased in the wild-type strain, supporting the notion that PitA can negatively impact EKO-35 fitness. Additionally, inventors also detected a higher abundance of RspB in the wild-type proteome, which is encoded in an operon with rspA, further supporting that the nonsynonymous mutation in rspA may mitigate an associated fitness cost. Indeed, EKO-35 can have both reduced protein abundance and mutated the corresponding genes to alleviate the fitness cost associated with loss-of-efflux.
E. coli is a versatile organism withstanding diverse and challenging environments. It is suggested the majority of efflux pumps are not constitutively produced under optimal growth conditions, which can underpin the observed dispensability of the drug efflux system (
To ascertain whether the E. coli efflux system becomes conditionally essential, inventors profiled EKO-35 growth under a range of different conditions. First, inventors confirmed the essentiality of drug efflux pumps for survival under extreme acid and alkaline conditions (
Since efflux pumps have also been associated with biofilm formation, inventors next explored whether loss-of-efflux impacted biofilm formation in E. coli. EKO-35 was propagated statically under both nutrient-rich and nutrient-limited conditions (
Next, inventors measured the growth kinetics of EKO-35 at 25° C., and under reduced oxygen concentrations, in both nutrient-rich and -limited conditions (Table 1). The lag-phase was significantly extended in nutrient-rich medium; however, no significant differences were identified in nutrient-limited medium (
Most notably, inventors observed that the E. coli drug efflux system is essential for growth in a nutrient-limited low-oxygen environment (5% O2) (
In summary, by profiling diverse growth conditions, inventors reveal instances where the fitness of EKO-35 is impacted due to loss-of-efflux. EKO-35 also exhibited phenotypes distinct to the ΔtolC mutant, which provides important biological insight into the physiological functions of drug efflux pumps. Indeed, EKO-35 is an important tool to further characterize the role of drug efflux pumps in physiological processes. The present findings highlight potentially compensatory nonsynonymous mutations that require further investigation.
EKO-35 Displays Differing Susceptibility Levels to a ΔtolC Mutant
Due to the observed phenotypic differences observed between EKO-35 and the ΔtolC mutant, inventors explored whether the strains also exhibited differences in susceptibility to growth inhibitors. Since inventors disrupted all drug efflux pumps known to form complexes with TolC, in addition to single component IM pumps, inventors posited it was likely EKO-35 would exhibit comparable—if not increased susceptibility—to antimicrobial agents.
Inventors curated and profiled a collection of compounds (n=52) with diverse physicochemical properties, including molecular weight (138.059 g/mol to 1449.27 g/mol), lipophilicity (log P −7.8597 to 5.823), aqueous solubility (log S −9.422 to and total polar surface area (PSA) (0 to 530.49 A2) (Table 10). This collection included well-described antibiotics, dyes, detergents, antiseptics, bile acids, and a subset of poorly characterized synthetic compounds (
Overall, EKO-35 and the ΔtolC mutant had similar susceptibility profiles for ˜65% (n=34) of the compounds, which were largely well-characterized antibiotics. However, the ΔtolC mutant was more susceptible to 31% of the profiled compounds, and a majority of these were the synthetic and poorly-characterized compounds (Table 10). These compounds also included novobiocin (4-fold difference) (Table 10), which inventors predicted could be due to the EKO-35 nonsynonymous mutation in gyrB (Table 5 and Table 6). Finally, EKO-35 was more susceptible to two compounds (acriflavine and ethidium bromide), which display relatively lower log P values (−1.9857 and −0.102, respectively) and a narrow molecular weight range (˜390 to 460 g/mol) (
To investigate whether the differences in susceptibility between EKO-35 and ΔtolC were associated with the six missense mutations identified within the EKO-genome (
Finally, inventors also generated an EKO-35 porinated strain (EKO-35-Pore), through introduction of an open variant of the OM siderophore transporter FhuA, which herein will be denoted as a ‘pore’. The FhuA transporter is rendered non-selective through removal of a terminal plug domain and four large external loops, which enables production of a ‘pore’ with an internal diameter of ˜2.4 nm. This non-selective pore increases the influx of both hydrophilic and hydrophobic compounds, without affecting efflux. As described previously, inventors introduced the pore into the intergenic region between the glmS and pstS genes in the genome of EKO-35 and the wild-type strain, with gene expression under the control of a constitutive promoter. Growth of the EKO-35-Pore strain was comparable to EKO-35 under optimal conditions (
Next, inventors profiled the wild-type, EKO-35 and ΔtolC porinated strains against the same curated collection of physicochemically diverse compounds (n=52), to gain insight into the physicochemical properties of compounds retarded by the OM and/or those that are susceptible to efflux (
In summary, the porinated strains provide an additional tool kit to study the transport of compounds across the E. coli cell envelope, enabling dissociation between permeation across the OM and active efflux. Indeed, with an intact OM, the physicochemical properties of compounds that could be assessed using the presently disclosed platform would be limited to those that can permeate the OM; for example, the exclusion limit for porins is considered to be ˜600 g/mol. The pore overcomes the selectivity barrier of the OM, widening the physicochemical properties that could be profiled in this disclosure. In addition, the differing susceptibility levels of EKO-35 and the ΔtolC mutant emphasizes that while both strains can be considered efflux-deficient, inactivation of IM efflux pumps impacts susceptibility differently than inactivation of an OM channel. Therefore, functions distinct from these IM efflux pumps could be attributed to the TolC-associated resistance phenotypes. This observation highlights the utility of EKO-35 for the study of drug efflux across the cell envelope.
