Characterizing Multidrug Efflux Systems in Acinetobacter baumannii Using an Efflux-Deficient Bacterial Strain and a Single-Copy Gene Expression System

Summary

We describe a facile procedure for the single-copy chromosomal complementation of an efflux pump gene using a mini-Tn7-based expression system into an engineered efflux-deficient strain of Acinetobacter baumannii. This precise genetic tool allows for controlled gene expression, which is key for the characterization of efflux pumps in multidrug resistant pathogens.

Abstract

Acinetobacter baumannii is recognized as a challenging Gram-negative pathogen due to its widespread resistance to antibiotics. It is crucial to comprehend the mechanisms behind this resistance to design new and effective therapeutic options. Unfortunately, our ability to investigate these mechanisms in A. baumannii is hindered by the paucity of suitable genetic manipulation tools. Here, we describe methods for utilizing a chromosomal mini-Tn7-based system to achieve single-copy gene expression in an A. baumannii strain that lacks functional RND-type efflux mechanisms. Single-copy insertion and inducible efflux pump expression are quite advantageous, as the presence of RND efflux operons on high-copy number plasmids is often poorly tolerated by bacterial cells. Moreover, incorporating recombinant mini-Tn7 expression vectors into the chromosome of a surrogate A. baumannii host with increased efflux sensitivity helps circumvent interference from other efflux pumps. This system is valuable not only for investigating uncharacterized bacterial efflux pumps but also for assessing the effectiveness of potential inhibitors targeting these pumps.

Introduction

Acinetobacter baumannii is a World Health Organization top priority pathogen due to its encompassing resistance to all classes of antibiotics¹. It is an opportunistic pathogen mostly affecting hospitalized, injured, or immunocompromised people. A. baumannii largely evades antibiotics via efflux pumps, the most relevant being the Resistance-Nodulation-Division (RND) family of exporters². Understanding how these efflux pumps work mechanistically will allow one to develop targeted therapeutic options.

One common way that cellular processes can be specifically distinguished is through genetic manipulation. However, the tools available for A. baumannii genetic studies are limited, and to further confound experimental design, clinical isolates often are resistant to the antibiotics routinely used for selection in genetic manipulations³. A second hurdle encountered when studying efflux pumps specifically is that they are strictly regulated-often by unknown factors-making it difficult to accurately isolate and attribute function to a single pump⁴. Seeing this need to expand the research toolbox, we developed a mini-Tn7-based, single-copy-insertion, inducible expression system that incorporates a Flp recombinase target (FRT) cassette, which allows for the removal of the selection marker^5,6,7 (Figure 1). First created for Pseudomonas^8,9,10, this elegant cloning and expression system was used to generate single-copy efflux pump complements into an RND efflux pump-deficient strain of A. baumannii (ATCC 17978::ΔadeIJK,ΔadeFGH,ΔadeAB: hereafter referred to as A. baumannii AB258) that we generated¹¹. Being able to study one efflux pump at a time and not overwhelm the bacterial cells with high-copy expression (as generally seen with plasmid-based expression systems), one can better learn about the critical, physiological aspects of each efflux pump with minimal interference and reduced complications.

This article describes how to use the mini-Tn7 system to complement a deleted gene of interest, RND efflux pump adeIJK, into the chromosome of A. baumannii AB258 through a series of uncomplicated steps performed over the course of 9 days⁷. The first set of steps re-introduces the deleted efflux pump genes cloned into the mini-Tn7-based insertion plasmid (Figure 2A) at the single attTn7 insertion site downstream of the well-conserved glmS gene (Figure 3A). This process is facilitated by a non-replicative helper plasmid (Figure 2B) that encodes for the transposase genes needed for Tn7-driven insertion. The second set of steps uses an excision plasmid (Figure 2C) for Flp recombinase-mediated removal of the gentamicin gene flanked by FRT sites (Figure 3B) to create an unmarked strain. Though this system is used to elucidate the essential roles and possible inhibitors of RND efflux pumps with respect to antibiotic resistance, it can be used to investigate any gene of interest.

