Functional Evaluation of a Conserved HNH Endonuclease Gene in Lactococcal Skunavirus Genomes

TL;DR

This study identifies a key HNH endonuclease gene, HrdP, essential for DNA packaging in Skunavirus phages, which could help prevent dairy fermentation failures.

Contribution

The study reveals HrdP's essential role in Skunavirus DNA packaging and its potential as a target for anti-phage strategies.

Findings

HrdP is essential for producing infective Skunavirus sk1 virions.

Deletion of hrdP prevents plaque formation and intact virion production.

HrdP deletion causes prohead conformational changes, indicating partial DNA packaging.

Abstract

Skunavirus is the most frequently encountered Lactococcus‐infecting (bacterio)phage genus in dairy fermentations and may cause milk acidification failure or inefficiencies. During phage infection, DNA packaging into the phage prohead is a crucial step for the successful maturation of phage progeny. In the Skunavirus genus of lactococcal phages, a predicted HNH endonuclease‐encoding gene adjacent to two putative DNA packaging genes is highly conserved. In the current study, we show that this gene, designated here as hrdP (HNH endonuclease related to DNA packaging), is essential for the production of infective Skunavirus sk1 virions. HrdP exhibits non‐specific endonuclease activity, while deletion of hrdP from the sk1 genome results in loss of plaque‐forming ability without hindering its DNA replication or protein production. Transmission electron microscopy analysis showed that the hrdP deletion mutant in sk1 is unable to produce intact virions. However, the proheads in this mutant displayed conformational changes associated with capsid expansion, suggesting that DNA packaging had initiated and progressed to a certain extent. This study provides valuable insights into the specific role of HrdP in Skunavirus members, highlighting it as a potential target for developing anti‐phage strategies to enhance starter culture robustness.

The HNH endonuclease HrdP is a highly conserved and essential determinant of infectivity among Skunavirus phages, specifically required for the successful completion of DNA packaging. This key role highlights HrdP as a promising and specific anti‐phage target to protect dairy starter cultures from failure.

Figures

HrdP exhibits typical HNHE protein characteristics. (A) Alignment of amino acid sequence of sk1 HrdP with other reference HNHEs from phage E2, SLT and HK97. Red‐coloured parts indicate conserved residues for enzymatic activity of HNHE (H49, N69 and H78). (B) Protein structure of sk1 HrdP predicted by AlphaFold3. The metal ions are shown as grey spheres and associated residues are indicated with sticks.

HrdP exhibits non‐specific endonucleolytic activity. (A) Produced HrdP visualised with SDS‐PAGE. (B) HrdP enzymatic activity (10 μM) against different DNA substrates. (C) HrdP activity with different enzyme concentrations (10, 1, 0.1 and 0.01 μM) and divalent metal ions (Mn2+, Mg2+ and Zn2+) against a PCR product containing the sk1 cos site.

Deletion of hrdP severely diminishes plaque‐forming ability of phage sk1. (A) Schematic overview of hrdP deletion from sk1 genome. The deleted genetic part was indicated with the striped pattern. (B) Plaque forming ability of sk1ΔhrdP and sk1 wild type with Lactococcus cremoris NZ9000 [pPTPi_hrdP] (HrdP complemented) or NZ9000 [pPTPi] (HrdP non‐complemented). Statistical significance was assessed using pairwise two‐tailed t‐tests. p < 0.001; N/S, not significant.

HrdP is not involved in DNA replication or structural protein synthesis. (A) Quantitative PCR of terS gene from different host cell DNA extracts infected with sk1ΔhrdP. (B) Proteins production (major capsid and tail) of sk1ΔhrdP propagated in different host cell. As controls, NZ9000 [pNZ44_abiP] and NZ9000 [pNZ44 + Aristaios] were used to inhibit DNA replication and protein production, respectively.

Deletion of hrdP affects maturation of progeny phages. (A) Growth curve of different host cell infected by sk1ΔhrdP with different MOI (1, 5 and 10). (B) Images of sk1ΔhrdP phage particles propagated in different host cell. (i) sk1wt with NZ9000 [pPTPi], (ii) sk1ΔhrdP with NZ9000 [pPTPi_hrdP] (complementation), (iii) sk1ΔhrdP with NZ9000 [pPTPi] (non‐complementation)—100 nm scale bar, (iv) sk1ΔhrdP with NZ9000 [pPTPi] (non‐complementation)—500 nm scale bar.

DNA packaging initiation occurs in the absence of HrdP. (A) Images of sk1 mutant phages lacking hrdP, htc1 or terS, propagated in complemented or non‐complemented host. (B) Diameter (nm) of head of defective phage particles. Statistical significance was assessed using one‐way ANOVA followed by pairwise two‐tailed t‐tests. p < 0.001; N/S, not significant.

Potential complex formation of HrdP with terminase proteins. (A) Coelution of HrdP with his‐tagged TerL. (B) Coelution of TerL with his‐tagged TerS. The red‐box indicates the co‐eluted protein along with his‐tagged protein.

Tables

TABLE 1. Plasmids, bacteria and phages used in this study.

