Strains, Media, and Growth conditions
All the strains used in this study are listed in Table S1 and all the oligos used in gene amplification, genetic manipulations, and construct verifications are listed in Table S2. All of the Bacillus erythromycin knockout (BKE) strains in 168 background were obtained from BGSC or from the National BioResource Project (NBRP, Japan).
Experiments for E. coli and B. subtilis growth were performed in LB medium (lysogeny broth; 5 ml in 25 ml tube) or agar (20 ml in petri plate) with appropriate antibiotics. We used kanamycin in final concentrations of 30 μg ml−1 for E. coli and 15 μg ml−1 for B. subtilis. The erm gene encodes macrolide lincosamide streptogramin (MLS) resistance and is selected with erythromycin (1 μg ml−1) and lincomycin (25 μg ml−1). 96 well plates with 0.1 ml of LB were inoculated with mid-log (OD 600 nm = 0.4) cultures at 50-fold dilutions in final volume for growth measurements. These aerobic cultures were monitored periodically every 60 min for 37 °C growth at 600 nm in Synergy H1 plate reader (BioTek Instruments Inc, VT).
Erythromycin specific-guide RNA (erm-gRNA) cloning
To target the MLS (erm) cassette we selected the 20 nt sequence 5’TTTGAAATCGGCTCAGGAAA3’ followed by AGG of PAM sequence (supplemental file S2), which spans the 35th-41st codon of erm. We added BsaI compatible ends to this sequence and its reverse complement sequence (Table S1) and oligos were mixed in 1:1 ratio (10 μM each) in TE buffer (pH 7.0). This mixture was heated at 95 °C for 10 min. and cooled to room temperature to generate double stranded erm-gRNA with BsaI compatible 5’ and 3’ ends (Fig 1), prior to cloning into pJOE8999 (11) digested with BsaI. The ligation reaction was transformed into E. coli DH5α and plated on LB agar with X-gal (40 μg ml−1) plus kanamycin. White clones (pJOE8999 containing erm-gRNA) were selected and subjected to PCR verification and sequencing at Cornell Biotechnology Resource Center facility to verify the erm-gRNA insertion. Culture of E. coli harboring erm-gRNA in pJOE8999 plasmid (pAJS23) was stored as HEAS25630 (Table S1).
Genome editing in Bacillus with a cloned repair template
Repair templates were generated using overlap extension (SOEing) PCR or PCR amplification of a suitable region from genomic DNA. The outermost primers were designed to include SfiI restriction sites (see SI Fig 2). The PCR product was digested with SfiI and then ligated into pAJS23 plasmid that had also been digested with SfiI and column purified to remove the small (12 bp) excised fragment. The repair template and vector were incubated with T4 DNA ligase enzyme at room temperature for 1 hr (or 18 hr at 16 °C), and then transformed into E. coli DH5α with selection for AmpR. The desired clone was confirmed with analytical PCR, and then transformed into E. coli TG1 to generate multimeric plasmid DNA. The resulting pAJS23 derivative can then be used for genome-editing of the appropriate strain containing an integrated erm cassette.
For each application, we transformed the appropriate recipient strain from the BKE collection with the corresponding pAJS23 derivative. Recipient cells (5 ml) were grown in competence medium to OD 600 nm ~0.8, 1 ml was removed as a negative control, and 1 μg plasmid was added to the remaining and culture and incubated for 2 hrs at 30 °C with shaking (250 rpm). Cells (from 1 ml culture) were plated on LB agar containing 15 μg ml−1 kanamycin (to select for plasmid) and 0.2% mannose (to induce gRNA expression) and incubated for 24-48 hrs at 30 °C. The resulting colonies were passaged three times on LB plates (without any antibiotics) at 45 °C for 18-24 hrs each time. The resulting clones were then tested for loss of the plasmid by patching on LB plates with MLS, kanamycin, or no antibiotics. Alternatively, colonies may be first patched onto LB plates at 50 °C, and then colony purified at 42 °C as previously described (11). Clones that failed to grow when patched onto both MLS and kanamycin test plates were selected. The presence of the desired genome change was confirmed using PCR amplification and Sanger sequencing.
We amplified three overlapping fragments by PCR including Bacillus prkC-fragment (767bp), the rsgA gene (cpgA ortholog) from S. aureus Newman (876bp), and rpe (799bp). These PCR fragments were fused to generate a single PCR fragment consisting with SfiI restriction sites for cloning into pAJS23.
For this FLAG epitope fusion, we chose LFH PCR based in-frame sodA ORF fusion to nucleotides GATTATAAAGATGATGATGATAAA coding for an extended DYKDDDDK fusion tag on Mn-SOD protein.
For yqgC gene fusion to gfp, we generated chimeric PCR by individually amplifying three overlapping fragments. The fragment one consisting of yqgB-yqgC (1205bp) and second fragment consisting of mostly 712 bp long yqgC adaptors-gfp (taken form pGFP-star plasmid (34)-sodA adaptors and sodA-yqgE fragment (1467bp).
To introduce a point mutation in yceF, we employed LFH primers encoding an Ile206Thr in yceF (yceF*). In this approach we generated and joined two fragments into a single fragment.
