Bacterial strains, plasmids and growth conditions
Strains and plasmids used in this study are listed in Supplementary Table 6. Primers used in this study are listed in Supplementary Data 5. All bacteria were routinely grown at 37 °C in broth (LB; BDTmDifco, USA). For extraction of fragin and/or valdiazen, as well as promoter activity measurements, strains were grown in ABG minimal medium (component A: 16 g (NH4)2SO4, 48 g Na2HPO4, 24 g KH2PO4, 24 g NaCl in 800 ml dH2O; component B: 2 ml 1 M MgCl2 × 6 H2O, 0.2 ml 500 mM CaCl2 × 2 H2O, 0.3 ml, 10 mM FeCl3 × 6 H2O in 800 ml dH2O; component G: 20% glycerol v/v in dH2O; ABG: 100 ml component A, 800 ml component B and 100 ml component G). Yeast extract (0.005%) was routinely added to boost growth except when the effect of iron on the activity of the ham promoter was tested. S. cerevisae BY4741 was grown in Yeast Peptone Dextrose (YPD) broth (BDTmDifco, USA) at 30 °C. Fusarium solani strain DS185 was routinely grown on Malt Extract Agar (MEA) (BDTmDifco, USA) plates at room temperature in the dark.
When required, media were supplemented with antibiotics at the following concentrations: kanamycin (Km) at 100 µg/ml, gentamycin (Gm) at 20 µg/ml, trimethoprim (Tp) at 25 µg/ml, chloramphenicol (Cm) 20 µg/ml.
B. cenocepacia H111 mutants were constructed using three different methods. PCR amplification was carried out using the Phusion® High-Fidelity DNA Polymerase (New England BiolabsInc). Mutation of hamG was accomplished by single crossover. To this end an internal fragment of hamG was PCR amplified using the primers listed in Supplementary Data 5 and cloned into pSHAFT2Gm. The resulting plasmid was subsequently transferred into B. cenocepacia H111 by tri-parental mating. Correct insertion into the chromosome was verified by PCR.
Markerless mutations were either introduced using the pGPI-SceI/pDAI system or a FRT/FLP-based system. Targeted unmarked deletions with the pGPI-SceI/pDAI system were carried out as described previously53. In brief, upstream and downstream regions (~1 kb in size), flanking the region targeted for deletion, were amplified using the primers listed in Supplementary Data 5. The PCR products were digested with appropriate enzymes and purified with the QIAquick PCR purification kit (Qiagen, Germany). After triple ligation of pGPI-SceI with the two homology arms upstream and downstream of the targeted deletion region, the resulting plasmid was electroporated into E. coli SY327 and subsequently transferred into B. cenocepacia strains by tri-parental mating. Correct integration of the plasmid into the host genome was verified by PCR. To enforce a second homologous recombination event, plasmid pDAI-SceI was transferred into the target strain. Double crossover mutants were verified by PCR. Mutations in hamA, hamB, and hamF were introduced by the aid of the pGPI-SceI/pDAI system. For targeted unmarked deletions with the FRT/FLP system, upstream and downstream regions of ~1 kb in size were amplified using the primers listed in Supplementary Data 5. The PCR products were digested with the appropriate restriction enzymes and purified with the QIAquick PCR purification kit (Qiagen, Germany). The upstream fragment was cloned into plasmid pSHAFT-FRT and electroporated into E. coli CC118λpir. The downstream fragment was cloned into plasmid pEX-FRT and electroporated into E. coli MC1061. Both resulting plasmids were sequentially transferred into the target strain by conjugation. Integration of both plasmids into the chromosome via homologous recombination was verified by PCR. Plasmid pBBR5-FLP was introduced into the recipient and deletion of the targeted DNA region was verified by PCR. Mutants carrying the deletion were subsequently incubated on M9 minimal medium agar plates containing 10% sucrose as sole carbon source to cure the strain from plasmid pBBR5-FLP. Mutations in hamC, hamD, hamE, and afcA were generated by the aid of the FRT/FLP system.
In silico methods
The architecture search mode of the MultiGeneBlast20 software was used to identify gene clusters similar to the ham cluster in other bacteria. The amino acid sequences of all Ham proteins (HamA–HamG) were combined in one fasta file and used for the search against a database created from the full bacterial GenBank subdivision. All parameters of the MultiGeneBlast software were used with default settings.
