Materials and Methods

Function and Application of the CRISPR-Cas System in the Plant PathogenErwinia amylovora

MATERIALS AND METHODSExtraction, alignment, and phylogenies of E. amylovora CRISPR arrays. Genomic sequences for the 127 E. amylovora isolates was accessed through the NCBI nucleotide database as per (7). CRISPR spacers were extracted and aligned from the whole-genome sequences using a pipeline developed in Biopython (this study). CRISPR spacers were obtained through identification of the flanking CRISPR repeats of:CRR1 (5′-GTGTTCCCCGCGTGAGCGGGGATAACCG-3′).CRR2 (5′-GTGTTCCCCGCGTATGCGGGGATAAACCG-3′).CRR4 (5′-GTTCACCTGCCGTACAGGCAGCTTAGAAA-3′).Sequences less than 29 bp, more than 35 bp, or that contained null base calls were excluded due to the high probability of sequencing/assembly errors. All CRISPR spacers were then used to create a consensus alignment for each CRISPR array (CRR1, CRR2, and CRR4). The isolate with the largest CRISPR array, EaOR1, acted as the seed array. The seed array was expanded by sequentially aligning additional CRISPR arrays. Unique spacers were added into the consensus if the flanking spacers were homologous. Otherwise, they were added to the end of the consensus (Fig. 5). After the consensus sequence was completed, the CRISPR arrays were then aligned to the consensus sequence to generate a binary code representing spacers in a array which were congruent with the consensus array. The scripts used to generate the E. amylovora CRISPR consensus sequence is available at Using the binary sequences which represent the presence and absence of spacers compared to the consensus array, a phylogeny was constructed using IQ-TREE (32) using the GTR2 model, with a bootstrap value of 1000, and a bootstrap cut off 70%. The phylogeny was visualized using iTOL (33).Open in a separate windowFIG 5Visual representation of the CRISPR spacer alignment pipeline designed in Biopython. Whole-genome sequences are first parsed to identify CRISPR spacers and form a CRISPR array. The largest CRISPR arrays then acted as a seed, and additional arrays were added to the seed in descending order based on size to form a consensus array. A separate consensus was formed for CRR1, CRR2, and CRR4.Identification of the genetic source of E. amylovora CRISPR spacers. Isolates were clustered based on the CRISPR group identified within the phylogeny and the CRISPR spacers were pooled. All unique spacers were identified within a given group and compared to the NCBI nucleotide database (34) using a cutoff E value of 1e-3 (identity of approximately 26 of the 32 bp of the spacer). The results were parsed to identify matches to known phages, plasmid, and other iMGEs such as transposons. The plasmid maps identifying the genetic position of the protospacers were created using AngularPlamid (35).Design of the artificial CRISPR system and plasmids. Two artificial CRISPR arrays were synthesized that consisted of the endogenous leader sequences for CRR1 and CRR2 followed by 12 CRISPR spacers within the multiple cloning site of pUC18. The leader sequences were predicted to contain promoters using BPROM (36). The spacers were separated using the CRR1 and CRR2 CRISPR repeats into their own respective plasmids (pEA-CRR1Φ and pEA-CRR2Φ). The CRISPR spacers were designed from the genome of four different genera of Erwinia phages (three spacers each): Kolesnikvirus Erwinia virus Ea214 ({"type":"entrez-nucleotide","attrs":{"text":"NC_011811.1","term_id":"219681259","term_text":"NC_011811.1"}}NC_011811.1), Agricanvirus Erwinia virus Ea35-70 ({"type":"entrez-nucleotide","attrs":{"text":"NC_023557.1","term_id":"589889409","term_text":"NC_023557.1"}}NC_023557.1), Johnsonvirus Erwinia virus Ea9-2 ({"type":"entrez-nucleotide","attrs":{"text":"KF806588.1","term_id":"583926751","term_text":"KF806588.1"}}KF806588.1), and Zindervirus Erwinia virus Era103 ({"type":"entrez-nucleotide","attrs":{"text":"NC_009014.1","term_id":"125999960","term_text":"NC_009014.1"}}NC_009014.1). The artificial iMGE (pEA-iMGE) consisted of the pUC57-Kan plasmid with one protospacer from each phage for a total of four and separated with an upstream “AAG” PAM sequence (12). qPCR primer/probe target sequences were incorporated at the beginning of synthesized DNA regions to flank the leader sequences. All sequences were synthesized and inserted into vectors by GenScript (Piscataway, NJ, USA).Quantification of Ea6-4, pEA-iMGE, pEA-CRR1Φ, pEA-CRR2Φ, and ΦEA21-4. Random, optimized primers and probes were designed using the genome of Loxodonta africana for the detection of pEA-CRR1Φ of pEA-CRR2Φ to ensure no cross-reactivity with E. amylovora (Table 3). Previously developed primers and probes were used for Ea6-4 and ΦEa21-4 (30). In this system, a plasmid containing the Ea6-4 and ΦEa21-4 PCR amplicons was diluted to 1011, 108, and 105 copies/mL to create a standard curve for quantification (30). qPCRs contained 2 µL of sample, 200 nM each primer, and 100 nM each probe in EVOlution Probe qPCR mix (Montreal Biotech Inc., Montreal, QC, Canada). Reactions were performed in a qTOWER G3 (Analytik Jena, Jena, Germany) or a Stratagene Mx3005P (Agilent Technologies, Santa Clara, CA, USA) qPCR thermocycler under the following conditions: 15 min at 95°C followed by 40 cycles of 15 s at 95°C and 45 s at 54°C. Prior to the quantification of ΦEa21-4, phage samples were treated with DNase to removed non-encapsidated phage genomes as previously described (31). Briefly, an 8 µL sample of phage was combined with 1 µL of 10x DNase I buffer (B0303D, NEB, Ipswich, MA, USA) and 1 µL DNase I (M0303S, NEB, Ipswich, MA, USA) in a 96-well plate. The samples were then incubated for 40 min at 37°C, followed by 20 min at 95°C, and a hold at 4°C. Phage were also quantified through plaque assays using a soft agar overlay (27, 37).TABLE 3Primers and probes used for molecular quantification in this studyTargetNameSequence Erwinia amylovora Ea-Lsc-FCGC TAA CAG CAG ATC GCAEa-Lsc-RAAA TAC GCG CAC GAC CATEa-Lsc-P/5Cy5/CTG ATA ATC CGC AAT TCC AGG ATG/3IAbRQsp/ΦEa21-4END37-FTTC AGC TTT AGC GGC TTC GAG AEND37-RAGC AAG CCC TTG AGG TAA TGG AEND37-P/56-ROXN/AGT CGG TAC ACC TGC AAC GTC AAG AT/3IAbRQSp/pEA-CRR1Φ & pEA-CRR2ΦpEA-CRR-FCTG GTC