Engineered Reporter Phages for Rapid Bioluminescence-Based Detection and Differentiation of Viable Listeria Cells

Abstract

The pathogen Listeria monocytogenes causes listeriosis, a severe foodborne disease associated with high mortality. Rapid and sensitive methods are required for specific detection of this pathogen during food production. Bioluminescence-based reporter bacteriophages are genetically engineered viruses that infect their host cells with high specificity and transduce a heterologous luciferase gene whose activity can be detected with high sensitivity to indicate the presence of viable target cells. Here, we use synthetic biology for de novo genome assembly and activation as well as CRISPR-Cas-assisted phage engineering to construct a set of reporter phages for the detection and differentiation of viable Listeria cells. Based on a single phage backbone, we compare the performance of four reporter phages that encode different crustacean, cnidarian, and bacterial luciferases. From this panel of reporter proteins, nanoluciferase (NLuc) was identified as a superior enzyme and was subsequently introduced into the genomes of a broad host range phage (A511) and two serovar 1/2- and serovar 4b/6a-specific Listeria phages (A006 and A500, respectively). The broad-range NLuc-based phage A511::nluc_CPS detects one CFU of L. monocytogenes in 25 g of artificially contaminated milk, cold cuts, and lettuce within less than 24 h. In addition, this reporter phage successfully detected Listeria spp. in potentially contaminated natural food samples without producing false-positive or false-negative results. Finally, A006::nluc and A500::nluc enable serovar-specific Listeria diagnostics. In conclusion, these NLuc-based reporter phages enable rapid, ultrasensitive detection and differentiation of viable Listeria cells using a simple protocol that is 72 h faster than culture-dependent approaches.IMPORTANCE Culture-dependent methods are the gold standard for sensitive and specific detection of pathogenic bacteria within the food production chain. In contrast to molecular approaches, these methods detect viable cells, which is a key advantage for foods generated from heat-inactivated source material. However, culture-based diagnostics are typically much slower than molecular or proteomic strategies. Reporter phage assays combine the best of both worlds and allow for near online assessment of microbial safety because phage replication is extremely fast, highly target specific, and restricted to metabolically active host cells. In addition, reporter phage assays are inexpensive and do not require highly trained personnel, facilitating their on-site implementation. The reporter phages presented in this study not only allow for rapid detection but also enable an early estimation of the potential virulence of Listeria isolates from food production and processing sites.

Keywords: CRISPR-Cas; Listeria monocytogenes; bacteriophage; bioluminescence; food safety; phage engineering; reporter phage; synthetic biology.

FIG 1

NLuc is a superior enzyme for the construction of synthetic and highly efficient Listeria-specific reporter bacteriophages. (A) Schematic representation of the synthetic phage engineering approach used to construct A500-derived reporter phages. (B) Comparison of the genetic location and size of the different luciferase genes integrated into the genome of A500 ΔLCR. (C) Bioluminescence time course assays of WSLC1042 cells infected with the indicated reporter phage at 30°C (unless indicated otherwise). (D) Reporter phage sensitivities were determined by infecting serial host cell dilutions and quantifying luminescence at 3 h p.i. Detection limits were calculated as the cell number required to produce a signal that is 3-fold above the standard deviation of the background luminescence (indicated as vertical dotted lines). All values are fold changes calculated by dividing the signal produced by A500::luciferase ΔLCR divided by the signal produced from infections with the parental phage lacking the luciferase gene (A500 ΔLCR). Ply, phage lysin; RLU, relative light unit. Data are mean ± standard error of the mean (SEM) from biological triplicates.