Dual-culture plate assay
F. solani was prepared by placing a fungal plug in the middle of a MEA plate and incubated for 8 days at room temperature in the dark. Bacteria from overnight LB cultures were pelleted, washed and adjusted to an OD600 of 1. Three 10 µl samples of the bacterial suspension were spotted on MEA plates equidistant from the center and the plates were allowed to dry for 10 min. Following overnight incubation at 37 °C, plugs from the edge of 8 day old F. solani plates were taken and placed in the center of the MEA plates with the spotted bacteria. The plates were sealed with parafilm and incubated for 8 days at room temperature in the dark. Pictures were taken with a Nikon D90 camera with AF-S Micro Nikkor 60 mm objective and used to determine the antifungal activity of each bacterial strain. The distance between bacterial colonies and the fungus (mm) was calculated from the pictures using the ImageJ software. Chemical complementation of H111 ∆cepI was achieved by supplementing MEA plates with 200 nM C8-homoserine lactone (Fluka, Buchs, Switzerland).
Fungal spray assay
A single plug from an F. solani plate was incubated in a shaking flask containing 100 ml LB with 70 rpm shaking at room temperature for 2 to 3 days. The liquid F. solani culture was homogenized with glass beads by thorough vortexing. Bacterial strains were grown overnight at 37 °C with 220 rpm agitation, pelleted and resuspended in LB (OD600 of 1.0) and 20 µl samples were spotted on MEA plates. After incubation at 37 °C for 20 h, the homogenized fungus was sprayed on the MEA plates. The plates were sealed with parafilm and incubated in the dark at room temperature for 48 h. The antifungal activity of the bacterial strains resulted in a fungus-free halo around the bacterial colony. Pictures were taken with a Nikon D90 camera with an AF-S Micro Nikkor 60 mm objective. Antifungal activity was defined as areas (mm2) with no fungal growth surrounding bacterial colonies and was calculated from pictures with the ImageJ software.
Assessment of promoter activity in liquid culture
Promoter activity of transcriptional lacZ fusions in liquid cultures was assessed by β-galactosidase assays as described before57 with minor modifications. Briefly, bacterial cells were grown overnight in ABG minimal medium. Where indicated, synthetic valdiazen or (−)-fragin dissolved in methanol was added to the cultures. To assess the influence of iron on promoter activity, no yeast extract was added to the ABG medium and the FeCl3 was replaced with the concentrations indicated in the results section. Bacterial overnight cultures were routinely adjusted to an OD600 of 2, centrifuged at 5000 rpm for 5 min and resuspended in 2 ml Z-buffer. A total of 1 ml bacterial suspension was used to determine the exact OD600 value. To permeabilize the cell membrane, 25 µl of chloroform and 0.1% SDS were added to the residual 1 ml of bacterial suspension, vortexed for 10 s and incubated at 28 °C for 10 min. Overall, 200 µl of o-nitrophenyl-β-d-galactoside (ONPG) solution (4 mg/ml in Z-buffer) were added to each sample, vortexed briefly and incubated at room temperature. The reaction was stopped by the addition of 500 µl 1 M Na2CO3. The samples were centrifuged at 16,000 rpm for 10 min and 1 ml of cell debris-free supernatant was used to measure the absorbance at 420 nm and 550 nm. Specific activities (Miller Units) are shown. To validate the cell density-dependent production of valdiazen, growth (OD600) and promoter activity (Miller Units) was monitored throughout the growth curve from samples collected every 2 h for a period of 28 h.
