Strains, media, and chemicals
Twelve actinobacterial strains (Table 1) were purchased from the German Collection of Microorganisms and Cell Cultures (DSMZ, Braunschweig, Germany). The MRSA ATCC 43300 and E. coli ATCC 25922 were purchased from American Type Culture Collection. Soybean powder was purchased from Wugumf, Shenzhen, China. Soluble starch was purchased from Affymetrix, Santa Clara, CA, USA. Magnesium sulfate hydrate was purchased from Riedel-de-Haën, Seelze, Germany. Bacteriological peptone and tryptone soya broth (TSB) were obtained from Oxoid, Milan, Italy. Mueller-Hinton broth (MHB) was purchased from Fluka Chemie AG, Buchs, Switzerland. Phosphate-buffered saline (PBS) was purchased from Thermo Fisher Scientific Inc., San Jose, CA, USA. Lysogeny broth (LB), glucose, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), and 1-butanol were purchased from VWR International Ltd, Leicestershire, UK. Antibiotics, stains, and all other chemicals were supplied by Sigma-Aldrich Corporation, Saint Louis, MO,USA.
Compound isolation and purification
Elasnin and natural products from 12 Actinobacteria strains were isolated and purified as previously described46. In a typical procedure, stock cultures were inoculated into 50 mL of AM4, AM5, and AM6 media (Table S2) containing glass beads and incubated at 22°C and 30°C on a rotary shaker (170 rpm) for 3, 5, and 7 days. The crude extracts were extracted with 1-butanol and dissolved in DMSO for storage and bioassay. Pure compounds were isolated by reversed-phase high-performance liquid chromatography (HPLC) (Waters 2695, Milford, MA, USA) using a semi-prep C18 column (10×250mm). Elasnin was extracted from the secondary metabolites produced by Streptomyces mobaraensis DSM 40847 after 5 days of incubation in the AM4 medium at 30°C and purified with HPLC.
MICs and MBCs were determined with MRSA ATCC 43300 and E. coli ATCC 25922, according to the Clinical and Laboratory Standards Institute guideline CLSI M100 (2018). In a typical procedure, a 105 CFU/mL overnight culture of test strains was inoculated into MHB and treated with testing compounds at a series of concentrations. After incubation for 24 h, the minimum concentrations at which no bacterial growth was visible were recorded as the MICs. MBCs were measured following the MIC assay by plating 1 mL of suitably diluted culture broth from each well on Mueller-Hinton agar (MHA) plate. MBC was defined as the lowest concentration at which an antimicrobial agent caused > 99.9% reduction in cells. Each assay was performed in duplicate and repeated three times.
MRSA ATCC 43300 was used for the concentration-response curve study. A culture of 4×105 CFU/mL MRSA in the exponential phase was inoculated into MHB with various concentrations of elasnin and vancomycin in 15 mL falcon tubes. Tubes were incubated at 37°C on a rotary shaker for 24 h. Then 1 mL of culture broth in each tube was diluted with MHB, and 1 mL of diluted bacteria was plated on MHA plates for CFU counting. Culture broth from each well was inoculated on two plates, and the experiments were repeated three times.
MBICs and MBECs were determined as previously described 46,60,61. The time-course biofilm formation on MRSA cells is shown in Fig. S2b. An overnight culture of test strains was diluted into approximately 107 CFU/mL with LB and 0.5% glucose and treated with various concentrations of testing compounds in 96-well cell-culture plates. There plates were then incubated at 37°C for 24 h and rinsed twice with 1×PBS to remove non-adhering and planktonic cells. After rinsing, MTT staining assay was conducted to measure viable cells in the biofilms because MTT can react with activated succinate dehydrogenase in viable cell mitochondria to form blue-violet formazan, which can be read at 570 nm after dissolving in DMSO. MBIC50 and MBIC90 were defined as the lowest concentration required to inhibit 50% and 90% of biofilm formation, respectively.
For MBEC assay, an overnight culture of test strains was incubated for 24 h in 96-well cell-culture plates to form mature biofilm before twice rinsing with 1×PBS and compound treatment. After 24 hs incubation at 37°C, each well was rinsed twice with 1×PBS, and OD570nm was recorded after MTT assay as described above. The lowest concentration of a compound resulting in 50% decreases in OD570 nm were recorded as MBEC50. Biofilm inhibition/eradication efficiency was calculated using the following equation: Biofilm inhibition/eradication (%) = [1-(OD570nm of test compound) / (OD570nm of control)] × 100%. Experiments were performed in triplicate and repeated three times.