AGC ATC ACT AGC ATA ApEA-CRR-RACC TCG AAG AAG GCG GAT AGpEA-CRR-P/5Cy5/TTT CTG CGC/TAO/GTA ATC TGC TGC TTG/3IAbRQSp/pEA-iMGEpEA-iMGE-FCAT CAC TGG CCT CCT ACT TTA CpEA-iMGE-RCCA AGG CAC CTC ACA TAC TTpEA-iMGE-P/56-FAM/TCC ACT ACG/ZEN/GCC ATC TGT TTC ACG/3IABkFQ/Open in a separate windowTransformation of E. amylovora. CFU and qPCR copy numbers have been previously correlated to allow CFU/mL to be determined using qPCR for E. amylovora (30). A standard curve for OD600 based on qPCR quantification was created using Ea2-95, Ea6-96, and Ea3-97 during exponential growth. A total of 61 measurements were taken at a range of 107 to 1010 on a Thermo Spectronic Genesys 20 (ThermoFisher Scientific, Waltham, MA, USA) then quantified through qPCR. From this standard curve, the equation y = 1010 · ×1.4518 (R2 = 0.9116) was derived where x is the OD600 measurement and y is the CFU/mL as determined through qPCR (data not shown).Isolates were plated on Difco nutrient agar (NA) (BD, Sparks, MD, USA) from frozen cultures stored on microbeads (MicrobankTM, ProLab Diagnostics, Richmond Hill, ON, Canada), incubated at 27°C overnight, and stored at 4°C. A culture of E. amylovora was grown in Difco nutrient broth (BD, Sparks, MD, USA) amended with 0.5% sucrose (NBS), and 100 ppm of kanamycin if required, to 108 CFU/mL in a programmable, Innova 44 shaking incubator (New Brunswick Scientific, Edison, NJ, USA) at 27°C (160 rpm). Following incubation, the bacterial suspension was centrifuged at 12 000 × g (4°C) for 8 min and the supernatant was discarded. The bacterial pellet was washed in 40 mL of iced 10% glycerol and centrifuged at 12 000 × g (4°C) for 12 min twice. The pellet was resuspended in 10% glycerol and adjusted to 2 × 109 CFU/mL. Each transformation consisted of 400 µL of bacterial suspension and 50 ng of plasmid DNA. Electroporation occurred in 2 mm electroporation cuvettes using a Bio-Rad Gene Pulser Electroporator (Bio-Rad Laboratories, Hercules, CA, USA) with the following settings: 800 Ω, 25 µF, and 2.5 kV for 4s (38). Transformants were immediately diluted with 600 µL of SOC media (39) and incubated at 27°C for 1 h. A 100 µL aliquot of the transformed bacteria was plated on NAS amended with ampicillin, kanamycin or both ampicillin and kanamycin. All antibiotics were applied at 100 ppm.CRISPR-Cas mediated interference against plasmids. Isolates were first transformed with piMGE and were secondarily transformed with pUC19 (control), pEA-CRR1Φ, or pEA-CRR2Φ to test if the introduction of CRISPR spacers homologous to pEA-iMGE resulted in curing of this plasmid. Secondary transformation reactions were plated on NAS amended with ampicillin, kanamycin, or both antibiotics. Efficiency of CRISPR-mediated curing was estimated by the number of colonies on NASKana+Amp relative to number of colonies on NASAmp. Transformations were enumerated after 30 h. Transformations were considered to be valid only if growth was observed on NASAmp. Experiments were performed in triplicate. ORFs of pEU30 were identified using ORFfinder and compared to the proteins of the AntiCRISPR Database (AcrDB) (25, 40).Propagation and infection using phage ΦEA21-4. Phage ΦEA21-4 was propagated as previously described with minor amendments (30). Briefly, 100 mL of 108 CFU/mL of the Ea6-4 was prepared in NB and grown at 27°C (160 rpm) in a Innova 44 shaking incubator. After 1 h, 108 PFU of ΦEA21-4 was added to the bacterial culture. The culture was incubated at the conditions listed above overnight. Following incubation, the culture was treated with 2 mL chloroform, centrifuged at 12 000 × g (4°C) for 8 min, and passed through a 0.22 µm filter under vacuum (Millipore, Billerica, MA, USA). Phage cultures were stored with 1 mL of chloroform in amber vials (Wheaton Industries, Millville, NJ, USA) at 4°C.To determine the effect of the CRISPR-Cas system against ΦEA21-4, cultures of Ea6-4 which had been transformed using plasmid pUC19, pEA-CRR1Φ, or pEA-CRR2Φ were grown in NBS amended with 100 ppm ampicillin to 108 CFU/mL as described above. Cultures for OD600 sampling were created by diluting the bacterial cultures to 107 CFU/mL in 50 mL falcon tubes with sterile sponge stoppers for aeration. Phage stocks were then diluted and added to the culture at the designated MOI for a total volume of 25 mL. At the same time, a set of paired samples was created at a volume of 150 µL in 96-well plates to quantify phage ΦEA21-4. The infected cultures in 50 mL falcon tubes and sealed 96-well plates were incubated at 27°C (150 rpm) for 8 h. A 1 mL sample was taken from the 25 mL cultures for OD600 measurements following phage infection, and every hour thereafter. Fifty µL of chloroform was added to each culture in the paired 96-well plate associated with each time point to kill the culture. Experiments were performed in triplicate. Growth rates were determined using an exponential line of best fit from the equation y = a·ekt. The phages were then quantified using the protocol qPCR previously described. After 8 h of incubation, the bacterial genomic DNA was extracted from the cultures infected at an MOI of 100 using the Bacterial Genomic DNA isolation kit (17900, Norgen Biotek Corp., St. Catharines, ON, Canada) as per the manufacturer’s instructions.Sequencing of the Ea6-4 transformants infected by phage ΦEA21-4. The extracted genomic DNA was sequenced using the Nanopore MinIon platform (Oxford Nanopore Technologies, Oxford, UK). Samples were prepared using the manufacture’s instructions for the Rapid DNA Sequencing kit (SQK-RAD004, Oxford Nanopore Technologies, Oxford, UK) on a Spoton flow cell (Oxford Nanopore Technologies, Oxford, UK). Sequencing data were acquired using MinKnow and the genomes were assembled using Flye with 4 polishing steps (41). Coverage of each assembly was at least 50x. The chromosomal and plasmid sequences were assessed for the insertion of new CRISPR spacers using the CRISPR aligner pipeline. The sequencing data were also parsed to identify individual reads which contained any CRISPR repeats. The reads were then cross-referenced to the phage ΦEA21-4 genome using blastn to determine if any novel spacers had been acquired which didn’t appear in the genomic assemblies (42).