RNA-Seq and data analysis
For RNA extraction, 50 ml ABG aliquots were inoculated from starter cultures (5 ml LB broth) with an OD600 of 0.03 and incubated at 37 °C with agitation (220 rpm). Synthetic valdiazen resuspended in methanol was added from the start to a final concentration of 50 µM when appropriate. An equal volume of methanol was added to all cultures not containing synthetic valdiazen. The final methanol concentration in all cultures was 0.1%. Three independent cultures were grown to an OD600 of 0.9 to 1 and harvested, using 1/10 of stop solution (10% phenol buffered with 10 mM Tris-HCl pH 8). RNA extraction and genomic DNA removal were performed as previously described58. A total of 150 ng good quality RNA (RNA integrity factor > 6) were further processed (cDNA synthesis and library preparation) using the Ovation® Complete Prokaryotic RNA-Seq Library System from NuGEN. After quantification of the obtained cDNA libraries59, Illumina single-end sequencing was performed on a HiSeq2500 Instrument. The CLC Genomics Workbench v7.0 (QIAGEN CLC bio, Aarhus, Denmark) was used to map the sequencing reads to the B. cenocepacia H111 reference sequence60. Statistics and differential analysis was done using the DESeq software61
Extraction of fragin and valdiazen
Bacterial strains were grown in ABG minimal medium containing 0.005% yeast extract for 72 h with agitation (220 rpm) at 37 °C. Bacterial cultures were centrifuged at 5000 rpm with an Eppendorf Centrifuge 5804R. To remove all bacterial cells, supernatants were subsequently filtered with a Millipore ExpressTm Plus 0.22 µm system. For fragin extraction, the cell free supernatant was extracted twice with 0.5 equivalents v/v chloroform in a separation funnel. The two chloroform phases were combined, dried with anhydrous magnesium sulfate (Sigma-Aldrich, Switzerland), and filtered through folded filters (grade: 3 hw; Sartorius, Switzerland). The chloroform extracts were stored at 4 °C. For valdiazen extraction, cell free supernatant was alkalized to a pH of 11 with 10 M NaOH and extracted twice with 0.5 volumes dichloromethane. The dichloromethane phases were discarded and the water phase was subsequently acidified to pH 5 with 10 M HCl and extracted twice with 0.5 volumes dichloromethane. The two dichloromethane phases were combined, dried using anhydrous magnesium sulfate (Sigma-Aldrich, Switzerland) and filtered. The dichloromethane extracts were stored at 4 °C. Chloroform and dichloromethane extracts were dried using an Eppendorf Concentrator 5301 under vacuum and routinely dissolved in 0.01 volumes methanol. The concentrated extracts were stored at −20 °C.
All reactions were performed in dry solvents under an argon atmosphere unless stated otherwise. The anhydrous solvents were obtained from commercial suppliers and used without any purification other than that they were filtered and passed through activated anhydrous alumina columns (Innovative Technology solvent purification system) prior to use. Syringes and stainless steel cannula were used to transfer air and moisture sensitive liquids and solutions. Analytical thin layer chromatography (Merck silica gel 60 F254 plates) was used to monitor reactions and the compounds were detected by UV light (254 nm and 350 nm) and by staining using ceric ammonium molybdate (CAM) solution followed by gentle heating with a heat gun. Flash chromatography was performed using SiliCycle silica gel 60 (230–400 Mesh) and Rf values of compounds are indicated. NMR experiments were performed at 25 °C on a Varian Gemini Bruker DPX operating at 400 MHz, on a Bruker Avance III NMR spectrometer operating at 400 MHz proton frequency, a Bruker Avance III NMR spectrometer operating at 500 MHz proton frequency and at 126 MHz carbon frequency, a Bruker Avance III NMR spectrometer operating at 600.13 MHz proton frequency and 1H-decoupling on a Varian Gemini 101 MHz spectrometer. The spectra were calibrated using the residual solvent proton and carbon signals (δH 7.26, δC 77.16 for CDCl3, and δH 3.31, δC 49.00 for CD3OD). Melting points (Mp) were determined using a Büchi B-545 apparatus in open capillaries and are uncorrected. IR spectra were recorded on a Varian 800 FT-IR ATR spectrometer or on a Spectrum Two (UATR) FT-IR Spectrometer (Perkin Elmer) and data are reported in terms of frequency of absorption (ν, per cm). Optical rotations were recorded on a Jasco P-2000 digital polarimeter with a path length of 1 dm, using the 589.3 nm D-line of sodium. Concentrations (c) are quoted in g/100 ml. High-resolution masses (HRMS-ESI) were recorded on a Bruker max is 4G QTOF ESI mass spectrometer or a QExactive instrument (Thermo Fisher Scientific, Bremen, Germany) equipped with a heated electrospray (ESI) ionization source. HPLC purifications were performed on a UHPLC Dionex Ultimate 3000 system equipped with an Ultimate 3000 pump, an Ultimate 3000 autosampler, an Ultimate 3000 thermostated column compartment and an Ultimate 3000 photodiode array detector or a HPLC Agilent series 1100 equipped with a Quaternary pump, a degasser, an autosampler, a thermostated column compartment and a VWD UV detector. HPLC-MS analyses were performed on a Dionex HPLC system equipped with a P680 pump, an ASI-100 automated sample injector, a TCC-100 thermostated column compartment, a PDA-100 photodiode array detector and a MSQ-ESI mass spectrometric detector. Lyophilization was performed using a Christ Alpha 1–2 LD plus system.
