REBOOTING OF SYNTHETIC BACTERIOPHAGE GENOME IN L-FORM BACTERIA

Abstract

The present invention related to a method for producing or propagating an engineered bacteriophage, comprising the steps of: providing a functional synthetic genome of an engineered bacteriophage being able to infect a target bacterium; providing a recipient bacterium; transforming the recipient bacterium with the functional synthetic genome in a transformation step, yielding a transformed recipient bacterium; incubating the transformed recipient bacterium in a first incubation step, wherein the engineered bacteriophage is propagated within the transformed recipient bacterium; and further incubating the transformed recipient bacterium or the propagated engineered bacteriophage released from the transformed recipient bacterium with the target bacterium in a second incubation step, wherein the propagated engineered bacteriophage infects the target bacterium and is further propagated within the target bacterium, and wherein the recipient bacterium is a cell wall-deficient bacterium.

Claims

1. Method for producing or propagating an engineered bacteriophage, comprising the steps of: providing a functional synthetic genome of an engineered bacteriophage being able to infect a target bacterium, providing a recipient bacterium, transforming said recipient bacterium with said functional synthetic genome in a transformation step, yielding a transformed recipient bacterium, incubating said transformed recipient bacterium in a first incubation step, wherein said engineered bacteriophage is propagated within said transformed recipient bacterium, and further incubating said transformed recipient bacterium or said propagated engineered bacteriophage released from said transformed recipient bacterium with a target bacterium in a second incubation step, wherein said propagated engineered bacteriophage infects said target bacterium and is further propagated within said target bacterium, wherein said recipient bacterium is a cell wall-deficient bacterium.

2. The method according to claim 1, wherein said recipient bacterium is a cell wall-deficient variant of or derived from a Gram-positive bacterium, particularly a cell wall-deficient variant of or derived from a bacterium of the genus Listeria, particularly Listeria monocytogenes.

3. The method according to claim 1, wherein said target bacterium is a Gram-positive bacterium, particularly selected from the genus of Listeria, Bacillus, Enterococcus, Streptococcus, Clostridium or Staphylococcus, more particular Listeria monocytogenes, Listeria ivanovii, Listeria innocua, Bacillus subtilis, Bacillus cereus, Bacillus thuringiensis or Staphylococcus aureus.

4. The method according to claim 1, wherein said recipient bacterium and said target bacterium belong to or are derived from different species or members of different genera.

5. The method according to claim 1, wherein said functional synthetic genome is provided by in vitro or in vivo assembly of fragments thereof or by de novo synthesis.

6. The method according to claim 1, wherein said functional synthetic genome has a length of at least 10,000 base pair, particularly at least 30,000 base pairs, more particularly at least 40,000 base pair.

7. The method according to claim 1, wherein said transformation step is conducted in presence of a polyethylene glycol, particular with a mean molecular weight in the range of 1,000 g*mol.sup.1 to 30,000 g*mol.sup.1, more particular in the range of 7,000 g*mol.sup.1 to 20,000 g*mol.sup.1, particularly PEG-8000.

8. The method according to claim 1, wherein said first incubation step is conducted over a period in the range of 4 h to 9 h, particularly in the range of 24 h to 32 h.

9. The method according to claim 1, wherein said recipient bacterium is provided by incubating a cell walled precursor bacterium in presence of a cell wall synthesis interfering antibiotic, particularly selected from beta lactam antibiotics, glycopeptide antibiotics, cycloserine, or fosfomycin, and optionally a osmoprotective medium, yielding said recipient bacterium.

10. A method for manufacturing a cell wall-deficient bacterium of the genus Listeria, comprising the steps of providing a bacterium of the genus Listeria characterized by the genotype [lmo0584 lmo1653-54 lmo01861] or by a genome not comprising functional homologues of lmo0584, lmo1653-54 and lmo01861, particularly Listeria monocytogenes, more particularly Listeria monocytogenes strain EGD-e, or providing a bacterium of the species Listeria innocua, particularly a bacterium of the strain 2021 of Listeria innocua; and cultivating said bacterium in presence of a cell wall synthesis interfering antibiotic and optionally in presence of an osmoprotective medium.

11. The method according to claim 10, wherein said cell wall synthesis interfering antibiotic is selected from the group comprising beta lactam antibiotics, glycopeptide antibiotics, cycloserine, or fosfomycin, particularly penicillin G.