Protocol

1. Experimental preparation

1. Purify the plasmid pUC18T-mini-Tn7T-LAC-Gm⁹ (insertion plasmid, Figure 2A) with the gene of interest.
   NOTE: Here, the gene of interest is adeIJK. The final plasmid concentration should be ≥100 ng/µL.
2. Purify the helper plasmid (pTNS2)⁹ and the excision plasmid (pFLP2^ab)⁶ (Figure 2B,C, respectively), ideally to a final plasmid DNA concentration of ≥100 ng/µL.
3. Prepare 50 mL of sterile ultrapure water and 25 mL of sterile LB (Lennox) broth (see Table of Materials).
4. Prepare a minimum of 10 LB agar plates, each with the following additives: plain (no additives), gentamicin (Gm) at 50 µg/mL, carbenicillin (Cb) at 200 µg/mL (for selection via the ampicillin resistance gene), and 5% sucrose (see Table of Materials).
5. Streak out the strain to be used for insertion on LB agar and incubate at 37 °C for 16-18 h. Here, A. baumannii AB258 is used.
   NOTE: This protocol must be performed in a sterile environment as much as possible by using a Bunsen burner at the bench or a biological safety cabinet. All consumables (pipette tips, inoculating loops, microfuge tubes, etc.) need to be sterile.

2. Culture preparation

1. Inoculate a single colony of A. baumannii AB258 into 4 mL of LB broth in a sterile 13 mL culture tube using a sterile inoculation loop or sterile wooden inoculating stick.
   NOTE: A single 4 mL culture is used for one sample and one control. Increase the number of cultures depending on the number of samples needed.
2. Incubate overnight at 37 °C with shaking (250 rpm).
3. Place a bottle of sterile distilled water (25-50 mL) at 4 °C overnight for use in step 3.

3. Preparation of electrocompetent cells

1. Place all sterile 1.5 mL microfuge tubes (two per culture) and sterile electroporation cuvettes (two per culture) on ice (see Table of Materials). Keep the samples on ice as much as possible throughout the procedure.
2. Place the bottle of sterile water that was stored at 4 °C on ice.
3. Transfer 1.5 mL of the overnight bacterial culture into one of the 1.5 mL microfuge tubes.
4. Centrifuge at 13,000 x g for 2 min to pellet the cells.
   NOTE: Centrifugation would ideally be performed at 4 °C, but ambient temperature centrifugation is acceptable and not detrimental.
5. Using a 1 mL pipette, remove all of the supernatant without disturbing the cell pellet.
6. Add another 1.5 mL of bacterial culture into the same microfuge tube. Centrifuge at 13,000 x g for 2 min, then remove all of the supernatant.
7. Repeat step 3.6 one final time with the remaining 1 mL of culture.
8. Add 1 mL of ice-cold sterile water to the cell pellet and resuspend with gentle pipetting until the pellet no longer sits at the bottom of the microfuge tube.
9. Centrifuge the resuspended cells at 13,000 x g for 2 min.
10. Carefully remove the supernatant using a 1 mL pipette. Do not pour off the supernatant, especially in the subsequent wash steps when the cells tend to form less compact pellets.
11. Repeat this wash step with ice-cold sterile water, steps 3.8 to step 3.10, twice more.
12. Gently resuspend the final cell pellet in 200 µL of ice-cold sterile water.
13. Transfer 100 µL of the final cell suspension into the second ice-cold 1.5 mL microfuge tube. This second aliquot will become the negative control for electroporation. Keep the cell samples on ice.