FIG 2

CRISPR-Cas9-aided construction of broad host range reporter phages A511::nluc_PLY and A511::nluc_CPS. (A) Schematic representation of the two-step, CRISPR-Cas-assisted phage engineering approach using a programmable type II-A CRISPR-Cas system from L. ivanovii. Homologous recombination between the incoming A511 genome and an editing plasmid (pEdit) carrying the nluc sequence and flanking homology arms occurs at low frequency. The homology arms contain a protospacer sequence, which can be targeted by Cas9. However, a silent point mutation is introduced into the PAM motif of the pEdit homology arms, which allows for subsequent CRISPR-mediated, sequence-specific cleavage of nonmodified phage DNA, while recombinant genomes are protected and the corresponding phages enriched. (B) Schematic representation of the editing plasmids used to incorporate mutated PAMs and nluc gene sequences either downstream of the A511 major capsid protein (cps) or downstream of the A511 endolysin gene (ply). (C) Counterselection of wild-type phages. Serial dilutions of phages harvested after propagation on strains containing the indicated editing plasmids were spotted on L. ivanovii 3009 p_help cas9 containing the indicated CRISPR targeting plasmids (crRNA encoding vector pLRSR). A511::nluc candidate phages that escape CRISPR-Cas interference are shown in red boxes. (D) Schematic representation of the resulting reporter phages A511::nluc_PLY and A511::nluc_CPS, including PCR-mediated genotype validation and assessment of plaque morphology. PAM, protospacer-adjacent motif; pLRSR, crRNA expression plasmids; p_help, strong, constitutive Listeria promoter; pEdit empty, empty vector control without nluc or flanking homology arms; WT, wild type.

FIG 3

Ultrasensitive, A511-based NLuc reporter phages enable single-cell detection. The NLuc reporter phages A511::nluc_PLY and A511::nluc_CPS were benchmarked against the previously published phage A511::luxAB. All reporter phages featured comparable infection kinetics as determined by bioluminescence time course assays (A). However, dose-response curves demonstrate superior sensitivity of the NLuc-encoding phages, both of which are able to detect a single cell in this assay (B). Detection limits were calculated as the cell number required to produce a signal that is 3-fold above the standard deviation of the background luminescence (indicated as vertical dotted lines). All values are fold changes calculated by dividing the signal produced by A511::luciferase by the signal produced from infections with the parental wild-type phage. Data are mean ± SEM from biological triplicates.

FIG 4

Serovar-specific reporter phages reliably differentiate Listeria strains. (A) Schematic representation of the luciferase gene insertion sites in the genomes of SV 1/2-specific phage A006::nluc ΔLCR, SV 4b/6a-specific phage A500::nluc ΔLCR, and broad host range phage A511::nluc_CPS. (B) Time course luminescence assays were used to compare infection kinetics of phage A006::nluc ΔLCR (on EGDe) to those of phages A500::nluc ΔLCR (on WSLC1042) and A511::nluc_CPS (on WSLC3009). Three hours postinfection was selected as an optimal endpoint for bioluminescence assays with all reporter phages (vertical dotted line). Data are mean ± SEM from biological triplicates. (C) Reporter phage-mediated serovar differentiation of 54 Listeria strains covering the major SVs. Indicated strains were infected with reporter phages, and transduced bioluminescence was quantified at 3 h postinfection. Brothothrix thermosphacta, Staphylococcus aureus, and Enterococcus faecalis served as genus specificity controls. The heat map shows fold changes calculated by dividing the background-corrected signal produced by reporter phage infections by the signal produced from infections with the parental phage lacking the luciferase. ATCC, American Type Culture Collection; WSLC, Listeria strains from the Weihenstephan Microbial Strain Collection.

FIG 5

Bioluminescence-based detection of Listeria in artificially contaminated foods. (A) Schematic representation of the workflow and duration of the reporter phage-based detection assay compared to classical culture-dependent detection. (1) Preparation of the food sample; (2) Listeria-selective enrichment; (3) phage infection; (4) substrate addition; (5) detection of bioluminescence. (B to D) Bioluminescence output (RLUs) of artificially spiked food samples (samples 1 to 11) and unspiked controls is shown after selective enrichment in 1/4 LEB (16, 18, and 20 h) and compared to results from a culture-based detection approach performed with the same samples: +, culture positive for Listeria; −, culture negative for Listeria. Samples were contaminated with ± 1 CFU/25 g of food. Quantified values are 1.2 ± 0.2 in milk, 1.1 ± 0.04 in chicken cold cut (errors are SEM of biological duplicates), and 1.0 ± 0.2 in iceberg lettuce (error is standard deviation of technical triplicates). Dotted lines indicate threshold value.