To delete the S936 RNA feature in the non-coding region between yqgC-sodA element, we engineered a 177 bp deletion spanning reference positions from 2586043 to 2586220 in the Bacillus genome NC_000964.3. We ordered a gBlock with SfiI recognition sequences at both ends from IDT (Integrated DNA Technologies, Inc. IA, USA), which was cloned into pAJS23 to generate pAJS25.
pAJS29 and pAJS30 (ΔyqgC)
To convert yqgC::erm to ΔyqgC we selected primers such that the length of the repair template was either 546 bp or 1594 bp (Table 1). These repair template were PCR amplified from ΔyqgC genomic DNA (generated from yqgC::erm using pDR244 as described; (9)). The resultant strain has an unmarked yqgC deletion (ΔyqgC) with a 150 bp scar in place of the erm resistance cassette.
Genome editing by co-transformation with a repair template
6 ml of recipient Bacillus subtilis strain was transformed as above with 1 μg of pAJS23 plasmid DNA mixed with either 1 to 2 μg of PCR DNA or with 1 μg of sheared/sonicated (20 sec. 40% amplitude) genomic DNA. 1 ml of culture was removed as a negative control. The competent cells were incubated with DNA at 30 °C overnight and then split into 5 × 1ml aliquots. Cells were centrifuged for 3 min at 5000 rpm, resuspended in sterile PBS/LB (0.1 ml), and plated onto LB containing kanamycin and 0.2 % mannose for 36-48 hrs at 30 °C. Plasmids were cured at 42 °C and the recovered clones confirmed by Sanger sequencing as described above.
Spot dilution assay
Assay was performed as described in (17). Briefly, cultures were streaked onto LB agar plate and incubated at 37 °C for growth. These cultures were then grown in liquid LB broth (5 ml) till OD at 600 nm was 0.4. Then serial dilution was performed for cells in 96 well microtiter plates. 10 μl of serially diluted cells were spotted onto LB agar and allowed to aseptically air dry. These plates were then incubated at either 37 or 30 °C. Plates were imaged after 18 hours.
The sodA-FLAG tagged HBYL1249 cells were grown in LB at 37 °C and 0.5ml of cell culture was harvested at 0.4, 0.6, 0.8 and 1.0 OD 600nm by centrifugation at 13,000 rpm for 5 min. Cell pellets were suspended with 50 μl of PBS and 1.0 μg of lysozyme and incubated at 37 °C for 20 min. This culture lysate was centrifuged at 13,000 rpm for 1 min and soluble fraction was collected, protein concentration was determined by Bradford reagent (BioRad) and separated by SDS-PAGE. Proteins were transferred to PVDF membrane using a Trans-Blot Turbo transfer system (BioRad). The membrane was blocked (for 1 hr) and incubated for 1 hr with primary rabbit anti-FLAG antibody (Sigma-Aldrich). Following buffer wash the membrane was incubated for 1 hr with HRP-conjugated IgG mouse anti-rabbit antibodies (ThermoFisher Sci.). The blot was stained with 1:1 enhancer:substrate reagents (BioRad) for 5 min, visualized with ChemiDoc MP imaging system (BioRad) and ImageLab software.
Candidate overnight grown cells were collected from the plate and mixed in LB to get 0.1 OD. These cells were spotted onto agarose pad and gently covered with a cover-slip and observed using Olympus BX61 epifluorescence (for GFP) microscope and images were captured.
The clustered regularly interspaced short palindromic repeat (CRISPR)-Cas9 system from Streptococcus pyogenes has been widely deployed as a tool for bacterial strain construction. Conventional CRISPR-Cas9 editing strategies require design and molecular cloning of an appropriate guide RNA (gRNA) to target genome cleavage and a repair template for introduction of the desired site-specific genome modification. Here, we present a streamlined method that leverages the existing collection of nearly 4000 Bacillus subtilis strains (the BKE collection) with individual genes replaced by an integrated erythromycin (erm) resistance cassette. A single plasmid (pAJS23) with a gRNA targeted to erm allows cleavage of the genome at any non-essential gene, and at sites nearby to many essential genes. This plasmid can be engineered to include a repair template, or the repair template can be co-transformed with the plasmid as either a PCR product or genomic DNA. We demonstrate the utility of this system for generating gene replacements, site-specific mutations, modification of intergenic regions, and introduction of gene-reporter fusions. In sum, this strategy bypasses the need for gRNA design and allows the facile transfer of mutations and genetic constructions with no requirement for intermediate cloning steps.
Importance Bacillus subtilis is a well-characterized Gram-positive model organism and a popular platform for biotechnology. Although many different CRISPR-based genome editing strategies have been developed for B. subtilis, they generally involve the design and cloning of a specific gRNA and repair template for each application. By targeting the erm resistance cassette with an anti-erm gRNA, genome-editing can be directed to any of nearly 4000 gene disruptants within the existing BKE collection of strains. Repair templates can be engineered as PCR products, or specific alleles and constructions can be transformed as chromosomal DNA, thereby bypassing the need for plasmid construction. The described method is rapid, free of off-target effects, and facilitates a wide-range of genome manipulations.
“Healing is a matter of time, but it is sometimes also a matter of opportunity.” – Hippocrates