4. Electroporation

1. Pre-warm 1 mL of LB broth and one LB + Gm⁵⁰ agar plate (prepared in step 1.4) for each sample and control in a static incubator set to 37 °C.
2. In a combined volume of 5 µL or less, add 100-200 ng each of the pTNS2 helper plasmid and the pUC18T-mini-Tn7T-LAC-Gm-adeIJK insertion plasmid to an aliquot of electrocompetent cells.
3. Mix with gentle fingertip tapping to ensure complete mixing of the plasmids with the electrocompetent cells without introducing bubbles.
4. Add an equivalent volume of sterile distilled water into the negative control cell aliquot and mix gently as above.
5. Incubate the samples on ice for 20 min.
6. Transfer the entire cell sample into an ice-cold electroporation cuvette, then place the cuvette back on ice. Repeat for the negative control cell sample.
7. Electroporate the cell sample.
   1. Turn on the electroporator and set it to 2.0 kV (25 µF, 200 Ω) (see Table of Materials).
   2. Wipe the surface of the cuvette with a soft tissue to remove any adhering ice or moisture.
   3. Insert the cuvette into the electroporator and deliver the electric shock.
   4. Immediately add 0.9 mL of pre-warmed LB broth to the cells in the cuvette and gently pipette up and down to mix the cells with the media.
   5. Transfer the entire cell suspension into a new 1.5 mL microfuge tube (room temperature).
   6. Check the time constant value on the electroporator; for best results, this value should be between 4 and 6.
8. Repeat the electroporation procedure (steps 4.7.1 to step 4.7.6) for the negative control cell sample.
9. Incubate the electroporated samples at 37 °C for 1 h at 250 rpm to allow for cell recovery.
10. Using an inoculation spreader, spread 100 µL of each electroporated cell sample onto a pre-warmed LB + Gm⁵⁰ agar plate.
    1. Centrifuge the remaining samples at 13,000 x g for 2 min to pellet the cells.
    2. After removing all of the supernatant using a pipette, resuspend the cells in 100 µL of LB broth.
    3. Spread each entire sample onto individual pre-warmed LB + Gm⁵⁰ agar plates.
       NOTE: Electroporation efficiency can be strain-dependent. When electroporating a strain for the first time, it can be informative to plate out different volumes, or even dilutions, of the electroporation sample to ensure the resulting plates have isolated colonies.
11. Incubate the plates at 37 °C for 16-18 h.

5. Selecting transformed colonies for PCR-based screening

1. Check the electroporation plates. The negative control should have no colonies; the sample should have distinct colonies.
   NOTE: Plates with colonies, at this step and at all subsequent steps where agar plates with colonies or patches are produced, can be stored at 4 °C for up to 3 days before proceeding.
2. Using sterile toothpicks, pick up to 10 single colonies from the LB + Gm⁵⁰ agar plates and patch them onto a fresh LB + Gm⁵⁰ agar plate.
3. Incubate the plates at 37 °C for 16-18 h.

6. Verifying chromosomal insertion by colony PCR

1. Remove the LB + Gm⁵⁰ agar plates containing the patched colonies from the incubator.
2. Using a sterile toothpick or sterile pipette tip, remove a small portion (about the size of a large colony) from a patch into 20 µL of sterile distilled water in a 0.2 mL PCR tube; mix well. The water sample should become visibly cloudy as the cells are released from the toothpick. Prepare any number of samples that need to be screened, but 6 should be sufficient.
3. Incubate the PCR tube containing the bacterial suspension at 100 °C for 5-10 min.
4. Using a mini-centrifuge (see Table of Materials), spin the sample at the fixed maximum speed for 2 min to pellet cellular debris.
5. Transfer the supernatant to a new 0.2 mL PCR tube and place it on ice. This sample contains the template DNA for the PCR reaction.
   NOTE: This template DNA can be stored at -20 °C before proceeding with PCR.
6. Prepare a PCR reaction mixture containing 1x polymerase buffer, 200 µM dNTPs, 0.18 µM ABglmS2_F_New forward primer, 0.18 µM Tn7R reverse primer (Table 1), 1 U of Taq DNA polymerase, and 1 µL of the prepared template DNA in a total volume of 25 µL. Prepare a no-template control (NTC), including all but the template DNA (see Table of Materials).
7. Enter the reaction conditions into a thermal cycler: 95 °C for 2 min; 95 °C for 30 s, 49 °C for 30 s, and 72 °C for 30 s for a total of 35 cycles; 72 °C for 10 min; 12 °C hold. Run the samples.
8. Following the addition of DNA loading dye to a final concentration of 1x, load 10 µL of each PCR reaction on a 2% agarose gel and run at 80 V for 40 min. The expected amplicon size for this primer pair is 382 bp. The NTC should have no bands.
9. Identify the samples that produced the expected PCR product. Prepare a streak plate on LB + Gm⁵⁰ agar plates from any of the PCR-positive patches; incubate at 37 °C for 16-18 h. The goal is to generate a plate with single colonies for the preparation of a glycerol stock to preserve this marked strain and as a starting point for creating an unmarked strain (described below).