Name	Antibiotic resistance	Function	References
Plasmids
pL2Cas9	Erythromycin		Lemay et al. (2017)
pNZ123	Chloramphenicol		Van Asseldonk et al. (1993)
pPTPi	Tetracycline		O'Driscoll et al. (2004)
pPTPi_hrdP	Tetracycline	hrdP complementation	This study
pPTPi_terS	Tetracycline	terS complementation
pPTPi_htc1	Tetracycline	htc1 complementation
pL2Cas9_2sphrdP	Erythromycin	Phage genome engineering (hrdP deletion)
pNZ123_dhrdP	Chloramphenicol
pL2Cas9_spterS	Erythromycin	Phage genome engineering (terS deletion)
pNZ123_dterS	Chloramphenicol
pL2Cas9_sphtc1	Erythromycin	Phage genome engineering (htc1 deletion)
pNZ123_2dhtc1	Chloramphenicol
pHTP1_hrdP	Kanamycin	HrdP protein production
pHTP1_terL & hrdP	Kanamycin	TerL and HrdP co‐expression
pHTP1_terL & terS	Kanamycin	TerS and TerL co‐expression
Phages
sk1			Chandry et al. (1997)
sk1ΔhrdP		sk1 mutant lacking hrdP	This study
sk1ΔterS		sk1 mutant lacking terS
sk1Δhtc1		sk1 mutant lacking htc1
Bacteria
Lactococcus cremoris
NZ9000			Linares et al. (2010)
NZ9000 [pL2Cas9_2sphrdP][pNZ123_dhrdP]	Erythromycin and chloramphenicol	Phage genome engineering (hrdP deletion)	This study
NZ9000 [pL2Cas9_spterS][pNZ123_dterS]	Erythromycin and chloramphenicol	Phage genome engineering (terS deletion)
NZ9000 [pL2Cas9_sphtc1][pNZ123_2dhtc1]	Erythromycin and chloramphenicol	Phage genome engineering (htc1 deletion)
NZ9000 [pPTPi]	Tetracycline	Empty vector control
NZ9000 [pPTPi_hrdP]	Tetracycline	hrdP complementation
NZ9000 [pPTPi_terS]	Tetracycline	terS complementation
NZ9000 [pPTPi_htc1]	Tetracycline	htc1 complementation
NZ9000 [pNZ44_abiP]	Chloramphenicol	Phage DNA replication inhibition (control)	Grafakou et al. (2024)
NZ9000 [pNZ44 + Aristaios]	Chloramphenicol	Phage protein production inhibition (control)
Escherichia coli
BL21(DE3)			Jeong et al. (2009)
BL21(DE3) [pHTP1_hrdP]	Kanamycin	HrdP protein production	This study
BL21(DE3) [pHTP1_terL & hrdP]	Kanamycin	TerL and HrdP co‐expression
BL21(DE3) [pHTP1_terS & terL]	Kanamycin	TerS and TerL co‐expression

Full text

Introduction

1

Bacteriophages (or phages) of the Skunavirus genus, formerly termed the 936 group, are the most problematic lytic lactococcal phages in the dairy fermentation industry due to their ubiquity as well as virulence (Mahony et al. 2012). Infection of a bacterial host by such phages typically causes host cell death (Salmond and Fineran 2015). During the lytic infection cycle, upon transfer of viral DNA into the host cell cytoplasm, phage DNA replication and virion production involve appropriation of relevant capabilities and resources from the host cell (Jamal et al. 2019; Sharma et al. 2017). This is followed by packaging of the replicated DNA into assembled viral components to generate complete and infective virions (Ahern et al. 2014; Bebeacua et al. 2013). Eventually, the host cell is lysed following the production of cell lysis‐enhancing proteins (holin and lysin), thereby releasing the newly formed virions that are poised to infect neighbouring cells. Efficient DNA packaging is therefore a critical step in the lytic phage infection cycle and directly determines successful virion production.

During the virion assembly process, phage DNA packaging relies on a complex molecular machine termed the ‘packasome’, which comprises the phage terminase, portal protein‐containing proheads and the DNA substrate (Dasgupta et al. 2024). The DNA packaging process, as mediated by this packasome, is highly conserved not only in most tailed phages but also across various eukaryotic double‐stranded DNA viruses, such as herpesviruses and adenoviruses (Ahi and Mittal 2016; Yang et al. 2020). Concatemeric DNA produced during DNA replication is a head‐to‐tail multimer of individual viral chromosomes. The terminase, which incorporates the terminase small (TerS) and large (TerL) subunits, initiates the packaging process via specific binding to the concatemeric DNA (Catalano 2000; Catalano et al. 1995). Subsequently, this nucleoprotein complex is matured through the endonucleolytic DNA cleavage activity of the TerL subunit, resulting in ‘complex I’. This complex subsequently docks onto the portal vertex in a preformed capsid (or prohead), thereby forming the aforementioned packasome (or ‘complex II’) (Rao and Feiss 2008). Once the packasome is formed, the terminase initiates translocation of (at least) a single phage genome‐length unit into the interior of the prohead (Isidro et al. 2004; Rao et al. 2023; Rao and Feiss 2015). Once the prohead is packaged with DNA, complex I with the remainder of the concatemeric DNA is detached from the portal vertex for DNA packaging of another prohead. The DNA‐filled prohead then assembles with other virion structural proteins, such as the tail structure, to produce mature progeny virions (Chen et al. 2020).

In addition, several other components associated with DNA packaging have been identified. For example, integration host factor (IHF) of Escherichia coli plays a crucial role in the DNA packaging of phage lambda (λ) by inducing DNA bending (Bear et al. 1984; Ortega and Catalano 2006; Sanyal et al. 2014). In addition, endonuclease VII of E. coli phage T4 is a Holliday‐structure resolvase that forms part of the packaging machinery, being responsible for debranching replicative DNA prior to packaging (Golz and Rries Kemper 1999). Furthermore, HNH endonucleases (HNHEs), which are endonucleases containing a typical enzymatic motif with three conserved amino acid residues (H‐N‐H/N) in its active site, have emerged as important players in phage DNA packaging (Kala et al. 2014; Quiles‐Puchalt et al. 2014).

In 2002, Crutz‐le Coq and colleagues identified four distinct genes in the lactococcal Skunavirus bIL170, each predicted to encode an HNHE (Coq et al. 2002). Most notable among these, Gpl3 of bIL170 was assumed to be involved in phage DNA packaging due to its encoding gene being located adjacent to terminase‐ and portal protein‐encoding genes. Although there is substantial interest in Skunavirus phages due to their negative impact on commercial dairy fermentations, the precise role of HNHE in their DNA packaging remains uncharacterized. In the present study, we investigated the enzymatic properties of this HNHE related to DNA packaging (HrdP) and the impact of its absence on the infection cycle of Skunavirus.