X-ray crystal-structure analysis of the fragin–copper complex from the University of Basel: data collection was performed at −150 °C using CuKα radiation on a Bruker Kappa APEX diffractometer. Integration of the frames and data reduction was carried out using the APEX2 software62. The structure was solved by direct methods using SIR9263. All non-hydrogen atoms were refined anisotropically by full-matrix least-squares on F using CRYSTALS64.
X-ray crystal-structure analyses of (1) and (12) from the Univeristy of Zurich: all measurements were made at −113 °C on an Agilent Technologies SuperNova CCD area-detector diffractometer65 using CuKα radiation. Integration of the frames and data reduction was carried out using CrysAlisPro65. The structures were solved by direct methods using SHELXS-201366. All non-hydrogen atoms were refined anisotropically by full-matrix least-squares on F2 using SHELXL-201467.
Isolation of (−)-fragin (1)
A comparison of the HPLC spectra of concentrated extracts of B. cenocepacia H111 wild type and the hamD mutant identified a compound, (−)-fragin (1), that is only present in the wild-type extract. The HPLC analysis was performed using a reversed phase column (Phenomenex Gemini-NX 5 µm; 250 × 4.6 mm with flow of 1 ml/min). The column was equilibrated for 5 min with 20% MeCN/H2O and the MeCN/H2O applied gradient was changed from 2 to 100% MeCN in 25 min. The column was then washed with MeCN for 7 min. A peak at 20.4 min was identified as (−)-fragin (1). The separation was achieved by HPLC using a reversed phase column (Phenomenex Gemini-NX 5 µm; 250 × 4.6 mm with flow of 1 ml/min). The column was equilibrated for 5 min with 20% MeCN/H2O and the MeCN/H2O applied gradient was changed from 20 to 80% and finally to 100% MeCN in 15 and 1 min, respectively. The column was then washed with MeCN for 8 min. A peak at 14.8 min was isolated and the antifungal activity of this fraction was tested. The procedure was repeated using multiple HPLC runs with the same conditions described above and (−)-fragin (1.4 mg) was obtained as a white solid. NMR experiments were performed in CDCl3 that was filtered through aluminum oxide (activated, basic). (−)-Fragin (1.4 mg) was dissolved in warm toluene (40 °C) and cooled slowly to RT to obtain a single crystal, which was analyzed by X-ray crystallography. The absolute configuration of (−)-(R)-fragin (1) was determined by the crystal structure analysis and by comparing the optical rotation values of the natural and synthetic compound.
(−)-(R)-Fragin (1): white solid; MS (ESI): [M+H]+: 244.2 and 274.2; [M−H]−: 272.2; Optical rotation: [α]D –122° (c1.9, EtOH); 1H-NMR: (500 MHz, CDCl3) δ = 11.72 (s, 1H), 5.69 (s, 1H), 4.20 (td, J = 9.2, 3.1 Hz, 1H), 3.86 (ddd, J = 14.4, 6.0, 3.1 Hz, 1H), 3.60 (ddd, J = 14.4, 9.4, 6.1 Hz, 1H), 2.25–2.17 (m, 1H), 2.14 (td, J = 7.4, 1.8 Hz, 2H), 1.61–152 (m, 2H), 1.31–1.23 (m, 8H), 1.07 (d, J = 6.8 Hz, 3H), 0.90 (d, J = 6.7 Hz, 3H), 0.87 (t, J = 6.8 Hz, 3H); 13C NMR: (from HMQC, CDCl3) δ = 173.7, 78.0, 39.1, 36.6, 31.8, 29.3, 29.1, 29.1, 25.7, 22.7, 19.1, 18.9, 14.2 (as determined from the 13C NMR of the synthetic compound).
Reactivity of the hydroxylamine 10a with sodium nitrite
To a solution of the hydroxylamine 10a (1.4 µM, 0.5 ml) in MeOH, Na15NO2 (6.9 mg) and water (0.5 ml) were added. The solution was filtered and analyzed by HPLC-MS (ESI positive mode). Three data points were recorded: after the preparation of the sample, after 24 h and after 44 h.