7. Removal of the Gm^R marker using pFLP2^ab

1. Follow the procedure mentioned in step 2: Culture preparation. The sample used to prepare the overnight culture is a single colony from the LB + Gm⁵⁰ agar plates prepared to generate discrete colonies in step 6.9.
2. Follow the procedure mentioned in step 3: Preparation of electrocompetent cells. Prepare the cells from the overnight culture for electroporation.
3. Follow the procedure mentioned in step 4: Electroporation. Here, the plasmid to introduce to the cells is pFLP2^ab (100-200 ng), which will remove the gentamicin resistance gene, flanking the chromosomally inserted gene of interest.
4. After the cells have recovered for 1 h (step 4.9), spread 100 µL of the electroporated cell sample onto a pre-warmed LB + Cb²⁰⁰ agar plate (prepared in step 1.4). This selective pressure ensures that only cells harboring the pFLP2^ab excision plasmid will grow.
5. Incubate the plates at 37 °C for 16-18 h.
6. Check the plate for colonies. The negative control plate should have no colonies; the LB + Cb²⁰⁰ agar sample plate should have distinct colonies.
7. Using sterile toothpicks, cross-patch up to 20 isolated colonies onto an LB + Cb²⁰⁰ agar plate and an LB + Gm⁵⁰ agar plate.
8. Incubate the plates for 16-18 h at 37 °C.
9. Check the plates for patches. Clones that grow on the LB + Cb²⁰⁰ agar plate but not the LB + Gm⁵⁰ agar plate have had the gentamicin resistance gene removed from the original insert.
10. Select the patches that are carbenicillin resistant and gentamicin susceptible for streaking onto individual LB agar plates supplemented with 5% (w/v) sucrose, which forces the expulsion of the pFLP2^ab plasmid from the bacteria. Choose up to 10 colonies.
11. Incubate the plates for 16-18 h at 37 °C.
12. Check the plates for colonies.
13. As a final confirmation of pFLP2^ab plasmid loss, cross-patch 4-6 isolated colonies from the 5% sucrose plates onto an LB + Cb²⁰⁰ agar plate and an LB agar plate.
14. Incubate the plates for 16-18 h at 37 °C.
15. Choose a clone that is growing on the LB agar plate, and that is carbenicillin sensitive for the preparation of a glycerol stock to preserve this unmarked strain.
    1. Genetic verification of the unmarked strain can be performed with colony PCR (step 6: Verifying chromosomal insertion by colony PCR) using the primers that target the gentamicin resistance gene.
    2. Prepare a PCR reaction mixture containing 1x polymerase buffer, 200 µM dNTPs, 0.18 µM Gm_F forward primer, 0.18 µM Gm_R reverse primer (Table 1), 1 U of Taq DNA polymerase, and 1 µL of the prepared template DNA in a total volume of 25 µL. Prepare a no-template control (NTC) including all but the template DNA, and a positive control using DNA from the marked strain.
    3. Enter the reaction conditions into a thermal cycler: 95 °C for 2 min; 95 °C for 30 s, 50 °C for 30 s, and 72 °C for 40 s for a total of 35 cycles; 72 °C for 10 min; 12 °C hold. Run the samples.
    4. Following the addition of DNA loading dye to a final concentration of 1x, load 10 µL of each PCR reaction on a 1% agarose gel and run at 80 V for 40 min. The expected amplicon size for this primer pair is 525 bp. The positive control should have a single band; the samples and the NTC should have no bands.

Results

The chromosomal insertion procedure takes only 2 h total across 3 days to see a result-colonies growing on a selective agar plate (Figure 1A-C). The expected number of colonies on the transformation plate is strain dependent: one may see 20-30 or even hundreds of colonies as insertion of Tn7 at attTn7 sites is specific and efficient⁹. Patching transformation plate colonies onto selective media (Figure 4A) preserves the transformed strain and provides starting material for colony PCR screening (Figure 1E and Figure 4B). Screening of colonies by PCR can be kept to a minimum-no more than 10 colonies should need to be processed, and most should yield a positive result for insertion. The PCR product for the screening primers, ABglmS2_F_New and Tn7R (Table 1), is 382 bp (Figure 4B lane 4); negative controls for the reaction include wild-type A. baumannii ATCC 17978 (Figure 4B lane 2), AB258 (Figure 4B lane 3), and no template (Figure 4B lane 5). Colonies that are PCR-positive represent complemented, marked strains.