Materials and Methods

2

Plasmids, Bacteria and Phages

2.1

Bacterial strains, plasmids and phages used in this study are listed in Table 1. Lactococcal strains were cultivated in M17 medium (Millipore, USA) supplemented with 0.5% glucose (and then termed GM17) at 30°C. For culture preparations of host bacteria harbouring the nisin‐inducible plasmid pPTPi or its derivatives, in which a gene of interest had been cloned downstream of the nisin‐inducible promoter, fresh overnight cultures grown in the presence of tetracycline (final conc. 10 μg/mL) were used to inoculate (1%) fresh medium (without tetracycline). Cultures were incubated until an OD_600nm_ of ~0.2 was reached, followed by nisin (Sigma, USA) induction (i.e., nisin addition to a final concentration of 20 ng/mL) for 3 h. For phage assays, 10 mM of CaCl_2_ (final conc.) was added to GM17 media (Lillehaug 1997).

Phage Genome Engineering

2.2

The hrdP (or orf3) gene from the Skunavirus sk1 genome (corresponding NCBI accession number is AF011378) was removed using a CRISPR‐Cas9 editing approach (Figure S1) as previously described (Lemay et al. 2017). To achieve specific Cas9‐mediated cleavage of a target site on the sk1 genome, a spacer sequence which aligns with a 30 bp section of the hrdP gene with 5 bp containing BsaI restriction enzyme cleavage site on either end was produced using a pair of synthesised complementary oligonucleotides (2spHrdP_F and 2spHrdP_R; Table S1). Subsequently, the fragment was inserted into BsaI‐treated pL2Cas9, a vector carrying the CRISPR‐Cas9 components, using T4 DNA ligase (Thermo Fisher Scientific, USA), yielding pL2Cas9_2sphrdP. A repair template vector containing a homologous region adjacent to hrdP on the sk1 genome was also constructed. Two insert sequences A and B (298 and 309 bp, respectively) containing complementary ends to both the linearized plasmid (pNZ123) ends and each other were amplified by a conventional PCR using specific primer pairs (insert A—inAHrdP_F and 2inAHrdP_R; insert B—2inBHrdP_F and inBHrdP_R; Table S1). PCR amplification was performed using a 2720 Thermal Cycler (Applied Biosystems, Waltham, Massachusetts, USA) employing the following amplification conditions: 5 min at 94°C followed by 40 cycles (30 s at 94°C, 30 s at 55°C, 30 s at 72°C) and a final step of 7 min at 72°C. These inserts were assembled with the linearized pNZ123 using Gibson Assembly Master Mix (New England Biolabs, USA) according to the manufacturer's instructions, yielding pNZ123_dhrdP. Both vectors were transformed sequentially into Lactococcus cremoris NZ9000 competent cells by electroporation (20 μF, 200 Ω, 2.5 KV) followed by recovery for 2.5 h at 30°C in GM17 supplemented with 20 mM MgCl_2_ and 2 mM CaCl_2_ (Holo and Nes 1989), yielding L. cremoris NZ9000 [pL2Cas9_2sphrdP][pNZ123_dhrdP]. Transformants were selected on GM17 agar plate supplemented with the appropriate antibiotics (final conc. erythromycin 5 μg/mL and chloramphenicol 5 μg/mL), followed by overnight incubation at 30°C. The sequence of recombinant vectors was confirmed by Sanger sequencing by Azenta/GENEWIZ (Leipzig, Germany).

L. cremoris NZ9000 [pL2Cas9_2sphrdP][pNZ123_dhrdP] was infected with the wild type sk1 using plaque assays, and up to 90 plaques were screened for the presence of the desired mutation. The mutation was confirmed by conventional PCR with specific primers (cHrdP_F and cHrdP_R; Table S1) and using phage plaques as a source for the DNA template for the reaction. Plaques that generated a double band amplicon as visualised by agarose gel electrophoresis, corresponding to the wild type (870 bp) and expected mutant (666 bp) were selected for subsequent purification and analysis. The selected plaques were subjected to a second screening for genotype purification of the mutant phage with NZ9000 [pPTPi_hrdP]. This strain carries a plasmid containing the hrdP gene regulated by a nisin‐inducible promoter to allow ectopic expression of HrdP. For plasmid pPTPi_hrdP construction, hrdP was amplified using PCR with specific primers (SalI _HrdP_F and BglII _HrdP_R; Table S1), containing restriction enzyme sites (SalI or BglII), followed by restriction enzyme treatment. Subsequently, the amplicon was ligated with similarly‐treated pPTPi using T4 DNA ligase (Thermo Fisher Scientific, USA). Plaques presenting a single amplicon, corresponding to the mutant (sk1ΔhrdP), were propagated and stored at 4°C until required for downstream analysis (i.e., phage genome DNA extraction and sequencing, phage plaque assay). Mutant sk1 phages lacking genes encoding either the terminase small subunit (terS or orf1) or the head‐tail connector 1 (htc1 or orf7) were also produced (sk1ΔterS or sk1Δhtc1), as described above.

Phage Genome DNA Extraction and Sequencing

2.3

Genome integrity of isolated mutant phages was confirmed by whole genome sequencing. Phage particles from fresh phage lysates were precipitated overnight at 4°C using 10% (final conc.) polyethylene glycol (PEG; Sigma, USA) 8000 and resuspended in 1 mL of SM buffer (50 mM Tris HCl pH 7.5, 200 mM NaCl, 10 mM MgSO_4_), followed by DNase I and proteinase K treatment, as described previously (Lavelle et al. 2018). Phage DNA was then extracted using a Phage DNA Isolation Kit (Norgen Biotek, Canada) according to the manufacturer's instructions.

Subsequently, extracted phage DNA was subjected to whole genome sequencing conducted by GenProbio srl (Parma, Italy). The phage DNA library was constructed using the Illumina Nextera XT DNA Library Preparation Kit (Illumina, Cambridge, UK), and sequencing was performed using the Illumina MiSeq platform with a 600‐cycle flow cell version 3 according to the supplier's protocol. The obtained Illumina paired‐end reads (250 bp) were analysed for single nucleotide polymorphism (SNP) with the sk1 wild type genome sequence using the Bowtie2 alignment tool (Langmead et al. 2009) and SAMtools (Li et al. 2009) performed in the MobaXterm server (https://mobaxterm.mobatek.net/).