To demonstrate the use of EKO-35 as a tool to investigate the physicochemical substrate specificities of efflux pumps, inventors constructed an efflux platform consisting of strains individually expressing efflux pump-encoding genes. Inventors selected genes encoding pumps that form tripartite complexes with TolC (AcrB, AcrD, AcrEF, MdtEF, MdtBC, EmrAB, EmrKY, and MacAB). In addition, inventors included mexCD from P. aeruginosa—which can function with TolC62—to show that EKO-35 can be used as a broad-spectrum tool to study efflux pumps from other bacterial species. Overall, the intent is for expansion of the platform to include additional efflux pumps of interest as needed.
First, inventors attempted to integrate each gene into the genome of EKO-35 using the pINT1 plasmid, which enables markerless integration of genes into araC, with gene expression under the control of the strong and constitutive PBla promoter. Inventors reasoned that integrated genes with constitutive expression would provide increased stability and circumvent the need for selective markers and inducers. However, inventors observed numerous deleterious mutations in a subset of the genes following ligation into this vector, and also following integration into the genome (data not shown). Consequently, inventors posited that the expression level could be too high and inventors instead integrated each gene using the pINT2 plasmid, which utilizes the relatively weaker and constitutive PLacI promoter. Following genomic integration into araC and removal of the resistance cassette, each gene integration was confirmed using Sanger sequencing. The EKO-35 integrated strains will herein be referred to as EKO-35 araC::X, where X represents the gene of interest. As a proof of principle, comparison of AcrB levels between the wild-type, EKO-35, and EKO-35 araC::acrB strains confirmed the complementation system was functional and the protein was produced at higher levels than when the gene is expressed at the basal level (
To highlight the utility of EKO-35 for the investigation of efflux pump substrate profiles, the efflux genes mentioned above were also integrated into the genome of the wild-type E. coli BW25113 strain, and single deletion mutants corresponding to the pumps of interest. These strains, and the EKO-35 integrated strains, were profiled against known substrates for each respective efflux pump (Table 12). Since the wild-type strain harbors an intact drug efflux network, there were no differences in susceptibility when the efflux pump-encoding genes were expressed (
In summary, due to the well-described differences in efflux pump basal expression levels, the developed efflux platform enables the study of efflux pumps in isogenic background(s) free of the masking effects of promiscuous pumps, with gene expression under the control of the same constitutive promoter. As such, this platform allows for the comparison of efflux pump substrate profiles, profiling of efflux pump inhibitors (EPIs), efflux pump interplay, and delineation of the molecular properties governing efflux, as described below.