12. A cell wall-deficient bacterium of the genus Listeria characterized by the genotype [lmo0584 lmo1653-54 lm01861] or by a genome not comprising functional homologues of lmo0584, lmo1653-54 and lmo01861.

13. The cell wall-deficient bacterium according to claim 12, wherein said bacterium is Listeria monocytogenes, particularly Listeria monocytogenes strain EGD-e.

14. The cell wall-deficient bacterium, according to claim 12, obtainable by a method according to claim 10.

15. Method for producing or propagating a bacteriophage, comprising the steps of: providing a functional genome of an bacteriophage being able to infect a target bacterium, providing a recipient bacterium, transforming said recipient bacterium with said functional genome in a transformation step, yielding a transformed recipient bacterium, incubating said transformed recipient bacterium in a first incubation step, wherein said bacteriophage is propagated within said transformed recipient bacterium, and further incubating said transformed recipient bacterium with said target bacterium in a second incubation step, wherein said propagated bacteriophage infects said target bacterium and is further propagated within said target bacterium, wherein said recipient bacterium is a cell wall-deficient bacterium, and said recipient bacterium and said target bacterium belong to or are derived from different species or members of different genera.

Description

DESCRIPTION OF THE FIGURES

[0066] FIG. 1 shows the rebooting of Listeria bacteriophage genomes in L-form strain Rev2L. (A-B) The ability to reboot Listeria phages in L-form strain Rev2L was assessed as depicted in (A) using genomic DNA of Listeria phage P35. L-form transformation reactions were prepared as indicated in (B), incubated at 32 C. and tested for plaque formation on the indicator strain at 24 h post transformation. DNAsel indicates a 30 min pre-digestion of P35 gDNA using DNAsel. (C) Efficiency of transformation and rebooting was determined using a dilution series of P35 gDNA. (D) A set of seven additional Listeria phages was rebooted in Rev2L using 500-1000 ng gDNA and detected using Listeria monocytogenes or Listeria ivanovii as indicator strains. Phage properties are listed in tabular form. (E) Bacteriophage production kinetics in L-forms were assessed for three phages over a period of 96 h. Data are meanSD (n=3). =phage, PEG=polyethylene glycol, gDNA=genomic DNA, tr=terminal redundancy, cp=circular permutation, cos=cohesive end site, nd=not determined.

[0067] FIG. 2 shows the rebooting of (A) Bacillus and (B) Staphylococcus phage genomes in Listeria L-forms. Rev2L cells were transformed with 1-5 g of phage gDNA, incubated for 24 h at 32 C. and assayed for phage production on their respective indicator strains. Transformation reactions lacking either the gDNA or the L-form cells served as controls. =phage.

[0068] FIG. 3 shows the rebooting of synthetic, in vitro assembled bacteriophage genomes in Listeria L-forms. The general workflow for rebooting of synthetic genomes in Rev2L is depicted in (A). Genomic phage DNA was extracted and purified from Listeria monocytogenes phage P35, Listeria innocua phage B025, and from Bacillus cereus phage TP21-L. Overlapping PCR-fragments covering the full genome were generated using a high-fidelity polymerase (B, D, F) and assembled in-vitro to generate circular molecules unless indicated otherwise. Assembly reactions were transformed and rebooted in Rev2L cells and plated on the respective indicator strains using incomplete assemblies as controls (C, E, G). For phage TP21-L, rebooting reactions were either assayed for phage production after 24 h using indicator strain or, where indicated (amplification), 10 l of stationary HER1399 culture was added to the rebooting reaction at 6 h post transformation and assayed for phage production at 24 h (G).

[0069] FIG. 4 shows life-style conversion of temperate Listeria phage B025 from temperate to virulent. An in-vitro genome assembly strategy was used to produce mutants of the temperate Listeria phage B025 lacking either the repressor of integration (rep) or the complete lysogeny control region (LCR). A workflow is shown in (A). Fragment three of the wild-type assembly contains the lysogeny control genes and was split into two overlapping fragments omitting either the repressor only or the complete LCR to yield five PCR fragments for genome assembly (B). Recombinant genomes were rebooted, the resulting phage mutants purified and assayed for correct genotype using PCR and sequencing (C). L. ivanovii WSLC3009 was infected with increasing numbers of wild-type, rep and LCR phages using the soft-agar overlay technique. Surviving bacterial lawn (B025 wt) or single colonies (B025 rep and B025 LCR) were assayed for the presence or absence of B025 prophage using the indicated primer pairs (E). LCR=lysogeny control region, PFU=plaque forming unit, attB=bacterial attachment site for prophage integration, attL=left prophage flanking region in WSLC3009::B025.