Plaque Assays

2.4

The mutant phage (sk1ΔhrdP) was subjected to double‐layer plaque assays (Lillehaug 1997), with host bacteria L. cremoris NZ9000 [pPTPi_hrdP] (hrdP complemented) or NZ9000 [pPTPi] (negative control) to assess the impact of the hrdP deletion on the plaque‐forming ability of sk1. After overnight incubation at 30°C, the plaque morphology and enumeration were examined. Three independent biological repeats were performed. The efficiency of plating (EOP) was calculated as (number of plaques with negative control strain/number of plaques with hrdP complemented strain).

Quantitative PCR (qPCR)

2.5

Phage DNA present in host cells was extracted as described previously with minor modifications (Patel et al. 2024). A nisin‐induced L. cremoris NZ9000 [pPTPi] culture was infected with sk1ΔhrdP at a multiplicity of infection (MOI) of 0.01. Following incubation at 30°C, 1 mL samples were collected at 10 min intervals over a period of 30 min and centrifuged at 15,000×g for 2 min at 4°C. The resulting bacterial pellets were washed with ¼ strength Ringer's solution (2.25 g/mL NaCl, 0.105 g/mL KCl, 0.12 g/mL CaCl_2_, 0.05 g/mL NaHCO_3_) followed by centrifugation to collect cells. The supernatant was discarded, and the cell pellets were frozen at −70°C prior to DNA extraction. As controls, extracts of sk1ΔhrdP infected with L. cremoris NZ9000 [pPTPi_hrdP] (complemented control) and L. cremoris NZ9000 [pNZ44_abiP] (an abortive phage infection system known to inhibit phage DNA replication) (Domingues et al. 2004) were also prepared.

Bacterial cell pellets were resuspended in 1 mL of lysis buffer (20 mM Tris–HCl pH 8, 2 mM Ethylenediamine Tetraacetic Acid [EDTA] pH 8, 2% PEG8000), to which lysozyme (50 μg/mL; Sigma, USA) and mutanolysin (100 U; Sigma, USA) were added, followed by incubation at 37°C for 3 h. After proteinase K treatment, phage DNA was extracted as described above. Quantitative PCR (qPCR) assays were performed with extracted phage DNA using SYBR select master mix (Thermo Fisher Scientific, USA) on a LightCycler 480 system (Roche, Switzerland). The reaction was performed with specific primers for Skunavirus (q936_F and q936_R; Table S1), as described previously (Muhammed et al. 2017). qPCR assays were performed employing the following thermocycling conditions: an initial incubation at 50°C for 2 min, followed by a hot‐start activation at 95°C for 2 min; 40 cycles of denaturation at 95°C for 15 s, annealing at 55°C for 30 s, and extension at 72°C for 30 s. The acquired cycle threshold (Ct) values were converted to DNA concentrations (ng/mL) based on the standard curve constructed with sk1 DNA (Y = −3.3642X + 39.696, R ^2^ = 0.9999) extracted from a phage lysate. DNA was quantified using a Qubit 1× dsDNA High Sensitivity (quantitation range 0.1–120 ng) (Thermo Fischer Scientific, USA). Fold changes in DNA concentration over time were calculated based on the average value at the 10 min point for each sample (X/average value of 10 min point). For this assay, three biological repeats were executed.

Detection of Phage Protein Production

2.6

The production of phage proteins in the host cell was examined by Western blot analysis, following a previously described procedure with minor adjustments (Grafakou et al. 2024). A nisin‐induced NZ9000 [pPTPi] culture was prepared and subsequently subjected to infection by sk1ΔhrdP at an MOI of 1. The infected cultures were incubated at 30°C and 1 mL samples were retrieved immediately (t = 0) and at 15 min intervals for 1 h, and the cell pellets were washed with ¼ strength Ringer's solution using centrifugation. Washed cell pellets were frozen at −70°C prior to DNA extraction. Cells of NZ9000 [pPTPi_hrdP] (complemented control) and NZ9000 [pNZ44 + Aristaios] (an Abi system that is known to inhibit phage protein translation) (Grafakou et al. 2024) infected with sk1ΔhrdP were similarly prepared.

Defrosted cell pellets were resuspended in 40 μL of Sodium dodecyl sulphate (SDS) loading buffer (10% glycerol, 3% SDS, 0.0625 M Tris–HCl pH 6.8, 0.1 mg/mL bromophenol blue), boiled at 100°C for 5 min, and then separated on 15% SDS–polyacrylamide gel electrophoresis (SDS‐PAGE) at 160 V for 90 min. Proteins were transferred from the SDS‐PAGE gel and fixed onto a nitrocellulose membrane (0.45 μm, Thermo Fisher Scientific, USA) using Mini‐PROTEAN 3 Cell apparatus (Bio‐Rad, USA) at 100 V for 30 min. The membrane was subsequently incubated with primary anti‐capsid or ‐tail antibodies (Davids Biotechnologie GmbH, Germany) diluted with Intercept (phosphate buffered saline; PBS) Blocking buffer (LI‐COR, UK) supplemented with 0.25% Tween‐20 (Sigma, USA) at 1:5000 at room temperature for 1 h with gentle shaking. After several washing steps with PBS (Sigma, USA), the membrane was treated with the secondary antibody (IRDye 680RD Goat anti‐Rabbit IgG; LI‐COR), diluted to 1:7500 with Intercept (PBS) Blocking buffer with 0.25% Tween‐20 at room temperature for 1 h with gentle shaking. After several washing steps with PBS, antibody‐protein complexes were visualised with ChemiDoc MP imager using the default settings (Bio‐Rad, USA).