To investigate the tripartite efflux pump substrate profiles, inventors determined the minimum inhibitory concentrations (MICs) of each compound in the curated collection against each strain within the efflux platform, for both the EKO-35 and EKO-35-Pore strains producing efflux pumps (
As anticipated, AcrB was associated with resistance to a significant subset (˜85%) of the compounds (
In contrast, AcrD and EmrAB decreased the susceptibility of EKO-35 to a smaller subset of compounds (30% and 33%, respectively) (Table 17). AcrD was associated with resistance to compounds with a narrower molecular weight range and PSA (
In contrast to the broad substrate profiles of the RND efflux pumps, the remaining pumps were associated with resistance to a much smaller range of compounds. For example, EmrAB primarily conferred resistance to non-polar compounds spanning a small range of lipophilicity, including the uncoupler CCCP, which is consistent with previous studies. EmrAB no longer provided resistance to CCCP in the porinated strain, and was instead associated with resistance to fusidic acid and synthetic compound 10 (
Finally, inventors also profiled the MexCD pump from P. aeruginosa to demonstrate the use of EKO-35 as a tool for the study of bacterial efflux pumps from other bacterial species. MexCD conferred resistance to ˜55% of compounds profiled, many of which were known substrates, in addition to six of the poorly-characterized compounds (
Overall, porinating the OM of the EKO-35 efflux-integrated strains did not substantially affect the activity of the efflux pumps, including their overall physicochemical substrate profiles. However, inventors did identify additional substrates when the OM was compromised, revealing that efflux pumps have expanded substrate profiles in environments that increase OM permeation. In the few instances (e.g., AcrEF, MacAB, and MexCD) where inventors observed decreased efflux-mediated resistance following introduction of the pore, it is possible that increased permeation overwhelmed the pump. However, the findings of the present disclosure indicate that efflux pumps can function robustly even when the OM is compromised.
Due to their role in antibiotic resistance, efflux pumps are attractive antibacterial targets. Indeed, an inhibitor of a polyspecific efflux pump, such as AcrAB, could simultaneously enhance the activity of numerous antibiotics against resistant strains. Phenylalanine-Arginine 8-Naphthylamide (PAβN) is one of the best-studied efflux pump inhibitors (EPIs), which inhibits AcrAB and its homologues, including numerous P. aeruginosa pumps (e.g., MexAB-OprM, MexCD-OprJ, MexXY-OprM, and MexEF-OprN). In addition, several arylpiperazines have been associated with E. coli efflux inhibition, including 1-(1-naphthylmethyl)-piperazine (NMP). Here, using PAβN and NMP as a proof of concept, inventors show EKO-35 and the developed efflux platform can be used to assess the specificity of EPIs.
First, inventors assessed PAM and NMP synergy with various antibiotics by checkerboard analysis against the efflux-deficient strains, EKO-35, EKO-35 acrBD408A, and ΔtolC, in addition to the wild-type strain (
Acknowledging that distinguishing EPI permeabilizing properties from efflux inhibition is a challenge, inventors next utilized the porinated strains to assess synergy, predicting that synergy would be lost in the porinated strains. Indeed, inventors observed that PaβN was no longer synergistic in combination with erythromycin, novobiocin, and oxacillin against the porinated efflux-deficient strains (
Next, inventors investigated EPI specificity against both EKO-35 and the EKO-35-Pore strain producing different efflux pumps, by assessing synergy with identified antibiotic substrates for each pump (
Overall, NMP is not as potent as PAβN, providing relatively higher fractional inhibitory concentration index values (FICIs) (
In summary, EKO-35 and the efflux platform are important tools for the assessment of candidate EPIs, enabling differentiation between compounds that permeabilize membranes, which increases the influx of antibiotics, and those that solely act as EPIs. In addition, there are major limitations associated with investigating EPI specificity in strains harboring intact drug efflux networks. EKO-35 and the efflux platform overcome these limitations, enabling users to systematically assess inhibition of each efflux pump within the platform.
Previous studies revealed the combination of structurally distinct efflux pumps—single component inner membrane efflux pumps and tripartite systems, which span the cell envelope—can confer multiplicative effects on resistance. Specifically, the combination of these two pump types confers a fold increase—or multiplicative effect—on resistance that is equal to or greater than the product of the fold increases conferred by the individual efflux pumps alone. An additive effect is observed when the fold increase in resistance equates to the sum of the individual efflux pumps alone. Interplay is considered to be the result of single component inner membrane pumps effluxing compounds to the periplasm, where tripartite systems can then access these compounds, extruding to the outside of the cell. Indeed, AcrB was shown to only provide robust resistance to ethidium bromide and acriflavine when single component pumps are present, since AcrB is considered to only access substrates from the outer leaflet of the inner membrane and the periplasm. Therefore, for compounds with cytoplasmic targets, such as acriflavine and ethidium bromide, it is possible that tripartite efflux systems rely on single component inner membrane pumps to first efflux substrates to the periplasm. However, inventors observed that with the exception of ethidium bromide and acriflavine, the overexpression of acrB alone restored the sensitivity of EKO-35 to levels comparable, if not greater, than those observed in the wild-type strain for the majority of compounds profiled in this disclosure (Tables 13A and 13B). Thus, AcrB provides robust efflux independent of single component inner membrane pumps for a wide range of compounds with diverse physicochemical properties. Indeed, findings of this disclosure show that permeation across the cell envelope is sufficiently rate limiting, since the tripartite pumps can access the drugs from the periplasm and the inner-membrane outer leaflet as they are diffusing.
