DNA cloning, BCM expression studies, and GFP assays were performed in NEB Turbo cells (New England Biolabs) grown in 2× LB media (BD Biosciences). All discrete infection assays, plaque assays, and PACE experiments were performed in E. coli S2060 cells11 grown in Davis Rich Media (DRM). For BCM production and LC/MS analysis, E. coli cells were grown in DRM, while P. fluorescens SBW25 cells were grown in BCM production media (BCMM)3. Plasmids and SPs were assembled with USER cloning, and amplified using Phusion U Hot Start DNA polymerase (Thermo Fisher Scientific). Antibiotics were used at the following concentrations: carbenicillin, 50 μg mL−1; tetracycline, 10 μg mL−1; chloramphenicol, 50 μg mL−1; kanamycin, 25 μg mL−1. DNA for Rho-dependent terminators and for assembling the P. aeruginosa BCM pathway was synthesized by Integrated DNA Technologies. A list of all primers used in this study is available in Supplementary Data 3, and a list of all plasmids used in this study is available in Supplementary Table 1.
Preparation and transformation of competent cells
For creating competent cells of NEB Turbo or S2060 E. coli, overnight cultures were inoculated into 50 mL 2× LB media (BD Biosciences) and grown in an incubator set at 37 °C and 290 rpm to mid-log-phase (absorbance at 600 nm = 0.5). Cells were harvested by centrifugation at 4000 g at 4 °C for 10 min. Cell pellets were resuspended in 1.25 mL of 2× TSS solution (LB media containing 10% PEG 3350, 5% v/v DMSO, and 20 mM MgCl2). Aliquots were frozen at −80 °C for later use.
For transformations, 50 μL aliquots of cells suspended in TSS were thawed on ice and 7.5 μL of KCM solution (100 mM KCl, 30 mM CaCl2, and 50 mM MgCl2) was added, along with the plasmid DNA or USER assembly reaction to be transformed. Cell–DNA mixtures were left on ice for 15 min before heat shock in a 42 °C water bath for 45 s. Transformations were cooled on ice for 1 min before the addition of 750 μL of 2× LB and transfer to a shaking 37 °C incubator for recovery over 45 min. Following recovery, cells were centrifuged at 4000 g, resuspended in a small volume of residual media, and plated on 2× LB plates containing appropriate antibiotics, growing out overnight at 37 °C.
To create electro-competent P. fluorescens SBW25 cells, single colonies were inoculated into 4 mL of 2× LB media (BD Biosciences) in a 15 mL polypropylene culture tube and grown in an incubator set at 28 °C and 300 rpm overnight. On the following day, cells were harvested by centrifugation at 4000 g at 4 °C for 10 min. Cells were washed twice with ice-cold sterile 10% glycerol before being resuspended in 200 μL 10% glycerol. Aliquots of 50 mL cell suspension mixed with DNA were electroporated with 2500 V, resuspended in 1 mL 2× LB, and recovered at 28 °C with shaking for 1 h before being plated on LB agar plates with appropriate antibiotics at 28 °C.