Lysis‐In‐Broth Assays

2.7

Bacterial growth of L. cremoris NZ9000 [pPTPi] and NZ9000 [pPTPi_hrdP] (complemented control) in response to sk1ΔhrdP infection at different MOIs was monitored. The bacterial culture was diluted 1:10 in GM17 supplemented with nisin (final conc. 20 ng/mL) and CaCl_2_ (final conc. 10 mM), and 100 μL of diluted culture was dispensed into a 96‐well microtitre plate. Subsequently, 100 μL of diluted sk1ΔhrdP in GM17 containing nisin and CaCl_2_ was added at various MOIs (1, 5 and 10) to maintain a final volume of 200 μL per well. The optical density at 620 nm (O.D.620) was measured every minute for 3 h using an Absorbance 96 Plate Reader (ENZO Life Sciences, New York, USA) with incubation at 30°C.

Transmission Electron Microscopy (TEM)

2.8

Phage samples for TEM analysis were prepared as described previously (Lavelle et al. 2020). Briefly, mutant phages (sk1ΔhrdP, sk1ΔterS and sk1Δhtc1) were propagated on a nisin‐induced, non‐complemented host strain (i.e., NZ9000 [pPTPi]), and their phage particles were precipitated with 10% of PEG8000 (Sigma, USA). The sample was clarified by adding an equal volume of chloroform, and the aqueous phase was separated by centrifugation at 2000×g for 15 min. Subsequently, phage particles were purified on a discontinuous caesium chloride (CsCl; Sigma, USA). As controls, mutant phages propagated with each corresponding nisin‐induced complemented host strains and wild type sk1 propagated on L. cremoris NZ9000 [pPTPi] were also prepared.

For transmission electron microscopy, purified phage particles were absorbed on carbon films for 20 min. The carbon films were washed twice with deionised water and stained for a few seconds with 2% uranyl acetate as described previously (Sprotte et al. 2022). Electron micrographs were taken on a Talos L120C transmission electron microscope (Thermo Fisher Scientific, USA) using a 4 k × 4 k Ceta camera (Thermo Fisher Scientific, USA) set to an acceleration voltage of 80 kV. Finally, images were optimised (only with respect to brightness and contrast) using Adobe Photoshop.

Bioinformatic Analysis

2.9

The molecular weight of HrdP encoded by sk1 (the protein ID is NP_044949) was calculated in silico using Protein Molecular Weight (https://www.bioinformatics.org/sms/prot_mw.html) (Stothard 2000). To identify homologous protein structures in the protein database (PDB), the amino acid sequence of HrdP was subjected to HHPred analysis (https://toolkit.tuebingen.mpg.de/tools/hhpred) (Söding et al. 2005). Also, the amino acid sequence of sk1 HrdP was aligned with the HNHEs from phages SLT (NP_075499), HK97 (NP_037756), and E2 (YP_001522898), using MAFFT version 7 (https://mafft.cbrc.jp/alignment/server/index.html) (Katoh and Standley 2013).

Computational Analysis of Protein Structure

2.10

The protein structure of HrdP was predicted with AlphaFold3 running on the DeepL server (https://golgi.sandbox.google.com) (Abramson et al. 2024), introducing the protein and two Zn^2+^ ions as input data. Visual representations of the structures were prepared with ChimeraX (Pettersen et al. 2021).

Protein Production and Purification

2.11

The hrdP gene of phage sk1 was amplified using specific primers (HrdP_F and HrdP_R; Table S1) by PCR under the same conditions as described above, and the generated amplicon was ligated into vector pHTP1 using the NZYEasy Cloning & Expression kit (NZYTech, Portugal), according to the manufacturer's instruction. The ligation mixture was introduced into competent E. coli BL21(DE3) by heat shock‐mediated transformation at 42°C for 40 s. Following heat shock, cells were recovered in SOC media (MP Biomedicals, USA) for 1 h at 200 rpm at 37°C and cells were then plated on LB (BD Biosciences, USA) agar supplemented with kanamycin (final conc. 50 μg/mL), followed by overnight incubation at 37°C. Transformants were checked for plasmid content. One of the expected recombinant colonies was selected, and the sequence of the expected recombinant plasmid, designated here as pHTP1_hrdP, was confirmed by Sanger sequencing. The corresponding transformant, named here as BL21(DE3) [pHTP1_hrdP], was used for protein overexpression.

The 1% of BL21(DE3) [pHTP1_hrdP] overnight culture in LB media supplemented with 50 μg/mL kanamycin was propagated at 24°C for 24 h at 200 rpm in 100 mL of NZY Auto‐Induction LB media (NZYTech, Portugal) supplemented with 50 μg/mL kanamycin and 1% glycerol. Cells were harvested by centrifugation at 4000×g for 30 min at 4°C and resuspended with 20 mL of protein buffer (10 mM Tris–HCl pH 7.5, 300 mM NaCl) supplemented with 10 mM imidazole (Sigma, USA). The cell suspension was sonicated with five cycles of 30 s each with 30 s intervals on ice between treatments. After the third cycle, a 2 min rest period was included to prevent protein aggregation due to heat. The cell lysate was centrifuged at 30,000×g for 25 min at 4°C to remove cell debris.

Proteins were purified using affinity chromatography using Ni‐NTA Agarose (Qiagen, Germany), as previously described (Lavelle et al. 2020). Proteins were eluted using protein buffer with increasing concentrations of imidazole (50, 100, 150, 200, 250 mM). The eluted proteins were separated using 15% SDS‐PAGE and visualised by staining with the buffer (30% methanol, 10% acetic acid) containing 2 mg/mL Coomassie Blue (Bio‐Rad, USA) for 30 min. The protein was dialyzed using a 0.03 mm dialysis membrane (Medical Membranes Ltd., UK) in protein buffer to remove imidazole.

The terL & hrdP encompassing regions of the sk1 genome were also cloned into pHTP1 to examine the possible assembly of HrdP with the TerL, and the proteins were overexpressed and eluted, as described above. As a control, the terminase small and large subunits were co‐expressed, and their possible protein complex formation was also assessed.