Isolation of valdiazen (2)
Concentrated valdiazen extracts were separated using multiple RP-HPLC runs (Phenomenex Synergi Hydro-RP 4 µm; 250 × 4.6 mm, 40 °C) at a flow of 1 ml/min. The MeCN/aq. NH4OAc (20 mM, pH = 4) gradient applied was changed from 2 to 50% and finally 50 to 100% in 15 and 7.1 min, respectively. The column was then washed with MeCN for 4 min. Valdiazen, which has a retention time of 10.2 min, was collected during each run. The fractions were combined, rotary evaporated, dissolved in CH2Cl2 (5 ml), filtered over wool and evaporated to dryness to afford 0.5 mg of valdiazen (2) as a white solid.
Valdiazen (2): white solid: 1H-NMR (500 MHz, MeOD): δ = 4.04–3.96 (m, 2H), 3.87–3.80 (m, 1H), 2.11 (ddh, J = 13.3, 7.2, 6.7, 1H), 1.02 (d, J = 6.8 Hz, 3H), 0.90 (d, J = 6.8 Hz, 3H); 13C-NMR: (126 MHz, MeOD) δ = 82.2, 61.2, 29.3, 19.4, 19.4.
HPLC chiral separation of valdiazen (2)
Valdiazen (2) was separated by HPLC using a chiral column (Chiralpak OD-H 10 µm; 250 × 4.6 mm, 38 °C) at a flow of N0.8 ml/min. The hexane/EtOH isocratic gradient was 80%. Control samples of pure (−)-valdiazen (12) eluted at 10.3 min, while (+)-valdiazen (13) eluted at 7.6 min. Two corresponding peaks at 10 min and 7.5 min were observed for valdiazen (2).
Transcriptional lacZ fusions
DNA fragments containing the putative promoters of the hamABCDE and hamFG operon were amplified with Phusion® High-Fidelity DNA Polymerase (New England BioLabs Inc.) according to the manufacturer’s instructions using the primers listed in Supplementary Data 5. Annealing temperatures were calculated with the NEB Tm calculator online program (New England BioLabs Inc.). The promoter fragments were separated on agarose gels, excised and purified with the QIAquick gel extraction kit (Qiagen). The purified fragments were digested with the restriction enzymes (New England BioLabs Inc.) indicated in Supplementary Data 5 according to the manufacturer’s instructions and purified by the QIAquick PCR purification kit (Qiagen). The lacZ-fusion vector pSU1168 was digested with the restriction enzymes XhoI and HindIII (New England BioLabs Inc.) according to the manufacturer’s instructions and subsequently purified by the QIAquick PCR purification kit (Qiagen). The promoter fragments were ligated into pSU11 with T4 ligase (New England BioLabs Inc.) according to the manufacturer’s instructions. The resulting plasmids were electroporated into E. coli Top10 and subsequently transferred into the target strain by tri-parental mating and selected on PIA plates containing gentamycin (20 µg/ml).
Quantitative real-time PCR
RNA was extracted from B. cenocepacia H111 wild type and ΔhamD cells grown to late exponential phase in ABG medium, as previously described69 and further purified using the RNeasy Qiagen kit (Qiagen, Germany). First strand cDNA was synthesized using random primers (Invitrogen, USA) and MLV reverse transcriptase (Promega, USA). qPCR was performed on the generated cDNA using Brilliant III Ultra-Fast SYBR® Green QPCR Master Mix (Agilent, USA) and a Mx3000P instrument (Agilent, USA). Primers used are listed in Supplementary Data 5. Each PCR reaction was run in triplicate and melting curve data was analyzed to determine the PCR specificity. Relative expression levels of several differentially regulated genes from the RNA-seq data were calculated using the ΔΔ CT method70 and the rpoD gene was used as the reference gene for normalization.