Removal of the gentamicin resistance gene (unmarking) takes less than 3 h, spanning 6 days as cells transformed with the excision plasmid need to be chosen through selective plating, and then the excision plasmid needs to be cured from the bacteria (Figure 1F). Flp-FRT recombination-based excision is precise and effective and should result in ≥20 colonies on the transformation plate. Colonies that are cross-patched onto carbenicillin (selecting for β-lactam resistance conferred by the excision plasmid) and gentamicin (looking for loss of gentamicin resistance) should all be carbenicillin-resistant and gentamicin-sensitive, respectively. The excision plasmid is forced out of the bacteria by growth on 5% sucrose agar plates. Growth on sucrose forces the cells to eliminate pFLP2^ab as the sacB gene on the plasmid promotes the conversion of sucrose to levans, a polysaccharide toxic to the bacteria^12,13. All colonies that grow on 5% sucrose media should then grow only on plain LB agar plates; there should be no growth on carbenicillin agar plates. Colonies growing on the plain LB agar plates represent unmarked strains. Confirming loss of the gentamicin marker can be achieved by colony PCR using the Gm_F and Gm_R primers (Table 1 and Figure 5). This primer pair yields an amplicon of 525 bp only in the positive control (the initially created marked strain, Figure 5 lane 4); wild-type ATCC 17978 (Figure 5 lane 2), AB258 (Figure 5 lane 3), any tested colony (Figure 5 lane 5), and the no-template control (Figure 5 lane 6) should not show amplification.

Once the unmarked strain is confirmed, functional testing can commence with phenotypic assays. Here, the obvious first choice is determining the minimum inhibitory concentration (MIC) of a range of antibiotics: ciprofloxacin is a known substrate of AdeIJK, tetracycline can be removed by AdeIJK (the major efflux pump is AdeAB), and kanamycin has a relatively minor effect on A. baumannii ATCC 17978^2,14. Using the broth micro-dilution method according to the CLSI guidelines¹⁵, the complemented unmarked strain AB258::adeIJK was challenged with each antibiotic in the absence and presence of 50 µM IPTG; wild-type strain ATCC 17978 and RND efflux-deficient strain AB258 were included as controls (Table 2). Overall, the trend seen in the MIC values tells the expected story-decreased susceptibility of AB258::adeIJK to ciprofloxacin and tetracycline with induced expression of the efflux pump, verifying that the insertion of adeIJK was successful.

Figure 1: Overview of the procedure. (A) An overnight culture of the A. baumannii strain to be complemented is prepared. (B) The cells from the overnight culture are washed with water 3 times via centrifugation and kept on ice. (C) The delivery and helper plasmids are added to the cells and incubated on ice for 20 min. The sample is electroporated, LB media is added, and the cells are allowed to recover for 1 h at 37 °C. A 100 µL aliquot of cells is spread onto LB + Gm⁵⁰ agar plates and incubated at 37 °C overnight. (D) Colonies from the transformation plate are patched onto an LB + Gm⁵⁰ agar plate and grown overnight at 37 °C. (E) Patched colonies are prepared for PCR to screen for the presence of an amplification product spanning the chromosomal insertion site. PCR amplification is visualized by agarose gel electrophoresis. PCR-positive samples represent successful insertion of the gene of interest into the chromosome and the creation of a marked strain. (F) A colony positive for gentamicin is prepared as in steps (A-C), with electroporation of the pLFP2^ab plasmid to remove the gentamicin cassette from the chromosomal insertion. Selective plating on LB + Cb²⁰⁰ agar confirms uptake of the plasmid. Duplicate patching on LB + Cb²⁰⁰ and LB + Gm⁵⁰ agar plates reveals colonies that are Cb^R and Gm^S confirming loss of the gentamicin cassette from the insertion. Growth of selected Cb^R colonies on 5% sucrose cures the pLFP2^ab plasmid from the cells. Colonies from the 5% sucrose plate are patched onto LB + Cb²⁰⁰ agar and LB agar to reveal desired Cb^S and Gm^S colonies and confirm the creation of the unmarked strain. Gm⁵⁰, gentamicin at 50 µg/mL; Cb²⁰⁰, carbenicillin at 200 µg/mL; ^R, resistant; ^S, sensitive. Please click here to view a larger version of this figure.