HNH Endonuclease Activity

2.12

The enzymatic activity of HrdP was examined with various DNA substrates (i.e., PCR‐generated DNA template, phage sk1 genomic DNA, and plasmid DNA) as described previously with minor modifications (Zhang et al. 2017). The PCR‐derived amplicon (1998 bp) containing the sk1 cos site was produced using the specific primers (Cos_F and Cos_R; Table S1) by conventional PCR. Genomic DNA of phage sk1 was extracted and treated with T4 DNA ligase (Thermo Fisher Scientific, USA), according to manufacturer's instruction, to generate circular phage genomes, and plasmid pNZ123 was also prepared as a substrate for the assay. A 10 μL reaction mixture was prepared, containing 100–200 ng DNA substrate and 10 μM HrdP in the reaction buffer (20 mM Tris–HCl pH 8.0, 1 mM DTT, 2 mM MnCl_2_, 0.1 mg/mL BSA, and 10% glycerol). The reaction was performed at 30°C for 30 min and terminated by the addition of 100 mM EDTA. The treated DNA was visualised by electrophoresis on a 1% agarose gel and UV transillumination. Furthermore, the activity of HrdP was examined with different enzyme concentrations (10, 1, 0.3 and 0.01 μM) and divalent metal ions (Mn^2+^, Mg^2+^ and Zn^2+^).

Results

3

HrdP Exhibits Typical HNHE Protein Characteristics

3.1

Genomic analysis of 228 dairy lactococcal‐infecting Skunavirus members revealed the universal presence of a gene encoding a predicted HNHE, which is located between the genes that encode the terminase large subunit and portal protein in this phage genus (insertion location H2) (Yu et al. 2025). Also, the protein sequence of this gene product is highly conserved (> 90% sequence identity) among all analysed Skunavirus members based on a BLASTP search (https://blast.ncbi.nlm.nih.gov/), demonstrating structural homologies with GVE2 HNHE based on HHPred search (Yu et al. 2025; Zhang et al. 2017). Therefore, we hypothesized that this conserved HNHE plays a crucial role in phage DNA packaging, and we gave this HNHE the name HrdP (HNHE related to DNA packaging).

To understand the structural features of HrdP, the deduced amino acid sequence of HrdP from phage sk1 (equivalent to ORF3; 94 aa) was analysed. Based on HHPred analysis, sk1‐encoded HrdP shares significant structural similarity with the HNHE encoded by Geobacillus phage GVE2, with a probability of 98.53%. Sequence alignment with reference HNHEs derived from various phages (Figure 1A), including GVE2, SLT (Staphylococcus phage), and HK97 (Escherichia phage), showed that sk1 HrdP contains the three conserved amino acid residues (H49, N69 and H78) which are known to be critical for enzymatic activity of HNHEs, that is H49: activating a water nucleophile, N69: positioning enzymatic domain, H78: metal ion binding (Zhang et al. 2017).

HrdP exhibits typical HNHE protein characteristics. (A) Alignment of amino acid sequence of sk1 HrdP with other reference HNHEs from phage E2, SLT and HK97. Red‐coloured parts indicate conserved residues for enzymatic activity of HNHE (H49, N69 and H78). (B) Protein structure of sk1 HrdP predicted by AlphaFold3. The metal ions are shown as grey spheres and associated residues are indicated with sticks.

Structural prediction of sk1 HrdP indicates that the protein is comprised of two α‐helices and two β‐sheets (Figure 1B). Furthermore, it revealed the typical enzymatic motif of HNHE, referred to as ββα‐metal fold, which consists of two anti‐parallel β‐sheets (β1 and β2), one α‐helix (α2), and a Zn^2+^ ion binding site. The metal ion binding site was presumed to be coordinated by D48 (in β1) and H78 (in α2), based on observations in other HNHEs (Shen et al. 2004; Zhang et al. 2017). Furthermore, AlphaFold3 predicted an additional Zn^2+^ ion binding site associated with four cysteine residues (Cys34, 37, 74 and 77), which may be necessary for DNA stability, DNA binding, and/or DNA digestion, as suggested in a previous study (Moodley et al. 2012).

HrdP Exhibits Non‐Specific Endonucleolytic Activity

3.2

To evaluate its enzymatic activity, the HrdP protein of sk1 was purified and analysed (Figure 2A). The predicted molecular weight of the HrdP monomer with a hexa histidine‐tag was 13.29 kDa. The produced HrdP was tested against various possible DNA substrates (Figure 2B). When treated with cos site‐containing DNA fragments (both PCR products and sk1 genome), no specific cleavage fragments were observed, but significant degradation of DNA was observed. Furthermore, circular DNA without the cos site, pNZ123, was also susceptible to degradation by HrdP, indicating that HrdP exhibits non‐specific endonucleolytic activity. The influence of divalent metal ions and the concentrations of protein on the specificity and activity of HrdP was furthermore examined (Figure 2C). The highest activity of HrdP was observed with Mn^2+^, demonstrating activity up to a concentration of 1 μM. Conversely, when using Mg^2+^ or Zn^2+^, residual DNA was still present at a concentration of 10 μM. However, specificity was not achieved with any of the three metal ions, even at lower concentrations, indicating that neither the type of metal ion nor the enzyme concentration influences the specificity of HrdP.

HrdP exhibits non‐specific endonucleolytic activity. (A) Produced HrdP visualised with SDS‐PAGE. (B) HrdP enzymatic activity (10 μM) against different DNA substrates. (C) HrdP activity with different enzyme concentrations (10, 1, 0.1 and 0.01 μM) and divalent metal ions (Mn2+, Mg2+ and Zn2+) against a PCR product containing the sk1 cos site.