To demonstrate the use of EKO-35 to study efflux pump interplay, the inner membrane efflux pump EmrE was introduced into EKO-35 strains with integrated RND tripartite efflux pumps (AcrB, AcrEF, AcrD, and MdtEF) using the pGDP2 plasmid (pGDP2:emrE), which features the constitutive PLacI promoter. Inventors then assessed resistance to ethidium bromide and acriflavine, since AcrB alone did not restore resistance to wild-type levels. Overall, inventors observed that the combination of EmrE with all three tripartite systems conferred multiplicative effects for both ethidium bromide and acriflavine (
Next, inventors predicted that interplay can be more apparent when the OM is compromised, causing a greater influx of the compounds into the cell. To investigate, the EKO-35-Pore strains were profiled using the same approach. Overall, the multiplicative effects were maintained for the majority of the pump combinations, with the exception of AcrD with acriflavine, where the multiplicative effect was lost, and AcrD did not increase resistance to acriflavine when produced alone or in combination with EmrE (
Bacterial drug efflux networks are expansive and poorly characterized, extending beyond archetypal pumps (e.g., AcrB) that are well-studied due to their polyspecific transporting capabilities. However, the substrate specificity and functions of many efflux pumps remain poorly understood, which can be attributed to the complexities of these systems, including differential gene expression and a high degree of functional redundancy. Here, inventors describe the generation of EKO-35, a simplified genetic background to address the described limitations of the efflux field. This strain can be used to study the functions and physicochemical substrate specificities of individually introduced efflux pumps, to assess the mechanism of action and efficacy of EPIs, and to study efflux pump interplay.
Inventors' ability to inactivate such a significant number of efflux pumps within the E. coli genome has provided important biological insight. Despite the high degree of functional redundancy, at least in terms of antibiotic detoxification, the E. coli drug efflux system is highly conserved. Conservation shows these proteins could contribute to physiologically essential roles, in addition to their well-described ability to extrude clinically important antibiotics. Indeed, numerous physiological functions have been associated with drug efflux pumps. For example, the TolC OM channel is broadly implicated in enterobacterial physiology. Inactivation causes pleiotropic phenotypes, including severe growth defects in nutrient-limited conditions. As described, MdtEF-TolC has been associated with the detoxification of nitrosative derivatives produced during anaerobic respiration. In addition, many E. coli efflux pumps have been associated with the extrusion of enterobactin, MdtJI exports the polyamine spermidine, and SugE is a guanidinium ion efflux pump.
Despite the contribution of these proteins to the maintenance of cellular homeostasis, here inventors show the E. coli drug efflux system is dispensable under optimal growth conditions (
In support of efflux-associated physiological functions, inventors show the E. coli efflux network is conditionally essential. Indeed, the introduction of wild-type copies of the above mentioned genes did not restore the conditionally essential phenotypes described in this disclosure, showing these phenotypes correlate with loss-of-efflux pumps directly. The fitness of EKO-35 was significantly impacted under acid and alkali stress (
To the best of inventors' knowledge, EKO-35 is the most efflux-deficient mutant to be reported. As such, this strain is highly susceptible to numerous antimicrobial agents (Tables 13A and 13B). Similar to the observed phenotypic differences, EKO-35 exhibited differing susceptibility levels to the ΔtolC mutant. Importantly, most of the well-characterized antibiotics inhibited the growth of both strains comparably, indicating tripartite efflux pumps are primarily responsible for extruding these compounds, and are the major contributors to resistance in the efflux network. However, numerous compounds exhibited increased activity against the ΔtolC mutant, despite the lack of tripartite efflux pumps in EKO-35. Most of these were the poorly characterized synthetic compounds (Tables 10A and 10B, compounds 1-19). Inventors have shown these differences were not associated with the nonsynonymous mutations, since the introduction of wild-type copies of these genes did not increase the susceptibility of EKO-35 (