[0070] FIG. 5 shows also the rebooting of Listeria bacteriophage genomes in L-form strain Rev2L. A detailed workflow for rebooting of phage genomes in which each parameter was optimized individually is shown in (A). Optimized parameters include: Time of Rev2L growth before transformation (B), adjusted optical density (600 nm) of L-form culture at the time of transformation (C), final PEG concentration before addition of DM3 medium (D), volume of DM3 added to the DNA/L-form/PEG-mixture before incubation (E), and the average molecular weight of the PEG-chains used (F). The selected optimal conditions are indicated (asterisk). The quality of the bacteriophage DNA used for rebooting was assessed using pulsed-field gel electrophoresis (G). Using a dilution series of phage DNA, rebooting efficiency of phages P35 and A511 were compared (H). The ability of various Listeria phages to infect walled Rev2 cells was assessed by spotting phage dilutions on soft-agar overlays containing Rev2 or the phage indicator strain. (I) PenG=penicillin G, OD=optical density, PEG=polyethylene glycol. Data are meanSD (n=3).

[0071] FIG. 6 shows the rebooting of synthetic, in vitro assembled bacteriophage genomes in Listeria L-forms. The efficiency of genome rebooting using genomic DNA and in vitro assembled genomes as input material was compared: Four-fold dilution series of P35 gDNA or Gibson-assembly reaction were used for rebooting in Rev2L starting with 2 g and 125 ng input material for gDNA and assembly reaction, respectively. Total pfu per transformation reaction were quantified and compared. Data is meanSD (n=3).

[0072] FIG. 7 shows that deletion of TP21-L lysogeny control genes. An in vitro genome assembly strategy was used to produce mutants of the temperate Bacillus phage TP21-L lacking either the repressor of integration (rep) or the complete lysogeny control region (LCR). A workflow is shown in (A). Fragment three of the wild-type assembly contains the lysogeny control genes and was split into two overlapping fragments omitting either the repressor only or the complete LCR to yield five PCR fragments for genome assembly. Recombinant genomes were rebooted, the resulting phage mutants purified and assayed for correct genotype using PCR and sequencing (B). LCR=lysogeny control region.

[0073] FIG. 8 shows shows the rebooting of Listeria bacteriophage genomes in L-form strain Rev2 (L. monocytogenes) and in a reversible L-form that was obtained from non-pathogenic Listeria innocua 2021 (SLCC 5639). Each of the two L-form cultures was transformed with genomic DNA (1 g) of the L. monocytogenes bacteriophage P70, incubated at 32 C. for 24 h, and tested for plaque formation using L. monocytogenes L99 as an indicator strain. Plaques on soft-agar overlays (24 h post infection) are shown.

EXAMPLES

[0074] The present invention shows that a novel Listeria monocytogenes L-form strain Rev2L can be transformed with intact, purified bacteriophage DNA which leads to genome rebooting, i.e. the production of infectious virions from naked DNA. Rebooting was applicable to a heterogeneous group of Listeria monocytogenes and Listeria innocua phages independent of phage tropism, morphology, genome size or genome structure. Remarkably, this genome rebooting approach was also successful for several phages infecting Bacillus and Staphylococcus, effectively bypassing the genus barrier of infection. To generate both wild-type and recombinant viruses in a synthetic biology approach, bacteriophage genomes were assembled in-vitro using amplified overlapping fragments and subsequently rebooted in Rev2L cells. As proof of concept for this strategy, the inventors deleted lysogeny control genes of the temperate Listeria phage B025 and demonstrated an acquired virulent phenotype of the recombinant viruses. Combining in-vitro genome assembly and L-form transformation to engineer bacteriophages will facilitate the development of tailored phages both for basic and applied research on phages infecting Gram-positive pathogens.