Deletion of

hrdP Severely Diminishes Plaque‐Forming Ability of Phage sk1

3.3

To understand the role(s) of HrdP in Skunavirus members, the hrdP gene was deleted from phage sk1 using an established CRISPR‐Cas9 editing procedure (Figure 3A) (Lemay et al. 2017). In an initial screening of mutant phages using L. cremoris NZ9000 [pL2Cas9_2sphrdP][pNZ123_dhrdP] (see relevant Section 2), two amplified bands were observed for certain plaques, corresponding to the wild type (870 bp) and the expected mutant phage (666 bp). However, subsequent genotype purification of the identified mutant proved very challenging (none among the 90 tested plaques), which may be attributed to the essential nature and functionality of the hrdP gene product. To account for that, the assay was repeated in the presence of a HrdP‐expressing plasmid during the genotype purification step using L. cremoris NZ9000 [pPTPi_hrdP], providing an ectopic genetic complementation. This approach indeed facilitated the isolation and purification of the desired mutant phage (designated here as sk1ΔhrdP). The approach employed allowed the identification of four validated sk1ΔhrdP mutants among 45 tested plaques. This plasmid‐based genetic complementation of hrdP further demonstrates that the observed phenotype is a direct consequence of hrdP loss and not of altered expression of neighbouring genes.

*Deletion of hrdP severely diminishes plaque‐forming ability of phage sk1. (A) Schematic overview of hrdP deletion from sk1 genome. The deleted genetic part was indicated with the striped pattern. (B) Plaque forming ability of sk1ΔhrdP and sk1 wild type with Lactococcus cremoris NZ9000 [pPTPi_hrdP] (HrdP complemented) or NZ9000 [pPTPi] (HrdP non‐complemented). Statistical significance was assessed using pairwise two‐tailed t‐tests. *p < 0.001; N/S, not significant.

To evaluate the plaque‐forming ability of sk1ΔhrdP, a plaque assay was performed using L. cremoris NZ9000 [pPTPi] (negative control; no in trans expression of HrdP) and L. cremoris NZ9000 [pPTPi_hrdP] (HrdP expression by in trans complementation) (Figure 3B). The titre of the sk1ΔhrdP lysate was above 10^9^ PFU/mL in the presence of the complementing plasmid, while, in the absence of plasmid‐mediated heterologous expression of HrdP, the titre was observed to be approximately 10^4^ PFU/mL. Furthermore, PCR‐based genotyping and Sanger sequencing of randomly selected plaques (n = 14) derived from the non‐complemented host confirmed that they all represent phages which had reverted to the wild type sk1 genotype through the acquisition of the hrdP gene (data not shown). These genetically reverted phages likely acquired the gene during the phage purification and propagation process through homologous recombination with the complementing plasmid (pPTPi_hrdP). However, this recombination appeared to occur at a very low frequency, as evidenced by the significantly lower titre with hrdP non‐complemented host (EOP < 10^−4^) compared to complemented host, likely due to the limited sequence homology (57 and 24 bp at each anticipated recombination end) as shown in Figure 3A. These results indicate that HrdP is critical for sk1‐mediated plaque formation.

HrdP Is Not Involved in DNA Replication or Structural Protein Synthesis

3.4

To elucidate the role of HrdP in plaque formation, critical steps of the sk1 phage cycle including DNA replication and structural protein synthesis were evaluated in the presence or absence of HrdP. Additionally, to compare with phages inhibited at either stage, strains containing plasmids with Abi systems that inhibit DNA replication (AbiP) (Domingues et al. 2004) or protein production (Aristaios) (Grafakou et al. 2024) were included as controls. As shown in Figure 4A, DNA replication of sk1ΔhrdP was not affected in the absence of hrdP when compared to the complemented mutant. In contrast, DNA replication was shown to be severely affected in control assays utilising a host harbouring the abortive infection system, AbiP, which is known to affect DNA replication.

HrdP is not involved in DNA replication or structural protein synthesis. (A) Quantitative PCR of terS gene from different host cell DNA extracts infected with sk1ΔhrdP. (B) Proteins production (major capsid and tail) of sk1ΔhrdP propagated in different host cell. As controls, NZ9000 [pNZ44_abiP] and NZ9000 [pNZ44 + Aristaios] were used to inhibit DNA replication and protein production, respectively.

Furthermore, the production of major capsid and tail proteins in host cells infected with sk1ΔhrdP was examined using Western blot analysis (Figure 4B). Based on qualitative comparison of the representative blots, no obvious difference in the production of these structural proteins was observed between the complemented and non‐complemented conditions. On the other hand, protein production was inhibited by Aristaios as expected, and in that case sk1ΔhrdP was no longer able to produce detectable structural proteins. Taken together, these observations suggest that HrdP is unlikely to play a major role in DNA replication or structural protein production.

Deletion of

hrdP Affects Maturation of Progeny Phages

3.5

To further investigate the infection cycle of sk1ΔhrdP, a phage infection profile was performed with the mutant and complemented mutant phage added at different MOIs (Figure 5A). Interestingly, at an MOI of 10, there was no increase in O.D.620 throughout the observation period (3 h), in the presence or absence of hrdP complementation. However, at an MOI of 5, the O.D.620 values of non‐complemented cells began to increase after 2 h, while growth of complemented host cells remained inhibited throughout the observation period. The observed inhibition activity in non‐complemented cells continued for ~110 min, after which the cell density started to increase, similar to the complemented cells. Furthermore, at an MOI of 1, the host bacteria without hrdP complementation began to grow around 30 min earlier than at MOI 5. Therefore, it is likely that the absence of HrdP affects the subsequent infection cycle, potentially hindering the spread or maturation of progeny phages.

Deletion of hrdP affects maturation of progeny phages. (A) Growth curve of different host cell infected by sk1ΔhrdP with different MOI (1, 5 and 10). (B) Images of sk1ΔhrdP phage particles propagated in different host cell. (i) sk1wt with NZ9000 [pPTPi], (ii) sk1ΔhrdP with NZ9000 [pPTPi_hrdP] (complementation), (iii) sk1ΔhrdP with NZ9000 [pPTPi] (non‐complementation)—100 nm scale bar, (iv) sk1ΔhrdP with NZ9000 [pPTPi] (non‐complementation)—500 nm scale bar.