With an intact OM, the physicochemical properties that can be assessed using EKO-35 and the efflux integration platform would be limited to those that can penetrate the OM, which would restrict the molecular weight range to compounds <600 g/mol. To overcome this limitation, a pore was introduced into the wild-type, ΔtolC, EKO-35, and EKO-35 efflux pump-integrated strains, which enabled uncoupling of the contributions of influx and efflux. Inventors show that efflux contributed more significantly than OM permeability to the intrinsic resistance of E. coli, and compromising both the OM and efflux increased susceptibility to a wide range of compounds (
Previously, different approaches have been taken to investigate the E. coli drug efflux system. Sulavik et al., individually inactivated 16 efflux pumps and profiled these strains against a panel of 20 toxic molecules. Due to efflux pump functional redundancies, and differential expression levels, there are limitations associated with this approach. Nishino & Yamaguchi individually expressed 37 efflux pump-encoding genes in an AcrAB-devoid host, and subsequently profiled 26 toxic molecules. Both studies substantiated AcrAB-TolC as the major contributor to intrinsic resistance. A subset of additional efflux pumps were associated with resistance to a proportion of the compounds, but a large number did not provide resistance phenotypes. Additional studies have also investigated E. coli efflux using tolC inactivated mutants, to ascertain molecular features of compounds amenable to efflux. Since TolC is the predominant gate keeper for extrusion across the OM in E. coli, the study of tolC mutants represents loss of efflux as a whole, and provides important information relating to the properties of compounds that are susceptible to efflux by tripartite systems. However, little insight is provided into the physicochemical substrate specificities of the IM pump components. In addition, as described above, tolC inactivation causes pleiotropic effects that appear to be distinct from loss of drug efflux pump functions. Therefore, there are limitations associated with the use of tolC mutants to study drug efflux, which are overcome by EKO-35.
Construction of the developed efflux platform enabled us to assess the physicochemical substrate profiles of E. coli efflux pumps forming tripartite complexes with TolC. Inventors also included MexCD from P. aeruginosa, showing that the platform can be used to study pumps from other organisms. The efflux platform was built upon an isogenic and highly susceptible background (EKO-35), with gene expression under the control of the same constitutive promoter. Such an approach enabled direct comparisons to be made between strains and inventors were able to summarize molecular properties that contribute to efflux in each pump (
Previous studies indicate that molecular weight and hydrophobicity are key factors impacting compound susceptibility to efflux; specifically, compounds exhibiting molecular weights between 300 and 700 g/mol are more susceptible to efflux. Findings from this disclosure were consistent with these observations, yet inventors observed a slightly larger molecular weight range through EKO-35 susceptibility testing, identifying compounds ranging from −227 to 750 g/mol as being susceptible to efflux (
It is possible that efflux pump interplay could impact the substrate profiles of tripartite efflux pumps; as described, interplay refers to synergistic relationships between tripartite systems and single component IM pumps. However, inventors show the production of AcrB alone can restore the susceptibility of EKO-35 to wild-type levels for most compounds profiled. Indeed, acriflavine and ethidium bromide were the only compounds that tripartite pumps appeared to rely on contributions from single component inner membrane pumps to provide robust resistance (
In addition to investigating efflux pump physicochemical substrate profiles, the efflux platform can also be used to evaluate EPIs. As a proof of principle, inventors assessed synergy of the EPIs PAβN and NMP with various antibiotics (Table 12 and
In conclusion, efflux pumps are a major contributor to the intrinsic antibiotic resistome of Gram-negative pathogenic bacteria. Understanding the molecular properties that influence efflux is key for overcoming these resistance mechanisms. It is important that all efflux pumps are considered since inventors have a poor understanding of the fitness advantage conferred by each pump during the bacterial lifecycle. The efflux platform represents an innovative tool kit to fully dissect the movement of compounds across the cell envelope, providing a unique opportunity for users to introduce desired combinations of efflux pumps, and to also probe the relationship and balance between permeation and active efflux. Overall, EKO-35 and the developed platform will be an important resource to the efflux field, which can be used to profile efflux pumps of interest, to assess physiological functions and substrate specificities, and ultimately assist the design of new antibacterial agents and EPIs.