Construction of plasmids for the complementation of mutants
Plasmids for complementation purposes were constructed by PCR of the respective gene using the Phusion® High-Fidelity DNA Polymerase (New England Biolabs Inc). For all genes, except hamB, the amplified fragment included the predicted ribosome binding site (RBS) of the gene. The 5′-end of hamB overlaps with the 3′-end of hamA and does not have its own RBS; therefore, a synthetic RBS was introduced with the forward primer upstream of the start codon of hamB. All primers used for the construction of these plasmids are listed in Supplementary Data 5. The annealing temperatures for the PCRs were calculated with the NEB Tm calculator online program (New England BioLabs Inc.). The amplified fragments were separated on agarose gels, excised and purified with the QIAquick gel extraction kit (Qiagen). The purified fragments were digested with the restriction enzymes (New England BioLabs Inc.) indicated in Supplementary Data 5 according to the manufacturer’s instructions and again purified by the QIAquick PCR purification kit (Qiagen). The expression vector pBBR1MCS2 was digested with the restriction enzymes HindIII and XbaI (New England BioLabs Inc.; expression of hamA, hamB, hamD, hamF, and hamG) or XhoI and XbaI (New England BioLabs Inc.; expression of hamC and hamE) according to the manufacturer’s instructions and subsequently purified by the QIAquick PCR purification kit (Qiagen). The promoter fragments were ligated into pSU11 with T4 ligase (New England BioLabs Inc.) according to the manufacturer’s instructions. The resulting plasmids were electroporated into E. coli Top10 and subsequently transferred into the target strain by tri-parental mating.
Plasmids were transferred from E. coli to B. cenocepacia strains by tri-parental mating using E. coli HB101/pRK600 as the helper strain as described previously71. E. coli strains (donor and helper strain) and B. cenocepacia (recipient strain) were grown in LB overnight with agitation (220 rpm) at 37 °C. The donor and helper strains were mixed and incubated at room temperature for 10 min before the recipient strain was added. The bacterial mixture was then spotted on LB agar plates and incubated at 37 °C for 4 to 7 h, washed off with saline (0.9% NaCl solution) and plated on Pseudomonas Isolation Agar (PIA; BDTmDifco, USA) containing appropriate antibiotics and incubated overnight at 37 °C. Transconjugants were verified by PCR using the primers listed in Supplementary Data 5.
HamC protein purification and enzyme assay
The hamC gene was amplified with Phusion® High-Fidelity DNA Polymerase (New England BioLabs Inc.) using the primers listed in Supplementary Data 5. The resulting PCR product was purified from an agarose gel by the QIAquick gel extraction kit (Qiagen), digested with the restriction enzymes indicated in Supplementary Data 5, and cloned into pQE30. The resulting plasmid, pQEhamC (Supplementary Table 6), was electroporated into E. coli M15. 200 ml LB were inoculated with an overnight culture of E. coli M15 pQEhamC to an OD600 of 0.02 and grown with agitation (220 rpm) at 37 °C. After 3 h of growth, the cultures were shifted to 30 °C and the expression of his-tagged HamC was induced with 500 µM IPTG. Every hour, starting at the induction, 100 µM ferrous ammonium sulfate was added to the cultures to ensure incorporation of iron into the protein. After 3 h, the cells were centrifuged (3000×g, 4 °C, 10 min) and the pellet was frozen in liquid nitrogen and stored at −80 °C. For protein purification, the cells were thawed on ice and resuspended in 5 ml lysis buffer (100 mM NaH2PO4; 500 mM NaCl, 5 mM Imidazol; 10% Glycerol; pH8) containing protease inhibitor (Roche Protean Tablet EDTA free) and lysed with a French press at 2 kBar. The cell lysate was centrifuged at 10,000×g at 4 °C for 60 min. Purification of his-tagged HamC was performed with ProfinityTM IMAC Resin (BioRad) according to the manufacturer instructions with slight modifications. In brief, the lysate was incubated with washed resin for 60 min at 4 °C with shaking, centrifuged (1000×g, 4 °C, 2 min) and washed twice with washing buffer (50 mM NaH2PO4, 500 mM NaCl, 30 mM Imidazol, pH8). The pellet was resuspended in 500 µl elution buffer (50 mM NaH2PO4, 500 mM NaCl, 250 mM Imidazol, pH8) and centrifuged (1000×g, 4 °C, 2 min). The supernatant was collected, flash frozen with liquid nitrogen and stored at −80 °C. The elution step was repeated three times and the collected supernatants pooled. Washing and elution steps of his-tagged HamC were examined by SDS-PAGE and Coomassie blue staining. The eluted protein was subsequently concentrated and the buffer exchanged (25 mM MOPS, 400 mM NaCl, pH7) using Amicon Ultra Centrifugal Filter Units (NMWL 10 kDa) (EMD MILLIPORE). The purity of the concentrated protein was estimated by SDS-PAGE and the protein concentrations were determined with Bradford Reagent (Sigma-Aldrich).