Biofilm eradication monitoring and resistance study
Mature biofilms of MRSA ATCC 43300 were first grown in 96-well cell-culture plates and treated with various concentrations of elasnin as described above. Plates were then collected after 0, 3, 6, 12, 18 and 24 h treatment and OD570nm were recorded after rinse and MTT assay. To assess the resistance development risks of elasnin, resistance study was conducted as previously described62. MRSA ATCC 43300 were treated with elasnin, vancomycin and ciprofloxacin at final concentrations of 0.5×, 1×, 2×, 4×, and 8× MIC in the MIC assays as described above. After 24 h of incubation at 37 °C, the MICs were recorded and 1 µL aliquots from the culture with the second-highest antibiotic concentration that showed visible growth were diluted 1000 times in MHB for the subsequent assay. This process was repeated for 45 days, and the final MIC was confirmed by the MIC assay. Experiment was performed in quadruplicate and on each day.
HT22 and Neuro2a (N2a) cells were used in the MTT assay to test the cytotoxicity of the compounds. Cells were grown in DMEM with 10% FBS and 1% penicillin-streptomycin at 37 °C with 5% CO2. Then, 5×103 cells were seeded in each well of 96-well plates and cultured for 24 h. After treating the the cells with different concentrations of the compounds dissolved in DMSO for another 24 h, 20 μL of MTT (5 mg/ml) was added to each well and incubated for 4 h at 37 °C before adding 100 μL of DMSO to dissolve formazan. The absorbance was measured using the Multiskan™ FC microplate photometer at 570 nm. IC50 data were analysed with GraphPad Prism software.
CLSM observation with biofilm staining
Biofilms were grown on glass cover slides as those described for the MBIC and MBEC assays. Treated biofilms were then rinsed twice with 1 × PBS and stained with FilmTracer™ FM® 1-43 green biofilm cell stain and FilmTracer™ SYPRO® Ruby Biofilm Matrix Stain at room temperature for 30 min in the dark. A Leica Sp8 confocal microscope was used to observe cells and matrix in the biofilm at 488 nm.
To visualise the changes in biofilm matrix components after elasnin treatment, biofilms were prepared as described above in the MBEC assays and stained with TOTO™-1 Iodide and Concanavalin A to observe eDNA and polysaccharides within the biofilm matrix according to the manufacturer’s instruction. A Zeiss LSM 710 confocal microscope was used for observing and ImageJ was used for quantification.
Total RNA extraction and transcriptomic analysis
Overnight cultures of 107 CFU/mL MRSA cells were inoculated into TSB complemented with 0.5% glucose (TSBG) at 37℃ to get mature biofilms. After 24 h of incubation, mature biofilms were rinsed with 1×PBS twice and treated with 5 μg/mL elasnin or media. Biofilm and released cells were collected at 6 and 12 h and RNA was immediately stabilised with RNAprotect Bacteria Reagent (Qiagen, Hilden, German) according to the manufacturer’s protocol. Total RNA was then extracted with RNeasy PowerBiofilm Kit (Qiagen, Hilden, German) and sequenced using Illumina Novaseq platform with 150bp short-insert library to generate 2 Gb paired-end reads for each sample. The raw reads were trimmed with Trimmomatic v0.3663 to remove adapters and low-quality bases with the setting ILLUMINACLIP: TruSeq3-PE.fa:2:30:10 and then mapped to the S. aureus ATCC 43300 genome (https://genomes.atcc.org/genomes/79691302ed634fef?_ga=2.259377226.1584810311.1616483300-1172888945.1616483300) by using Bowtie2 v2.3.564. Salmon v.0.13.165 was used to quantify the abundance of successfully mapped transcripts, and differential expression analysis was conducted using Perl scripts align_and_estimate_abundance.pl and run_DE_analysis.pl under edgeR66 method in Trinity v2.8.5 67,68 toolkits. Transcripts with false discovery rates < 0.05 and an absolute fold-change value > 2 were defined as DEGs.