0.781
25
3.125
0.195
0.195
0.098
0.098
0.098
187.5
12.5
6.25
320
320
320
320
320
320
20
320
320
320
320
320
80
160
320
160
40
5
320
320
320
320
320
320
320
320
12.5
12.5
12.5
25
20
6.25
3.125
3.125
12.5
25
12.5
10
12.5
25
25
25
100
1.563
0.781
125
1000
1500
1500
1500
50
320
320
320
320
40
1.25
320
320
160
160
320
320
2.5
1.563
0.781
1.563
1.563
0.039
10
3.125
6.25
3.125
3.125
3.125
3.125
12.5
0.781
0.391
125
125
750
750
1000
1000
20
40
320
320
160
160
1.25
0.625
80
80
320
20
4
4
256
256
256
32
8
64
4
16
8
8
32
16
16
4
32
8
128
32
16
16
4
4
4
4
8
4
128
64
128
16
4
8
4
128
64
64
64
64
64
64
16
8
32
8
32
32
32
32
32
4
4
4
4
4
4
4
16
16
16
16
16
16
16
4096
8
4096
128
8
8
8
8
32
32
16
64
64
64
64
32
4
4
4
4
8
4
8
8
8
16
32
8
8
4
32
4
4
4
16
16
16
16
512
16
8
8
16
16
64
64
8
4
8
4
4
4
8
4
8
4
4
4
4
16
8
4
4
4
4
4
4
4
8
16
16
16
16
16
16
8
16
16
256
16
−
8
−
16
7.20 × 10−64
16
256
16
256
8
128
16
128
9.79 × 10−64
16
16
9.79 × 10−64
16
7.26 × 10−64
4
32
−
4
−
4
2.45 × 10−31
64
256
64
256
16
64
16
64
16
8
8
32
8
32
512
512
256
256
128
64
32
32
128
64
64
16
16
8
8
4
4
4
4
4
4
Bacterial strains and plasmids used in this Example are provided in Table 21. E. coli K-12 str. BW25113, the parental strain of the Keio Collection (Baba, T. et al., 2006) was used as the background for generation of EKO-35v2. E. coli TOP10 or E. coli DH5a strains were used as routine cloning hosts. Plasmids for CRISPR-Cas9 mediated counterselection, pCas and pTargetF, were purchased from Addgene (Jiang, Y. et al., 2015). Strains were routinely grown in Lysogeny broth (LB) (Bioshop) at 37° C. or 30° C. For optimal aeration, broth cultures were grown with aeration at 220 rpm. For growth profiling, microtiter plates were incubated at 37° C. with continuous linear shaking at 600 rpm. Ampicillin (100 μg/mL) (Bioshop), kanamycin (50 μg/mL) (Sigma-Aldrich), spectinomycin (50 μg/mL) (Bioshop), and gentamicin (10 μg/mL) (BioBasic) were used at the listed concentrations for selection of resistance markers.
E. coli K-12 str.
E. coli TOP10
E. coli DH5α
E. coli EKO-35v2
Generation of EKO-35 was achieved using a CRISPR-Cas9 counter-selection (Jiang, Y. et al., 2015). The efflux genes were inactivated in the order denoted in Table 22. All PCR reactions and restriction enzyme digests were prepared according to manufacturers' guidelines. Amplicons were purified using a GeneJET PCR purification kit (Thermo Fisher Scientific) according to manufacturer's guidelines. The 2×GB-AMP™ high-fidelity PaCeR™ polymerase Master Mix (GeneBio Systems Inc) and Taq 2× polymerase Master Mix (FroggaBio) were used according to the manufacturer's suggested guidelines.