Figure 2: Plasmids used in this protocol. General plasmid maps of (A) pUC18T-mini-Tn7T-LAC-Gm (insertion plasmid), (B) pTNS2 (helper plasmid), and (C) pFLP2^ab (excision plasmid). Please click here to view a larger version of this figure.

Figure 3: Schematic of insertion and unmarking. (A) Insertion. The single Tn7 insertion site in the A. baumannii chromosome is located 24 bp from the end of the glmS2 gene. Co-electroporation of the insertion plasmid and the helper plasmid allows for complementation of the gene of interest (inserted gene, purple) along with the rest of the insertion cassette (FRT sites for marker excision, yellow; accC1 gene for gentamicin resistance, green; lacIq gene for inducible expression, blue) into the chromosome. (B) Unmarking. Electroporation of the complemented, marked insertion strain with the pFLP2^ab excision plasmid facilitates removal of the gentamicin resistance gene (accC1, green) via Flp-FRT recombination (FRT sites, yellow), creating an unmarked strain. Please click here to view a larger version of this figure.

Figure 4: Transformation, patching, and insertion confirmation by colony PCR. Representative result of (A) the growth of transformation colonies after patching, and (B) colony PCR amplification with ABglmS_F_New (grey) and Tn7R (orange) primers to confirm chromosomal insertion. Lane 1: Low molecular weight DNA ladder; Lane 2: ATCC 179798; Lane 3: AB258; Lane 4: AB258::adeIJK-LAC-Gm; Lane 5: no-template control. The expected band of 382 bp is labeled. Note that the gentamicin-specific primers (Gm_F and Gm_R, green) could also be used to affirm chromosomal insertion. Please click here to view a larger version of this figure.

Figure 5: Confirmation of loss of marker by colony PCR. Representative result of colony PCR amplification with gentamicin-specific primers (Gm_F and Gm_R) to confirm the loss of the antibiotic marker via pFLP^ab-based excision. Lane 1: low molecular weight DNA ladder; Lane 2: ATCC 179798; Lane 3: AB258; Lane 4: AB258::adeIJK-LAC-Gm; Lane 5: AB258::adeIJK; Lane 6: no-template control. The expected band of 525 bp is labeled. Please click here to view a larger version of this figure.

Strains, plasmids, and primers	Relevant characteristics	Reference
Stain
A. baumannii ATCC 17978	Type strain	ATCC
A. baumannii ATCC 17978 AB258	ΔadeAB,ΔadeFGH,ΔadeIJK	11
Plasmids
pUC18T-miniTn7T-Gm-LAC	GmR, AmpR	9
pUC18T-miniTn7T-Gm-LAC-adeIJK	GmR, AmpR, adeIJK	This study
pTNS2	AmpR	9
pFLP2ab	pWH1266 origin or replication, sacB, AmpR	7
Primers	Sequence (5′–3′)
ABglmS_F_New	CACAGCATAACTGGACTGATTTC	7
Tn7R	TATGGAAGAAGTTCAGGCTC	7
Gm_F	TGGAGCAGCAACGATGTTAC	This study
Gm_R	TGTTAGGTGGCGGTACTTGG	This study

Table 1: Bacterial strains, plasmids, and primers used in this protocol. Gm, gentamicin; Amp, ampicillin; ^R, resistant.

	Ciprofloxacin		Tetracycline		Kanamycin
IPTG	−	+	−	+	−	+
ATCC 17978	0.250	nd	0.500	nd	1.5	nd
AB258	0.031	nd	0.063	nd	4	nd
AB258::adeIJK	0.016	0.063	0.031	0.125	8	2
Fold change		4.01		4.03		0.25

Table 2: Testing the functionality of the inserted genes via antibiotic susceptibility. Comparison of minimum inhibitory concentration (MIC) values for A. baumannii ATCC 17978, AB258, uninduced AB258::adeIJK, and IPTG-induced AB258::adeIJK against ciprofloxacin, tetracycline, and kanamycin. Fold change = induced (+ IPTG)/uninduced (− IPTG); nd = not determined.