Sample preparation for proteomics analysis
Preformed biofilms were prepared the same way as those described for transcriptome analysis and then treated with 5 μg/mL elasnin (or media for control) for 2, 6, and 12 h followed by rinsing twice. Biofilm matrix and total proteins were extracted as previously described69 with slight modification. In a typical procedure, biofilms were collected from the bottom of the dish and washed with washing buffer comprising 10 mM Tris-HCl (pH 8.0) and protease inhibitor cocktail (Sigma-Aldrich) followed by centrifugation at 5,000 g for 10 min. The pellet was dissolved in a matrix-extraction buffer comprising 10 mM Tris-HCl (pH 8.0), 1 M NaCl, and protease-inhibitor cocktail followed by incubation at 25°C for 30 min with gentle rotation. The mixture was centrifuged at 5,000 g for 10 min after incubation, and the supernatant was collected as the biofilm-matrix protein. To extract the total protein, the pellet was lysed with B-PER™ bacterial protein extraction reagent (Thermo Scientific) according to the manufacturer’s instructions and sonicated with a Q125 Sonicator (Qsonica) set at 65% amplitude (five blasts each lasting 15 s with 30 s pauses). The supernatant was collected as the total protein after centrifugation. For all proteomics experiments, three biological replicates were performed for each sample including the control sample.
Collected proteins were desalted with Thermo Pierce C18 spin tips and digested with trypsin (Pierce™ Trypsin Protease, MS Grade) before injecting into the Bruker timsTOF Pro Mass-spectrometer (Bruker Headquarters Billerica, MA, US) with capitve apray ion source. Approximately 200 ng of the digested protein was injected into the Bruker nanoElute system and separated on a C18 column (ionoptiks Aurora UPLC column, Part no. AUR2-25075C18A-CSI) that was eluted with a 30 min gradient of 2 - 95% aqueous acetonitrile containing 0.1% formic acid at a flow rate of 0.3 μL/min. The m/z range acquired in the MS full scan was 100 - 1700 Da.
Sequence database searching and label-free quantification of proteomics data
The generated raw data were converted to mgf files by Bruker Compass DataAnalysis, and subsequently converted to mzML files by msconvert of the ProteoWizard 70. The mzML files were searched using Comet (version 2016.01 rev.2)71 with a custom database. In a typical procedure, the genome sequence of MRSA ATCC 43300 was converted into a protein database using GeneMark72 gene-prediction tool. Proteins were then annotated using BLASTp from NCBI using MRSA NCTC 8325 as the protein database. The sequences of common contaminants such as trypsin and human keratins, as well as decoy sequences generated by shuffling amino acid sequences between tryptic cleavage sites, were added to the database. The decoy sequences in the database were used for the false FDR estimation of identified peptides. The search parameters criteria were set as follows: 15 ppm peptide mass tolerance, monoisotopic mass type, fully digested enzyme termini, 0.05 amu fragment bin tolerance, 0 amu fragment bin offset, carbamidomethylated cysteine, and oxidated methionine as the fixed and variable modifications, respectively. Search results from Comet were processed using PeptideProphet73, iProphet, and ProteinProphet of the Trans-Proteomics Pipeline74 in the decoy-assisted non-parametric mode. Every mzML run was analysed independently. Protein identifications were filtered at a FDR of 0.01 as predicted by ProteinProphet.
Label-free quantification of proteomics data was accomplished by spectral counting. Search results from the two technical replicates of each biological replicate were combined, and proteins identified in at least two out of three biological replicates were used for label-free quantification by spectral counting. Proteins were quantified using the normalised spectral-abundance factor (NSAF)75, where the number of peptide-spectrum matches (PSMs) for each protein divided by the length of the corresponding protein was normalised to the total number of PSMs divided by the lengths of protein for all identified proteins. The DEPs were filtered by the following cutoff: average spectral counts of at least three, p-value for Student’s t-test on the NSAF values was lower than 0.05, and the fold changes were higher or lower than ± 1.5-fold. Moreover, only unique proteins detected in the treatment samples or the control samples were retained for analysis as they were also likely to have increased/decreased expression owing to elasnin treatment. To minimise false positives, we further limit our attention to only uniquely detected proteins with spectral counts greater than 4. Here, we assume that these unique proteins with sufficiently high spectral counts were also induced/upregulated (if detected only in treatment samples and not in control samples) or repressed/downregulated (if detected only in control samples and not in treatment samples).