[0075] Particularly, the below two-step protocol may be used for the generation of genetically-modified bacteriophages: [0076] In a first step, synthetic bacteriophage genomes are assembled in-vitro using a method based on Gibson-assembly, which is commercially available. Bacteriophage genomes are assembled as circular, double-stranded molecules mimicking DNA intermediates that occur during natural phage replication. In principle, any method that leads to the formation of a synthetic genome may be used for this first step and it is thus not limited to Gibson-assembled genomes. [0077] In a second step, in-vitro assembled viral genomes/synthetic genomes are transformed into Listeria monocytogenes cells where these synthetic phages are rebooted to produce infectious particles. For this transformation reaction, the inventors do not use wild-type Listeria cells but instead they relied on a newly identified Listeria monocytogenes L-form strain Rev2L. This strain has the ability to grow and divide in the absence of an intact cell-wall as long as an osmotically stabilized medium is used to prevent lysis of the bacterium. Due to the lack of an intact cell-wall, Rev2L can be transformed with large DNA-molecules such as intact bacteriophage genomes. For this transformation reaction, the inventors used an optimized, polyethylene glycol (PEG8000) based protocol. Transformed L-form bacteria were incubated for 24 hours to allow for phage rebooting. To isolate and amplify synthetic phages from these reactions, L-forms were subsequently mixed and incubated with the natural propagation strain of the phage of interest. It is not necessary to actively release progeny phage from the L-form cells because they are either lysed due to the activity of phage-encoded enzymes or simply by osmotic destabilization of the membrane.

[0078] Importantly, this method is not restricted to Listeria phages, but can be applied to other phages of Gram-positive bacteria or other bacteria.

DETAILED DESCRIPTION OF THE PROOF OF CONCEPT

[0079] To address phage genome rebooting in Gram-positive cells, the inventors used the virulent Siphovirus P35 that infects the food-borne pathogen Listeria monocytogenes. The 35822 bp linear genomic DNA (gDNA) of P35 is too large for electroporation into walled Listeria cells which usually take up supercoiled plasmids of up to 10 kb at very low efficiencies. Therefore, the inventors were unable to address genome-rebooting using a conventional electroporation protocol with walled cells. Based on the assumption that the peptidoglycan envelope is the major barrier to transformation of larger DNAs, they investigated the possibility of using metabolically active, yet cell wall-deficient Listeria L-forms as recipients of purified, linear P35 gDNA. L-forms are induced by prolonged, repeated subcultivation in the presence of cell wall synthesis targeting antibiotics in an osmoprotective medium. For the method of the invention, the inventors used a novel Listeria strain Rev2 [lmo0584 lmo1653-54 lm01861] that was obtained by long-term exposure of L. monocytogenes EGD-e to penicillinG (penG) in DM3 medium and has the ability to switch between growth as an L-form (designated Rev2L) and a walled bacterium (Rev2).

[0080] The inventors devised a work-flow for polyethylene-glycol (PEG)-mediated transformation of L-form bacteria with phage gDNA (FIG. 1A and FIG. 5A) and found that P35 is rebooted in a gDNA-, L-form-, and PEG-dependent process (FIG. 1B). For this purpose, purified P35 gDNA was mixed with a growing penicillinG (penG)-induced Rev2L culture and PEG-8000 solution. PEG was subsequently diluted with osmotically stabilized DM3 medium and the mixture incubated for 24 h at 32 C. to allow for phage rebooting. The L-form transformation reaction was assayed for the presence of rebooted phage using the propagation strain of phage P35 as an indicator strain (L. monocytogenes Mack).

[0081] In order to get the highest phage yield, the inventors optimized each step of the rebooting protocol using P35 gDNA (see FIG. 5 and Methods section for details). Under such optimized conditions, the inventors found a linear correlation between input DNA and phage production (FIG. 10) with a detection limit of 2.6 pg of P35 gDNA corresponding to about 66000 genomes (assuming that the used DNA consisted of intact gDNA only). Besides P35, this L-form transformation protocol allowed for rebooting of a heterogeneous group of seven additional Listeria phages (FIG. 1D). These include phages with virulent (P70, P100, and A511) and temperate (B025, B035, B056, PSA) life-styles, some phages with large genomes of more than 130 kb (P100, A511), and morphologically diverse phages with either contractile tails (Myoviruses; P100, A511) or non-contractile tails (Siphoviruses; all other). Phage B025 has cohesive overlapping genome ends (cos), whereas the other rebooted phages have terminally redundant genomes, with or without circular permutation.