For error-prone PCR mutagenesis, the BCM operon was amplified with Taq DNA polymerase (New England Biolabs) in 96 × 50 μL reactions. PCR amplicons were isolated with a QIAquick PCR purification kit (Qiagen), treated with DpnI (New England Biolabs) to remove residual template DNA, and then gel extracted. This purified pool of mutated BCM operons was amplified again in 96 × 50 μL PCR reactions, using Phusion U polymerase and primers containing deoxyuridine to install USER-compatible termini for assembly. Amplicons were PCR purified, treated with DpnI, and gel extracted before USER assembly with an aTC-inducible plasmid backbone. 72 × 10 μL USER assembly reactions were pooled and transformed into fresh TSS-KCM competent NEB Turbo E. coli cells carrying our superfolder GFP BCM reporter plasmid (equivalent to 100 mL of culture with absorbance at 600 nm = 0.5). Following a 15 min incubation on ice, cells were heat shocked in a 42 °C water bath for 45 s. Transformations were cooled on ice for 1 min before being diluted into 500 mL of 2× LB, shaking in a 37 °C incubator for 45 min to recover. Following this recovery, antibiotics were added for selection of reporter and BCM plasmids. Following overnight growth, this pooled transformation was diluted 1:100 into 100 mL of fresh 2× LB with antibiotics and 100 ng mL−1 aTC to induce BCM production, and left growing for 12–16 h at 30 °C and 290 rpm. After overnight outgrowth and induction, cells were diluted 1:1000 into 1× PBS and sorted using a BD FACSAriaII (BD Biosciences), gating forward and side scatter to enrich for single cells, then sorting based on superfolder GFP fluorescence intensity. Sorted cells suspended in PBS were plated on 2× LB agar with appropriate antibiotics and allowed to grow for 12–16 h at 37 °C. Three hundred eighty-four individual colonies of sorted cells were picked and resuspended into individual wells of a deep 96-well plate containing 2× LB with antibiotics. Plates were grown at 37 °C and 900 rpm until cells reached early–mid log phase (absorbance at 600 nm = 0.2), after which individual wells were diluted into fresh media in four wells in a new plate, including duplicates with and without aTC. New plates were grown for 12–16 h at 30 °C and 900 rpm before cell density (absorbance at 600 nm) and superfolder GFP fluorescence from the reporter was measured on a SpectraMax plate reader (Molecular Devices) to assess BCM production from mutated plasmids recovered following Taq mutagenesis.
XL-1 Red evolution
Fifty microliters aliquots of XL-1 Red E. coli cells (Agilent) were thawed on ice for 15 min before 1 μL of a purified plasmid carrying an inducible copy of the BCM operon was added. Following 15 min incubation on ice, cells were heat shocked in a 42 °C water bath for 45 s. Transformations were cooled on ice for 1 min before the addition of 750 μL of 2× LB and transfer to a shaking 37 °C incubator for recovery over 45 min. Following this recovery, transformations were plated on 2× LB agar containing 50 μg mL−1 chloramphenicol and 10 μg mL−1 ketoconazole, incubating overnight at 37 °C. Colonies of transformants were washed off the plates and resuspended in 2× LB containing matching antibiotics, diluted into a 96-well plate, and grown overnight at 37 °C. Every 24 h for 3 days, the plate was diluted 1:100 into new media and the remaining culture was used to isolate mutated BCM plasmids with a QIAprep Spin Miniprep kit (Qiagen). Following serial passaging and DNA prep, a pooled, mutated plasmid sample was transformed into competent NEB Turbo cells carrying a BCM-sensitive superfolder GFP reporter plasmid, and was allowed to recover in 100 mL of 2× LB for 45 min before the addition of antibiotics, followed by outgrowth for 12–16 h at 37 °C and 290 rpm. Following overnight outgrowth, transformed NEB Turbo cells carrying our BCM-sensitive reporter plasmid and our mutated BCM biosynthesis plasmid were diluted 1:100 into 100 mL fresh 2× LB media with antibiotics and 100 ng mL−1 aTC to induce BCM production, and left growing for 12–16 h at 30 °C and 290 rpm After overnight outgrowth and induction, cells were diluted 1:1000 into 1× PBS and sorted using a BD FACSAriaII (BD Biosciences), gating forward and side scatter to enrich for single cells, then sorting based on superfolder GFP fluorescence intensity. Sorted cells suspended in PBS were plated on 2× LB agar with appropriate antibiotics and allowed to grow for 12–16 h at 37 °C. Three hundred eighty-four individual colonies of sorted cells were picked and resuspended into individual wells of a deep 96-well plate containing 2× LB with antibiotics. Plates were grown at 37 °C and 900 rpm until cells reached early–mid log phase (absorbance at 600 nm = 0.2), after which individual wells were diluted into fresh media in four wells in a new plate, including duplicates with and without aTC. New plates were grown for 12–16 h at 30 °C and 900 rpm before cell density (absorbance at 600 nm) and superfolder GFP fluorescence from the reporter were measured on a SpectraMax plate reader (Molecular Devices) to assess BCM production from mutated plasmids recovered from XL-1 Red transformants.