The enzymatic activity of purified HamC was studied as previously described for PsAAO38 with modifications. To a solution of water (76.1 µl), PABA (12.5 µl, 40 mM in DMSO), NaCl (1.8 µl, 1 M in H2O), MOPS (2.3 µl, 1 M in H2O) and a solution of the enzyme HamC (6.4 µl) in MOPS buffer (25 mM) with NaCl (100 mM) were sequentially added. To start the reaction H2O2 (5 µl, 30% in H2O) was added and the reaction was analyzed by HPLC-MS after 1 day. A control experiment was performed using water instead of the enzyme solution.
Chemogenomic profiling including HIP and HOP
The growth-inhibitory potency of test substances was determined using wild type S. cerevisiae BY4743. OD600 values of exponentially growing cultures in rich medium were recorded with a robotic system. Twelve-point serial dilutions were assayed in 96-well plates with a reaction volume of 150 µl, start OD600 was 0.05. Solutions containing dimethyl sulfoxide (DMSO) were normalized to 2%. IC30 values were calculated using logistic regression curve fits generated by TIBCO Spotfire v3.2.1 (TIBCO Software Inc.). HIP, HOP, and microarray analysis were performed as described previously30. Sensitivity was computed as the median absolute deviation logarithmic (MADL) score for each compound/concentration combination. z-scores are based on a robust parametric estimation of gene variability from over 3000 different profilings and were computed as described in detail in Hoepfner et al.30
Quantification of valdiazen
The concentration of valdiazen (2) in the supernatant of the H111 wild type and the H111 ∆hamF strain was determined by UHPLC/MS measurements (ultimate 3000 system equipped with a Kinetex® EVO C18; 1.7 µm; 100 Å, 50 × 2.1 mm as the column, a flow of 0.4 ml/min and a solvent system consisting of MeCN/H2O (both containing 0.1% formic acid)). The gradient used varied from 2 to 2% in 0.9 min, 2 to 30% in 0.3 min, 30 to 95% in 2.3 min, 95 to 100% in 0.05 min and the column was washed with 100% MeCN for 1.24 min. The samples used for the calibration curve were composed of four aqueous solutions of the synthetic valdiazen (Supplementary Methods) at 10, 20, 50, and 100 µg/ml and 1 µl was injected. The supernatants of the H111 wild type and the H111 ΔhamF mutant were filtered and 2 µl were injected. For the quantification, the chromatograms at wavelengths between 240 and 250 nm were analyzed and, to confirm the identity of the peak, SIM measurements at 149.1 Da were performed. A valdiazen concentration of 6.0 µg/ml (41 µM) was calculated in the supernatant of the H111 ΔhamF mutant and 3.3 µg/ml (22 µM) in the supernatant of the H111 wild-type strain.
Quantification of (−)-fragin
The quantification of fragin (1) in the supernatant of the H111 wild type strain was performed as described for the quantification of valdiazen with some modifications. The solvent system of the UHPLC was composed of MeCN/H2O (both containing 0.1% formic acid), the column was equilibrated at 30% for 0.4 min, the gradient was varied from 30 to 95% in 2.1 min, 95 to 100% in 0.05 min and the column was washed with 100% MeCN for 1.24 min. The samples used for the calibration curve were composed of synthetic fragin (Supplementary Methods) solutions (H2O/MeCN 1/1) from 50, 100, 200 and 500 µg/ml and 1 µl of these samples and the supernatant of H111 wild-type strain were injected. The quantification was done using the SIM chromatograms at 244.1 Da. The concentration of fragin was calculated to be 69 µg/ml (253 µM) in the supernatant of the H111 wild-type strain.
Statistical significance for differences of antifungal activity in dual-culture plate assays was calculated with either one-way ANOVA and Tukey’s multiple comparison as a post test or a paired two-tailed Student’s t-test. Differences were considered significant with p < 0.05.
Crystallographic data have been deposited in the Cambridge Crystallographic Data Centre (www.ccdc.cam.ac.uk/structures) with accession codes CCDC-1543487, CCDC-1543488, and CCDC-1812456. The RNA-Seq raw and processed data have been deposited in the GEO database with accession code GSE97171. The authors declare that all other data supporting the findings of this study are available in this article and its Supplementary Information files, or from the corresponding authors upon request.