SEM analysis of biofilms treated with elasnin
Samples for SEM analysis were prepared as previously described with a little modification76,77. Preformed biofilms on a copper strip surface were treated with elasnin (5 μg/mL) or TSBG for 6 h followed by overnight fixation with 4% (v/v) glutaraldehyde under 4 °C. Thereafter, biofilms were dehydrated in a graded ethanol series (30%, 50%, 70%, 90% v/v with distilled water and three times with 100% ethanol, 10 min each step) followed by air drying. Samples were then gold coated using a gold coater Scancoat Six (Edwards, Irvine, CA, USA) and observed using a SEM (JSM-6390, JEOL, Akishima, Tokyo, Japan).
PCA was performed to determine the correlation between individuals and expression level of transcripts on R, using DESeq278. Functional annotation, enrichment analysis of DEGs/DEPs was performed using The Database for Annotation, Visualization and Integrated Discovery (DAVID) v6.879,80 (ease=0.01). Cluster analysis was constructed to reveal the similarity of gene expression between control and elasnin-treated groups based on Bray-Curtis distance matrix in the software PAST (version 2.0)81. To construct the interaction network between the DEGs/DEPs, STRING v1182 was used to predict the protein-protein interactions.
Transcription inhibition and complementation of DEGs in the MRSA
The expression of upregulated DEGs was inhibited using CRISPR/Cas9 transcription-inhibition system pCasiSA, whereas genes downregulated by elasnin were complemented with a tetracycline-inducible expression vector pRMC2 in the wild-type MRSA ATCC 43300. All plasmids, bacterial strains, and primers used in this study are listed in Tables S3 and S4. pRMC2 was a gift from Tim Foster83 (Addgene plasmid #68940; http://n2t.net/addgene:68940; RRID: Addgene 68940). pCasiSA was constructed by mutating pCasSA plasmid as previously described84 and pCasSA was a gift from Quanjiang Ji (Addgene plasmid #98211; http://n2t.net/addgene:98211; RRID: Addgene_98211).
Constructed plasmid was transported into the wild-type MRSA ATCC43300 by Electroporation. In a typical procedure, competent cells were prepared as previously described8383 and stored at −80 °C. For electroporation, 50 μL of competent cells were thawed on ice for 10 min and mixed with 1-2 μg of plasmid, and transferred into a 1 mm electroporation cuvette (Bio-Rad, Hercules, CA, USA). Cells were then pulsed at 2.5 kV, 100 Ω, and 25 μF and incubated in 1mL of TSB at 30 ℃ for 1 h followed by plating on a TSB agar plate containing 7.5 μg/mL chloramphenicol. The plates containing pRMC2 plasmid were incubated at 37 ℃, whereas plates with pCasiSA plasmid were incubated at 30 ℃. Strains containing different plasmids were then used for MBEC and MBIC assay as described above (all strains containing pCasiSA plasmid were incubated at 30 ℃ throughout the entire assay).
Quantitative real-time PCR
A 3 mL overnight culture of mutant MRSA strains (0.2 μg/mL anhydrotetracycline was added in complemented strains) was harvested, stabilized, and total RNA was extracted as described above. cDNA was then synthesised with RevertAid H Minus First-Strand cDNA Synthesis Kit after the removal of genomic DNA using DNase I (Thermo Fisher Scientific Inc., Waltham, MA, USA) followed by quantification on a Roche Diagnostics GmbH LightCycler 480 Instrument II Realtime PCR System using SYBR Green RT-PCR Reagents Kit (Applied Biosystems) with the following procedures: (1) polymerase activation at 95 °C for 10 min, and (2) annealing and extension at 55 °C for 1 min for a total of 40 cycles. The specificity of primer pairs for PCR amplification was checked by the melting-curve method. Two biological replicates and three technical replicates were performed for each sample, and the relative gene-expression level was calculated based on the 2ΔΔCt using gyrB as the internal-reference gene.
Statistical analyses for all data were performed using the GraphPad Prism 8.0.2 software and Microsoft Excel 2012 Edition (Microsoft, Redmond, WA, USA).
The RNA-seq data used in the present study were deposited to GeneBank with the accession codes PRJNA740277, and the proteomics data were deposited to the ProteomeXchange through the PRIDE repository with the dataset identifier PXD026836.