GFP reporter assays
Single colonies of NEB Turbo E. coli cells carrying a reporter plasmid and potentially also a plasmid containing an inducible BCM operon were used to create glycerol stocks. For assays, 2× LB containing appropriate antibiotics was inoculated with an appropriate stock, sampled with a sterile wooden stick. Cultures were grown overnight at 37 °C and 250 rpm to recover. On the following day, fresh media was inoculated 1:1000 with the overnight cultures grown in similar conditions to absorbance at 600 nm = 0.6–0.8 before being diluted 1:100 into 300 μL of fresh media in a deep 96-well plate for assays. For reporter dose-response assays, 2× LB with antibiotics was serially diluted with BCM (dissolved in methanol; Santa Cruz Biotechnology) and warmed to 30 °C prior to the addition of cells. For assays involving the induction of the BCM operon, 2× LB with antibiotics and with or without 100 ng mL−1 aTC was added to a deep 96-well plate and warmed to the appropriate temperature (30 °C in all cases beyond temperature dependence assays) prior to the addition of cells. Following inoculation, the assay plate was sealed with a breathable membrane and shaken at 900 rpm for 12 h. After incubation with shaking, assay plates were sampled, measuring fluorescence, and absorbance of the cultures on a SpectraMax plate reader (Molecular Devices). For assays involving the BCM efflux pump (bcmT), IPTG for induction of pump expression was added simultaneously with aTC and/or BCM.
S2060 cells were transformed with the pTW004(s) and pTW006 plasmids of interest, carrying the T7 RNA polymerase N-terminal fusions of Bcm proteins, and the T7 C-terminal fragment and luciferase reporter, respectively19. Single colonies were grown overnight in DRM media supplemented with maintenance antibiotics, and were then diluted 1000-fold into DRM media with maintenance antibiotics in a deep 96-well plate (VWR). The plate was sealed with a porous sealing film and grown at 37 °C with shaking at 900 rpm for 4 h, whereupon the culture reached OD600 ~0.4–0.6. Cells were induced with 100 ng mL−1 aTC and 5 µM arabinose before incubation for an additional 1 h at 37 °C with shaking at 900 rpm, before 150 µL of cells were transferred to a 96-well black-walled clear-bottomed plate to measure absorbance at 600 nm and luminescence using a Tecan Spark microplate reader. OD600-normalized luminescence values were obtained by dividing raw luminescence by background-subtracted 600 nm absorbance. The background value was set to the 600 nm absorbance of wells containing DRM only.
Liquid chromatography and mass spectrometry
Mass spectrometry was performed using a Thermo QExactive mass spectrometer (ThermoFisher Scientific, USA) with a HESI II probe, operated in positive mode. Liquid chromatography was performed with a coupled Thermo Dionex UltiMate 3000 HPLC system (ThermoFisher Scientific, USA), using a Luna Omega 1.6 μm Polar C18 column (50 × 2.1 mm, Phenomenex) maintained at 25 °C. Mobile phases were water with 0.1% formic acid or methanol with 0.1% formic acid. For separations, organic solvent was increased from 0 to 35% over 6 min, followed by a wash at 100% starting at 6 min 6 s. BCM fragmentation and retention time were confirmed via analytical standard (Santa Cruz Biotechnology) prior to analysis. High-resolution mass spectrometry data were collected with full scan mode and with parallel reaction monitoring (PRM) for quantification. PRM targeting of BCM was accomplished by filtering for the associated ion (m/z = 285.1 [M + H–H2O]) and subsequent diagnostic MS/MS detection of diagnostic fragment ions. Intensities and peak areas of BCM were measured with MZmine2 software33 and concentrations were calculated by comparison with a standard curve.