Discussion

Even though this procedure for the chromosomal insertion of an inducible single-copy gene expression system in A. baumannii is technically straightforward and not labor-intensive, there are a few important steps that need to be emphasized. First, preparation of the competent cells needs to be done on ice as much as possible as the cells become fragile during the replacement of the media with ice-cold water. Ideally, the centrifugation steps are performed at 4 °C, but centrifugation at room temperature is acceptable. Given the increasing fragility of the cells during the water washes, gentle pipetting is also critical. Second, electroporation is sensitive to the presence of ions. Washing the cells with multiple rounds of pelleting and resuspending in water ensures the media is fully removed. Also, plasmids should be freshly purified and may be eluted in standard kit elution buffers (normally TE buffer) as long as the plasmid DNA concentration is high enough. We aim to add <5 µL of plasmid to 100 µL of cell suspension to keep the ionic strength of the sample very low, although up to 10 µL should be tolerated. Third, selective agar plates should be prepared as needed to ensure the efficacy of the added antibiotic. Note that carbenicillin was used instead of the usual ampicillin for selection during the transformation of the excision plasmid, pFLP2^ab. A. baumannii is intrinsically resistant to aminopenicillins (ampicillin)¹⁶; substituting a carboxypenicillin (carbenicillin) allows for continued selection with the plasmid-encoded β-lactamase.

Optimization of the experimental protocol is more nuanced and will vary between different species of Acinetobacter (or even genetically manipulated strains within the same species) and possibly the particular reagents used in the lab. For example, the voltage used for electroporation can vary between 1.8 and 2.5 kV, and thermocycling conditions may need to be altered slightly depending on the DNA polymerase used for PCR. Helpful hints to consider if cells are growing poorly after electroporation include reducing the concentration of gentamicin in the agar plates from 50 µg/mL to 30 µg/mL and/or extending the incubation time of the agar plates to ≥24 h. Regarding the steps to remove the gentamicin cassette, better success may be had using LB agar plates with 10% sucrose and/or incubating them at 30 °C for ≥24 h if carbenicillin resistance persists.

Numerous A. baumannii cell preparation methods for use with electroporation can be found in the literature, but they often include a subculturing step after the initial overnight culture and then a long growth phase to a prescribed optical density. We have found that a simple overnight culture can be used just as effectively. Electroporation of A. baumannii has been well described, and readers may gain further insight into this specific aspect of the protocol here^17,18. The key advantages of this mini-Tn7 chromosomal gene complementation system compared with a plasmid-based complementation system are the ability to regulate the level of expression of the complemented gene through the IPTG-controllable lacI^q repressor system and the choice to remove the gentamicin marker (aacC1 gene) via the flanking FRT sites. It was observed that the tolerance of the cells to the expression of efflux pumps can vary depending on the pump inserted. For example, cells are more sensitive to the expression of AdeIJK compared with AdeABC or AdeFGH; this can be addressed by modifying the concentration of IPTG in culture conditions. Removing the antibiotic marker reduces mutational risk to the strain due to constant selection pressure and also allows for unrestricted antibiotic susceptibility investigations³.

This mini-Tn7 system has been used successfully with Pseudomonas^9,10, Yersinia⁹, Burkholderia¹⁹, Xanthomonas²⁰, and Acinetobacter^5,6,21,22 species, however, some limitations exist. For example, in A. baumannii there is only one functional attTn7 insertion site in the genome⁵, so a strain can be created with multiple deletions, but only one gene at a time can be complemented. Also, this system has not yet been proven effective for Gram-positive bacteria¹⁰.

RND efflux pumps are important facilitators of antibiotic resistance in A. baumannii. What makes them so powerful is their three-part structure that spans across the inner and outer membranes, allowing the removal of antibiotics from the periplasm to outside the cell. The most studied and ubiquitous RND pumps-designated AdeABC, AdeFGH, and AdeIJK-have been shown to eliminate antibiotics encompassing all classes. Elucidating efflux pump function in detail will provide a good starting point to design novel therapeutic options against multidrug-resistant strains. Using the AB258 triple RND pump deletion strain and complementing back one pump at a time allows for the study of each pump independent of the others, discerning for each their unique substrate profiles and most effective inhibitors. Of course, efflux pumps also have a "day-to-day" role in the normal function of the bacterium. Understanding those roles could lead to indirect crippling of the pumps, which would, in turn, interrupt antibiotic efflux, possibly making multidrug-resistant bacteria susceptible to commonly-used antibiotics once again. Generally, the mini-Tn7 system can be used to introduce any gene of interest for detailed study or helpful markers, for example, fluorescent proteins for microscopic imaging⁶.