[0082] These results strongly suggest that rebooting is independent of virion morphology, phage genome size and genome replication strategy. Moreover, the inventors were able to produce Listeria phages that would not normally infect a Rev2 cell (B025, B035, B056, and PSA, FIG. 5) suggesting that receptor binding and/or genome translocation are the only barriers restricting the production of these phages during infection of walled cells. The diverse features of the rebooted Listeria phages are summarized in FIG. 1D (small table) and show that rebooting in Rev2L is broadly applicable to Listeria phages.

[0083] The inventors compared the rebooting kinetics of phages P35, P70, and A511 (FIG. 1E) and found that production of all three phages peaks at 24 h post transformation which is slow compared to infection kinetics in walled cells. For comparison, A511 has a burst time of 60 min when walled cells are infected. This large time difference is likely explained by slow metabolism in L-forms. Because L-forms are devoid of an intact cell wall, progeny virions are released by lysis of Rev2L cells either through the action of phage holin proteins or by osmotic destabilization when DM3 medium is diluted in osmotically non-stabilized soft-agar. This will also allow for the release of phages whose cell-wall lytic enzymes would not normally degrade Listeria serovar 1 cell walls. Released phages cannot bind to neighboring cells and their survival depends on stability in DM3 medium (FIG. 1E).

[0084] In addition to rebooting of phages that infect Listeria, the inventors demonstrate that Rev2L cells can be used to reboot phages that infect different genera of Gram-positive organisms within the Firmicutes phylum. Phylogenetically, Bacillus is most closely related to Listeria, and the inventors found that the small, 37.46 kb Siphovirus TP21-L of Bacillus cereus as well as the large, 153.96 kb Myovirus Bastille of Bacillus thuringiensis could be rebooted in Rev2L using genomic DNA as a substrate (FIG. 2A).

[0085] Next, the inventors found that Listeria L-forms could also be used as a rebooting platform for phages infecting the human pathogen Staphylococcus aureus: They successfully rebooted 2638A, a 41.32 kb Siphovirus as well as phage K, a large 127.40 kb Myovirus (FIG. 2B). The results so far demonstrate that Rev2L is a highly versatile host for the uptake of large viral genomes and the cross-genus rebooting of phages of Gram-positive organism from naked, linear gDNA.

[0086] Next, the inventors utilized the L-form platform of the invention to address whether synthetic genomes could also be used as a substrate for rebooting (workflow is shown in FIG. 3A). First, they amplified and purified overlapping segments of P35 DNA and assembled synthetic genomes using the Gibson method. Because most phages use circular replication-intermediates, the overlaps between fragments were designed to allow for end-joining (circular closure). Nevertheless, the formation of concatemeric or terminally redundant linear DNA cannot be excluded. To assemble the P35 genome, the inventors used either six fragments of about 6 kb or three fragments of about 12 kb (FIG. 3B) and it was found that P35 was efficiently rebooted from synthetic DNA (FIG. 3C). As control, they used an incomplete assembly lacking one fragment. For P35, using synthetic DNA was even more efficient compared to purified gDNA (detection limit=1.1 pg DNA, S FIG. 6), either because ring-closure is no longer required or because circular DNA is transformed more efficiently than a linear molecule.

[0087] Besides P35, the inventors also used this synthetic approach to assemble and reboot the genomes of the temperate L. innocua phage B025 (FIG. 3D-E). In-vitro assembly and cross-genus rebooting of B. cereus phage TP21-L (FIG. 3F-G) was also successful in Rev2L and it was found that the production of phage was amplified when host cells were added to the rebooting reaction at 6 h post transformation (FIG. 3G). This amplification strategy was not successful for any phages infecting Listeria or Staphylococcus because they fail to infect in DM3 medium. Also TP21-L was used to show that synthetic phages are produced even in the absence of a designed ring-closure (FIG. 3G).