For analysis of BCM production in E. coli, NEB Turbo cells were transformed with expression plasmid pBCM.1 or expression and efflux plasmid pBCM.2 bearing either the original BCM sequence or an evolved gene cluster recovered from PACE. Single colonies were grown overnight at 37 °C and 300 rpm in 4 mL DRM with appropriate antibiotics. Following overnight growth, a second set of 4 mL cultures was inoculated 1:100 and grown to absorbance at 600 nm = 0.6–0.8, then diluted 1:100 into 4 mL DRM with antibiotics and 100 ng mL−1 aTC and grown at 28 °C and 300 rpm for 16 h. Cultures were harvested by centrifugation at 3200 g for 15 min at 4 °C using an Eppendorf 5804R centrifuge. Supernatants were sterilized by syringe filtration (0.2-μm filter) and cell pellets were resuspended in methanol, followed by a second centrifugation.
For analysis of BCM production in P. fluorescens, we followed the protocol outlined by Vior et al.3. Briefly, cells were transformed with Pseudomonas protein expression vector pJH, carrying a wild-type or PACE-evolved BCM operon, along with the BCM efflux pump bcmT. Single colonies of the resulting transformations were used to inoculate 4 mL cultures of BCMM in 15 mL polypropylene culture tubes, growing overnight at 28 °C and 300 rpm Following overnight growth, a 400 μL aliquot was diluted into 10 mL fresh BCMM in a 50 mL Falcon tube and covered with a foam bung. Production cultures were grown for 16 h at 28 °C and 300 rpm Cultures were harvested by centrifugation at 3200 g for 15 min at 4 °C using an Eppendorf 5804R centrifuge. Supernatants were sterilized by syringe filtration (0.2-μm filter).
Single colonies of S2060 E. coli cells carrying the activity-independent persmissive plasmid pJC175e11 were grown overnight in DRM media supplemented with antibiotics, before being diluted 1:1000 into fresh DRM media with antibiotics and grown at 37 °C and 250 rpm to absorbance at 600 nm = 0.6–0.8 before use. Phage samples were serially diluted 1:100 four times in DRM media. 100 μL of cells were added to 10 μL of each phage dilution and allowed to sit for ~1 min before addition of 1 mL of warm liquid top agar (2× YT media + 0.4% agar) supplemented with 2% Bluo-gal (dissolved in DMF; Sigma) to a final concentration of 40 μg mL−1 was added and mixed by pipetting up and down once. The resuspended cells and phage dilution were then pipetted onto a quartered Petri dish already containing 2 mL of solidified bottom agar in each quadrant (2× YT media + 1.5% agar, no antibiotics). Plates were incubated at 37 °C overnight after the top agar was observed to have solidified.
Phage propagation assays
Single colonies of S2060 cells carrying an AP of interest were prepared similarly to those used in activity-independent plaque assays. Following dilution into fresh media and growth at 37 °C and 250 rpm to absorbance at 600 nm = 0.6–0.8, 4 mL cultures of cells were divided into two new cultures and prepared SP were added to a final titer of 1 × 105 pfu mL−1. Cultures were moved to a 30 °C shaker set to 250 rpm and left for phage propagation. Phage counts were assessed after 2, 4, and 6 h, collecting supernatants containing phage by transferring 100 μL of culture to a 1.7 mL Eppendorf tube, centrifuging at 2000 g for 5 min. Supernatants were filtered using sterile polyethersulfone 0.2-μm centrifuge filters (mdi Membrane Technologies), centrifuging again at 2000 g for 5 min. Samples were stored at 4 °C for up to 1 week before use.