The increasing prevalence of multidrug resistant bacteria is a concern for us all. Understanding the protective mechanisms provided by efflux pumps in pathogens like A. baumannii is critical for combatting serious infections. This chromosomal single-copy gene expression system is a powerful tool for mechanistic studies as well as for identifying inhibitors to thwart efflux pump function.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
0.2 mL PCR tube	VWR	20170-012	For colony boil preparations and PCR reactions
1.5 mL microfuge tubes	Sarstedt	72-690-301	General use
13-mL culture tubes, Pyrex	Fisher	14-957K	Liquid culture vessels
6x DNA loading buffer	Froggabio	LD010	Agarose gel electrophoresis sample loading dye
Acetic acid, glacial	Fisher	351271-212	Agarose gel running buffer component
Agar	Bioshop	AGR003	Solid growth media
Agarose	BioBasic	D0012	Electrophoretic separation of PCR reaction products; used at a concentration of 0.8–2%
Agarose gel electrophoresis unit	Fisher	29-237-54	Agarose gel electrophoresis; separation of PCR reaction products
Carbenicillin	Fisher	50841231	Selective media
Culture tube closures	Fisher	13-684-138	Stainless steel closure for 13-mL culture tubes
Deoxynucleotide triphosphate (dNTP) set	Biobasic	DD0058	PCR reaction component; supplied as 100 mM each dATP, dCTP, dGTP, dTTP; mixed and diluted for 10 mM each dNTP
Dry bath/block heater	Fisher	88860023	Isotemp digital dry bath for boil preparations
Electroporation cuvettes	VWR	89047-208	2 mm electroporation cuvettes with round cap
Electroporator	Cole Parmer	940000009	110 VAC, 60 Hz electroporator
Ethidium bromide	Fisher	BP102-1	Visualization of PCR reaction products and DNA marker in agarose gel
Ethylenediaminetetraacetic acid (EDTA)	VWR	CA-EM4050	Agarose gel running buffer component
Gentamicin	Biobasic	GB0217	For the preparation of selective media
Glycerol	Fisher	G33	Preparation of bacterial stocks for long-term storage in an ultra-low freezer
Incubator (shaking)	New Brunswick Scientific	M1352-0000	Excella E24 Incubator Shaker for liquid culture growth
Incubator (static)	Fisher	11-690-550D	Isotemp Incubator Oven Model 550D for solid (LB agar) culture growth
Inoculation loop	Sarstedt	86.1562.050	Streaking colonies onto agar plates
Inoculation spreader	Sarstedt	86.1569.005	Spreading of culture onto agar plates
Lysogeny broth (LB) broth, Lennox	Fisher	BP1427	Liquid growth media (20 g/L: 5 g/L sodium chloride, 10 g/L tryptone, 5 g/L yeast extract)
Microfuge	Fisher	75002431	Sorvall Legend Micro 17 for centrifugation of samples
Mini-centrifuge	Fisher	S67601B	Centrifugation of 0.2 mL PCR tubes
Petri dishes	SPL Life Sciences	10090	For solid growth media (agar plates): 90 x 15 mm
Pipettes	Mandel	Various	Gilson single channel pipettes (P10, P20, P200, P1000)
Power supply	Biorad	1645050	PowerPac Basic power supply for electrophoresis
Primers	IDT	NA	PCR reaction component; specific to gene of interest; prepared at 100 μM as directed on the product specification sheet
Sucrose	BioBasic	SB0498	For the preparation of counterselective media for removal of the pFLP2ab plasmid from transformed A. baumannii
Taq DNA polymerase	FroggaBio	T-500	PCR reaction component; polymerase supplied with a 10x buffer
Thermal cycler	Biorad	1861096	Model T100 for PCR
Toothpicks	Fisher	S24559	For patching colonies onto agar plates
Trizma base	Sigma	T1503	Agarose gel running buffer component
Ultrapure water	Millipore Sigma	ZLXLSD51040	MilliQ water purification system: ultra pure water for media and solution preparation, and cell washing
Wide range DNA marker	Biobasic	M103R-2	Size determination of PCR products on an agarose gel
Wooden inoculating sticks	Fisher	29-801-02	Inoculating cultures with colonies from agar plates