Article TitleFunction and Application of the CRISPR-Cas System in the Plant PathogenErwinia amylovora


In conclusion, the CRISPR-Cas system ofE. amylovorais far more diverse and complex than previous analyses suggested. The phylogenies produced using the CRISPR-Cas system resolve the same clades previously observed inE. amylovorabut show a strong connection between the ENA and WNA clades (7,8). The annotation of the spacers in this work showed that theAmydaloideae-infecting strains ofE. amylovoraare more frequently pressured by plasmids than phages, while theRubus-infecting strains appear to be equally pressured by both. The CRISPR-Cas system is active in the WP and WNA clades in the absence of pEU30, while activity in the ENA clade was not observed. While CRR1 spacers do provide some degree of immediate protection to phage infection, no spacer acquisition to ΦEa21-4 was detected. Interestingly, the control strain of Ea6-4 containing pUC19 was able to survive phage infection using an unidentified system which was complementary to the CRISPR-Cas system. Overall, this shows that while the CRISPR-Cas system is potentially important as a defense mechanism for plasmids, it is not the primary mechanism for phage resistance inAmydaloideae-infecting strains ofE. amylovora.

Login or Signup to leave a comment
Find your community. Ask questions. Science is better when we troubleshoot together.
Find your community. Ask questions. Science is better when we troubleshoot together.

Have a question?

Contact or check out our support page.