[0088] The unique combination of synthetic genome assembly and rebooting in L-forms offers a platform for phage genome engineering: To explore this approach, the inventors attempted to switch the life style of a temperate Listeria phage from lysogenic to lytic using a synthetic, modified genome. To this end, they assembled the genome of temperate Listeria phage B025 but omitted genes that control and mediate prophage integration. In B025, these genes are encoded on a 2.7 kb putative lysogeny control region (LCR), and either the repressor of the lytic cycle (B025 rep) or the whole LCR (B025 LCR) including the phage integrase were deleted. The genome assembly strategy is depicted in FIG. 4A and fragments for the assembly of phage mutants are shown in FIG. 4B. Recombinant phage genomes were assembled, rebooted successfully, and the resulting phages were validated by PCR and sequencing of the LCR region (FIG. 4C). Although there were no obvious differences in plaque morphology (FIG. 4C), the recombinants could no longer integrate and replicated as synthetic virulent phages (FIG. 4D): When rebooted phage from a B025 wild-type assembly was used, prophage integration effectively prevented killing of the host at high multiplicities of infection (FIG. 4D; 10.sup.5 pfu). As a result, a seemingly uninfected bacterial lawn was observed which consisted of B025 lysogens (VVSLC3009::B025) that were resistant to super-infection. In contrast, the B025 mutants displayed enhanced killing and were able to lyse almost all host cells in a soft-agar overlay assay at identical multiplicity of infection. Only a few survivors grew after infection with lysogeny-control mutants. When regrown, these survivors were not resistant to B025 and had an empty integration site (FIG. 4E) proofing that the LCR mutants were indeed lytic and unable to integrate into the host genome. A similar approach was successfully applied to B. cereus phage TP21-L, where the complete 2 kb LCR or the repressor only were deleted (FIG. 7). Switching phage life-style from temperate to virulent is a fast, broadly applicable method to increase the arsenal of lytic phages with enhanced antimicrobial activity that could potentially be used in a biomedical setting or for pathogen detection.

[0089] The genome engineering platform of the invention allows for a rational design and fast, reporter-free production of recombinant bacteriophages. In contrast to recombination-based technologies, cumbersome screening for recombinant phages is no longer required. In the future, this technology will enable us to tailor phages with enhanced antimicrobial properties, incorporate sensitive reporter genes into phage genomes for pathogen detection, and possibly allow for host-range modifications by switching receptor binding proteins. Many aspects of basic phage biology are still poorly understood, mostly due to the lack of efficient genetic tools for mutation, deletion, and molecular tagging of phage proteins. This is particularly true for virulent phages of Gram-positive bacteria. Therefore, the approach presented here will contribute substantially to an enhanced understanding of the biology of these nano-machines and pave the way for novel phage-based biotechnological applications beyond their use as biocontrol- and detection-agents.

Materials and Methods

Bacterial Strains and Growth Conditions

[0090] Listeria monocytogenes WSLC1042 and Mack, Listeria ivanovii WSLC3009, Bacillus thruingiensis HER1211, and Bacillus cereus HER1399 were grown at 30 C. in 0.5BHI medium. Staphylococcus aureus ATCC19685 and Staphylococcus aureus 2638A were grown at 37 C. in 0.5BHI medium. Novel Listeria monocytogenes L-form strain Rev2L was grown at 32 C. in a slightly modified version of DM3 medium for which we used tryptone instead of casamino acids (5 gL.sup.1 tryptone, 5 gL.sup.1 yeast extract, 0.01% BSA, 500 mM succinic acid, 5 gL.sup.1 glucose, 20 mM K.sub.2HPO.sub.4, 11 mM KH.sub.2PO.sub.4, 20 mM MgCl2 adjusted to pH 7.3).

Bacteriophage Propagation and DNA Extraction

[0091] Phages were propagated using the soft-agar overlay method and extracted from plates with SM buffer (100 mM NaCl, 8 mM MgSO.sub.4, and 50 mM Tris pH 7.4). Phage P35 was propagated in L. monocytogenes Mack at room-temperature using LC (LB agar supplemented with 10 mM CaCl.sub.2) as bottom and top agar (LC/LC). All other phages were propagated using 0.5BHI as the bottom- and LC as the top-agar. Phage PSA was propagated in L. monocytogenesWSLC1042 at 30 C. and all other Listeria phages were propagated on L. ivanovii WSLC3009 at 30 C. Bacillus phages Bastille and TP21-L were propagated at 30 C. on HER1211 and HER1399, respectively while Staphylococcus aureus phages K and 2638A were propagated at 37 C. on ATCC19685 and 52638A, respectively. For DNA extraction, filter-sterilized lysates were digested with DNasel (10 gml.sup.1) and RNaseA (1 U in 10 ml) for 30 min at 37 C. Phages were subsequently concentrated by PEG precipitation (7% PEG8000 and 1 M NaCl), digested with proteinase K (200 ug/ml, 50 C., 30 min, in SM buffer+10 mM EDTA pH 8), and purified using the High Pure Viral Nucleic Acid Kit (Roche Life Science). To purify DNA from large bacteriophage genomes (P100, A511, Bastille, K), PEG-precipitated phage was purified using stepped CsCI gradient ultracentrifugation, dialyzed against a 1000 excess of SM buffer, digested with proteinase K, and DNA was extracted using organic solvents as previously described.