For CRISPR-Cas9-mediated counterselection, the methodology described by Jiang et al. was modified for high-throughput screening of mutants (Jiang, Y. et al., 2015). Multiple efflux-encoding genes were targeted simultaneously by multiplexing two or three guide RNAs in the pTargetF vector. CRISPR guide software (Benchling) was employed for selection of appropriate N20 sequences. pTargetF was modified via PCR to introduce an N20 for the gene of interest (Table 23). Amplicon size (2100 bp) was verified via gel electrophoresis, and the remaining PCR product was purified. A second N20 sequence was amplified using PaCeR™ polymerase and inserted into pTargetF-gene1 vector through restriction digest with EcoR1 and XhoI and ligation with T4 ligase. A third N20 sequence was amplified using PaCeR™ polymerase and inserted into pTargetF-gene1-gene2 vector through restriction digest with XbaI and HindIII and ligation with T4 ligase. pTargetF vectors were verified using Sanger Sequencing at the Advanced Analysis Centre (AAC) University of Guelph. To enable rapid screening of positive mutants and to disrupt the target gene, ssDNA repair oligos (˜100 bp in length) were designed to contain an AseI restriction site and three tandem stop codons (Table 23). All ssDNA repair oligos were purchased through Integrated DNA Technologies (IDT). Electrocompetent cells of the mutant strain of interest were transformed with 50 ng of pCas. A broth culture of each strain was grown to the mid exponential phase (OD600 nm˜0.5) in the presence of kanamycin and 10 mM arabinose to induce recombinase expression. To recombinase induced electrocompetent cells, 100 ng of pTargetF that was modified to contain the desired N20 sequence, and 2000 ng of repair ssDNA targeting the gene of interest were electroporated (Bio-Rad MicroPulser, Ec1 setting, 1 mm electroporation cuvette (Fisher)). The cultures were recovered in LB with 1 mM arabinose at 30° C. and propagated on selective agar (LB with kanamycin and spectinomycin) to identify successful gene disruptions. For high-throughput screening of colonies, Taq polymerase was used with primers annealing to the target region of each gene (Table 23). The amplicons were digested with AseI and successfully inactivated genes were identified via gel electrophoresis by digestion relative to a wild-type negative control. Insertion of the three tandem stop codons into the gene of interest was verified using Sanger sequencing at the Advanced Analysis Centre (AAC) (University of Guelph). Genes disrupted using CRISPR-Cas9-mediated counter selection are indicated in Table 22.
ATT GGA AAT CGG GCC ATA AAA CAG CTG TGA GAC
TTA ATT GGC CCA GCC CAT TTG ATC AAC GAA GCG
TTA ATT ATC TGG CCC CAG CTC GGC GCT GAC TCC
Genomic DNA was extracted using the Purelink Genomic DNA Mini Kit (Invitrogen), according to the manufacturer's guidelines. Quality of the extracted gDNA was assessed using gel electrophoresis. Illumina DNA library preparation was performed using an Illumina Nextera kit by the Microbial Genome Sequencing Center (Pennsylvania, USA), which was followed by Illumina sequencing on a NextSeq 2000 platform. Analysis of the raw reads was performed using Geneious Prime 2021.0.2 (Kearse, M. et al., 2012). Low quality reads were trimmed using an in-suite BBDuk plug-in. Raw wild-type reads were assembled to an NCBI reference genome (Accession No. CP009273.1) with bowtie2. The resulting assembly was used as a reference to assemble the EKO-35v2 mutant reads. Differences between the wild-type BW25113 and EKO-35v2 strains were identified by searching for single nucleotide polymorphisms (SNPs) and deletions using the following thresholds: minimum variant frequency of 0.75, maximum variant P-value of 10−6, and minimum variant P-value of 10−5. The results were confirmed using the breseq (v 0.35.6) pipeline. Three intergenic mutations and three secondary mutations were identified, as summarize in Table 24. The nonsynonymous mutation in yhaM was repaired using CRISPR-Cas9-mediated counterselection, which introduced three intentional silent mutations (yhaM A390T, C393A, C432A) to remove the adjacent PAM site and introduce XhoI-guided screening.
For growth profiling in nutrient-rich conditions, strains were propagated on LB agar for 18 h at 37° C. Single colonies were inoculated into LB and grown at 37° C. until the mid-exponential phase (OD600 nm˜0.6) was reached. All strains were assessed with at least three biological replicates. The cultures were standardized to an OD600 nm˜0.1 in sterile 0.85% saline (w/v). Standardized cultures were diluted 1/200 into LB and 100 μL of the resulting dilution were applied to round-bottom 96-well microtiter plates (VWR). To prevent evaporation, the microtiter plates were sealed (labeling tape, Fisher Scientific). The OD600 nm was measured every 15 minutes over the course of 24 h using a BioTek Synergy H1 microplate reader. Growth was assessed at both 37° C. and 25° C.