Phage-assisted continuous evolution
The apparatus for PACE—including tubing, pumps, chemostats, and lagoons—was used as previously described9,11,12 unless noted otherwise. S2060 cells were transformed with the AP pAP.1 and plated on 2× LB agar containing appropriate antibiotics, incubating overnight at 37 °C. A single colony of S2060 with the transformed AP was grown in 4 mL DRM with antibiotics at 37 °C and 250 rpm until absorbance at 600 nm = 0.6–0.8, when it was added to a 100 mL chemostat containing fresh media. The chemostat culture was grown until absorbance at 600 nm = 0.6–1.0, after which pumps were started to supply fresh media at a rate of 1–1.5 vol h−1 to maintain cell density. Lagoons were filled with DRM and constantly diluted with culture from the chemostat following initiation of the pumps. Lagoons were allowed to equilibrate for ~8 h prior to the addition of phage. Unless stated otherwise, lagoons were maintained at a volume of 20 mL and fed at a flow rate of 1 vol h−1 from the chemostat. Any changes in dilution rate during PACE experiments were exclusively implemented through alterations to pumping rate, rather than adjusting lagoon volume. For PACE experiments involving the inducible BCM efflux pump plasmid pPUMP, chemostat DRM was modified with carbenicillin for plasmid maintenance and 5 μM IPTG for constitutive induction.
Following equilibration of lagoon and chemostat cultures, lagoons were seeded with starting titers of ~106 pfu mL−1 and grown for 12–16 h overnight without mutagenesis to reach high titers for the initiation of mutagenic PACE the following morning. After overnight culturing, lagoons were sampled prior to the addition of mutagen (t = 0 h) by removing 3 mL of culture via sterile syringe. For mutagenesis, MNNG (TCI America) was dissolved in DMSO for direct infusion and dilution into lagoons, pumped from 50 mL syringes through Versilon SE-200 chemical transfer tubing (Saint-Gobain) using a syringe pump (New Era Pump Systems) at a flow rate between 0.1 and 1.0 mL h−1. Following the initiation of mutagen addition, lagoons were sampled at regular intervals (roughly every 12 h).
Samples taken from lagoons were processed immediately to isolate phage particles and dsDNA. One milliliter of culture sample was transferred to a 1.7 mL Eppendorf tube, centrifuging at 2000 g for 5 min. Supernatants were filtered using sterile polyethersulfone 0.2-μm centrifuge filters (mdi Membrane Technologies), centrifuging again at 2000 g for 5 min. Samples were stored at 4 °C for up to 1 week before use. The remaining culture sample was centrifuged in the original 1.7 mL Eppendorf tube at 20,000 g for 1 min to collect the cell pellet for plasmid isolation. Genotype assessment and monitoring of the integrity of the BCM operon was performed regularly by PCR off of isolated phage dsDNA using primers BCM_CWJ_USER_F (5′-AGGGGCCCAAGdUTCACTTAAAAAGGAG) and BCM_CWJ_USER_R (5′-ACTCTAGATCAATAdUA-GACCCTGGGTATTCTC) and a standard thermocycler program (98 °C 1 min, followed by 30 cycles of 98 °C for 10 s, 60 °C for 30 s, 72 °C for 3 min 30 s, followed by 72 °C for 10 min). PCR products were assessed by 0.8% agarose gel electrophoresis. For operon recovery, the resulting PCR product was gel excised and cloned into an inducible plasmid backbone. Assembled BCM EPs were transformed into NEB Turbo cells carrying a fluorescent reporter and plated on 2× LB agar containing appropriate antibiotic and 100 ng mL−1 aTC, growing at 30 °C for 48–96 h. GFP expression from individual colonies was monitored via inspection on a SafeImager 2.0 UV imager (Invitrogen). Colonies that expressed GFP were isolated, retested in liquid culture, and grown for plasmid isolation and Sanger sequencing (Quintarabio) to determine genotype.
Phage titers were determined by plaquing filtered culture supernatant samples on S2060 cells carrying the activity-independent permissive plasmid pJC175e11. Samples were checked periodically for the presence of recombinant phage bearing pIII by plaquing on S2060 without pJC175e.
Statistics and reproducibility
All assays and LC–MS quantification presented in this work were repeated independently at least three times and provided reproducible results.
Further information on research design is available in the Nature Research Reporting Summary linked to this article.