Generation of Rev2 Strain and Reversible L. innocua L-Form Strains

[0092] Briefly, an overnight culture of Listeria EGD-e or Listeria innocua 2021 (SLCC 5639) was plated on DM3 agar supplemented with penicillin (200 g*ml.sup.1). After approximately two weeks of incubation, single colonies emerged which were transferred to liquid DM3 medium containing penicillin G (PenG) and further incubated without shaking. It took around one week until the bacteria grew in liquid medium for the first time. After growth of L-forms in liquid DM3+PenG medium, they were passaged into liquid DM3 medium without penicillin G. The resulting L-forms did not revert to their walled form in DM3 liquid medium, but only on DM3 agar plates (in absence of penicillin G). After two days of growth in liquid DM3 medium, the L-forms were plated on DM3 plates without penicillin, where revertant colonies (walled cells) usually emerged after 2-5 days. These were picked and inoculated into liquid BHI medium, in which any residual L-form cells would burst due to the absence of osmotic protectants. From this culture, a cryo stock was be prepared and the strain tested for its ability to grow in DM3 medium supplemented with penicillin G. The reversible L-form generated with EGD-e is referred to as Rev2 (walled) or Rev2L (L-form).

L-Form Transformation and Genome Rebooting

[0093] Induction of Rev2 cells into the L-form state was performed in the presence of 200 g*ml.sup.1 PenG in DM3 medium at 32 C., and L-form cells were passaged every 3-4 days by 1:1000 dilution in fresh DM3 medium. The different L-form passages were assayed for their ability to produce infective phage particles upon transformation. We observed that phage production was maximal for passage five and therefore used a cryo stock (80 C.) prepared from this passage for all experiments. Small volumes (4-5 ul) of the cryo-culture were inoculated into 1 ml prewarmed DM3+PenG, suspended using a vortex mixer at low speed, and incubated without agitation at 32 C. for 96 h. The resulting L-form culture was suspended by pipetting and adjusted to an OD.sub.600nm of 0.15 using DM3 medium. 100 l OD-adjusted Rev2L culture was mixed with 10-20 l phage genomic DNA in a 50 ml Falcon tube. 150 ul sterile 40% PEG8000 solution was added and mixed thoroughly by pipetting. After 5 min of incubation, 10 ml pre-warmed DM3 medium was added, suspended using a serological pipette, and the transformation reaction was incubated without agitation at 32 C. for 24 h unless indicated otherwise. The L-form transformation reaction was re-suspended by pipetting, and assayed for matured phage particles using the soft-agar overlay method: 5 ml of molten LC soft-agar were mixed with 50-500 l of the transformation reaction and 200 l of a fresh stationary-phase culture of a suitable phage propagation strain (the indicator strain). This soft-agar mixture was poured onto solid agar plates and incubated at the optimal growth temperature of the indicator strain to allow for phage propagation and visible plaque formation.

In-vitro Assembly of Synthetic Genomes

[0094] Bacteriophage genomes were divided in silico into three to six fragments of similar size carrying 40 bp overlapping ends. Fragments were chosen randomly for phage genomes with circular permutation. For phages with non-permuted genomes (TP21-L and B025), genome fragments were designed to allow for artificial circularization at their physical ends using long overlapping primers. Genome fragments were amplified from phage gDNA by PCR using Phusion DNA polymerase (Thermo Scientific) and subsequently purified using silica columns. Synthetic genomes were assembled for 1 h at 50 C. using the NEBuilder HiFi DNA Assembly Cloning Kit (NewEngland Biolabs). For every assembly reaction, 150-250 ng purified DNA was used per fragment in a 20 l reaction and 15 l were used for rebooting.