Generation of EKO-35: Inactivation of E. coli Drug Efflux Pumps
Inventors' first goal was to generate a simplified genetic background to overcome the challenges associated with the complexities of intact drug efflux networks. To generate the first-generation strain, inventors started with an ΔacrB mutant from the Keio Collection and removed a further 13 pumps using the λ-Red system and 22 additional pumps using CRISPR Cas9-mediated counterselection. As such, EKO-35 of Example 1 (i.e. EKO-35v1) contains 13 flippase recognition target (FRT) sites and 11 secondary mutations. To circumvent genomic mobility due to FRT site-mediated DNA translocation, the inventors created the second-generation strain, EKO-35v2, using only CRISPR Cas9-mediated counterselection. EKO-35v2 is a scarless genetic background with significantly fewer secondary mutations, representing a stable isogenic background devoid of 35 efflux-encoding genes (Table 24).
The efflux genes were inactivated in the following order: ΔyajR; mdtO; ydhC; emrE; yojI; mdtD, sugE; ynfM, emrD, ydeF; mdlA, emrY; mdtK, bcr; mdtG; mdtH, mdlB, macB, yddA; fsr; ydiM; yieO; mdfA; mdtM; mdtJ; emrB, mdtB, mdtL; yebQ; cusA; mdtF; ydeA; acrF; acrB. While generating the strain, inventors identified a missense mutation (K139T) within the gene encoding a putative L-cysteine desulfidase (YhaM). In effort to create a scarless genetic background devoid of secondary mutations, the inventors repaired this mutation. However, an additional missense mutation (N39H) arose within a gene encoding a xylose ABC transporter (XlyG).
Since many of the efflux pump-encoding genes have predicted start codons and are poorly characterized, it is possible that alternative start codons located downstream from the inserted tandem stop codons (Table 26A) could be utilized for a subset of the CRISPR-inactivated pumps. To investigate, inventors profiled the genome of the efflux-deficient strain, which included the inserted stop codons, using the Prodigal prokaryotic gene recognition and translation initiation site identifier (Hyatt et al, 2010). Overall, the program did not predict production of any potentially functional efflux pumps in EKO-35v1 and EKO-35v2, which supports the notion that the tandem stop codons were sufficient to prematurely terminate translation, and that alternate start codons would not be utilized (Tables 26A and 26B). Indeed, Prodigal analysis of the wild-type strain's genome predicted production of all pumps (Table 26B). In addition, inventors also took advantage of new developments in protein structure predictions, and carefully analyzed AlphaFold-generated models of each efflux pump to ascertain whether the predicted start codons were correct based on the structural features of the proteins. Such an analysis indicated the inserted stop codons were sufficient to inactivate the different genes. This strain was subsequently designated Efflux KnockOut-35 version 2 (EKO-35v2), and was used for phenotypic characterization and construction of an efflux platform, as described below. The EKO-35v2 genome sequence (SEQ ID NO: 255) confirmed successful disruption of the 35 efflux-encoding genes, including successful repair of yhaM, and also revealed two additional secondary mutations, one of which encoded missense mutations and one silent substitutions (Table 24).
With the intent of using EKO-35 as a simplified genetic background to study the functions of drug efflux pumps, inventors' second goal was to fully characterize EKO-35v1 and v2 prior to further investigations. Other efflux-deficient backgrounds (e.g., tolC mutants) display pleiotropic phenotypes, including severe growth defects in minimal M9 medium with glucose supplied as the carbon source. Indeed, tolC inactivation causes periplasmic accumulation of enterobactin in this low-iron growth medium, which underlies this conditionally essential phenotype. In addition, tolC mutants exhibit membrane stress and depletion of essential metabolites, resulting in ‘metabolic shutdown’ in minimal glucose medium. Phenotypic analysis of EKO-35v1 growth kinetics under standard laboratory conditions, nutrient-rich Lysogeny broth, revealed EKO-35v1 exhibited a 1 h extended lag phase. To assess the impact of the secondary mutations and genomic mobility on the growth kinetics of EKO-35v1, the growth kinetics of EKO-35v2 were also profiled. Compared to the wild-type strain, EKO-35v2 shows no significant difference in the lag phase of growth or doubling time (
While the present disclosure has been described with reference to what are presently considered to be the preferred example, it is to be understood that the disclosure is not limited to the disclosed example. To the contrary, the disclosure is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.
All publications, patents and patent applications are herein incorporated by reference in their entirety to the same extent as if each individual publication, patent or patent application was specifically and individually indicated to be incorporated by reference in its entirety.
The present disclosure claims priority from U.S. provisional application No. 63/352,569 filed on Jun. 15, 2022, which is hereby incorporated by reference in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 63352569 | Jun 2022 | US |