To confirm the role of HrdP in the maturation of phage progeny, TEM analysis was performed to observe phage particles (Figure 5B). When sk1ΔhrdP was propagated in the host bacteria without complementation of HrdP‐expressing plasmid, most observed particles appeared to be defective, with a very low number of apparently intact particles (possibly reverted phages, data not shown). Conversely, the complemented mutant was observed to produce intact particles with packaged DNA in the capsid structures. These results suggest that HrdP plays a role in phage particle maturation, possibly phage DNA packaging.

DNA Packaging Initiation Occurs in the Absence of HrdP

3.6

To investigate how deletion of hrdP affects DNA packaging in phage sk1, two sk1 mutants were created, each lacking one of two essential protein encoding genes: terS and htc1. These mutants exhibited severe defects in phage maturation, producing defective progeny due to specific deficiencies: sk1ΔterS could not initiate DNA packaging, while sk1Δhtc1 could not assemble the head and tail. As expected, neither deletion mutant was able to produce intact phage progeny, but upon in trans genetic complementation with the respective deleted gene, each mutant regained the ability to produce intact phage particles (Figure 6A).

*DNA packaging initiation occurs in the absence of HrdP. (A) Images of sk1 mutant phages lacking hrdP, htc1 or terS, propagated in complemented or non‐complemented host. (B) Diameter (nm) of head of defective phage particles. Statistical significance was assessed using one‐way ANOVA followed by pairwise two‐tailed t‐tests. *p < 0.001; N/S, not significant.

During DNA translocation, the phage capsid undergoes a major structural expansion of up to 15% in outer capsid dimension, typically occurring between 10% and 25% completion of DNA packaging (Rao and Feiss 2015). As expected, the sk1Δhtc1 proheads appeared considerably larger and more variable in size compared to those of sk1ΔterS, suggesting that the prohead structure experiences expansion during DNA packaging (Patel et al. 2024). Interestingly, this shell expansion was also observed in the proheads of sk1ΔhrdP (Figure 6B), indicating that while HrdP is not essential for initiating DNA packaging, its presence may be critical for successful completion of the process. In the absence of HrdP, DNA packaging likely stalls after partial progression, preventing the formation of fully mature phage particles.

Potential Complex Formation of HrdP With Terminase Proteins

3.7

The potential complex formation of HrdP with terminase proteins was assessed. The HrdP and His‐tagged TerL were produced together in E. coli , and a lysate of this E. coli strain following induction of target protein expression was applied to Ni‐NTA affinity chromatography. Following serial elution containing increasing imidazole concentrations, possible co‐eluted HrdP along with hexa‐His‐tagged (His‐6x) TerL was examined (Figure 7A). However, co‐elution of the protein corresponding to the size of HrdP (11.19 kDa) along with His‐tagged TerL was not observed. Instead, the majority of HrdP was observed in the flow‐through, indicating that TerL and HrdP do not form a (stable) complex. Also, computational prediction of a putative TerL and HrdP complex using AlphaFold3 did not yield a confident model for the interaction (data not shown). Conversely, co‐elution of a protein which corresponds to the expected molecular mass of TerL (62.94 kDa) along with His‐tagged TerS was observed, indicating their complex formation (Figure 7B).

Potential complex formation of HrdP with terminase proteins. (A) Coelution of HrdP with his‐tagged TerL. (B) Coelution of TerL with his‐tagged TerS. The red‐box indicates the co‐eluted protein along with his‐tagged protein.

Discussion

4

Although the structural and functional characteristics of the phage DNA packaging module have been extensively studied (Chechik et al. 2024; Coshic et al. 2024; Prevo and Earnshaw 2024; Prokhorov et al. 2024), knowledge pertaining to atypical or additional components involved in this process remains limited. Here, the essential role of HrdP encoded in Skunavirus, among genes responsible for phage DNA packaging, was demonstrated. The universal presence of HrdP in Skunavirus, but not in other lactococcal phage species (e.g., Ceduovirus, P335‐like), suggests a species‐specific evolutionary adaptation that may underlie its widespread prevalence and abundance in dairy niche.

Previous studies have suggested that HNHE in phages plays an auxiliary role in cos site cleavage by introducing a single‐strand nick, while the complementary nick on the opposite strand is coordinated by TerL (Kala et al. 2014; Quiles‐Puchalt et al. 2014). However, sk1 HrdP alone did not demonstrate sequence‐specific cleavage activity, consistent with previous studies on DNA packaging related‐HNHEs from E. coli and S. aureus phages (Moodley et al. 2012; Quiles‐Puchalt et al. 2014). Interestingly, similar non‐specific cleavage activity was also observed in TerL from T4 (Bhattacharyya and Rao 1993). This suggests that the nuclease domains involved in DNA packaging, such as HrdP and TerL, may require complex formation with TerS, which mediates DNA binding, in order to achieve DNA cleavage at the cos site. However, co‐elution of HrdP with His‐tagged TerL was not observed under the tested conditions in this study, while apparent TerL/TerS co‐elution was observed (Figure 7). Possibly, assembly of DNA packaging modules is tightly regulated and may be difficult to capture experimentally using heterologous expression of individual components of the packaging machinery under conditions that may not reflect the correct in vivo situation. However, it remains unclear whether the suggested cooperative nick generation and complex formation with the terminase complex will apply to all HrdP‐associated phages. Also, given that not all bacteriophages employ HNHE for DNA packaging, the evolutionary rationale for requiring an additional nuclease component beyond TerL remains obscure, highlighting the need for further research.

Alternatively, HrdP may serve as resolvase (Coq et al. 2002). In 1981, Kemper and Garabett elucidated that endonuclease VII (gp49) in phage T4 is essential for the maturation of the phage head (Kemper and Garabett 1981). Endonuclease VII demonstrates resolving activity on four‐way (Holliday) junction and three‐way junction (Y‐structures) in branched DNA (Jensch and Kemper 1986). This resolving activity clears branched DNA precursors, facilitating phage DNA packaging. Dasgupta and colleagues highlighted the importance of this resolving activity for phage DNA packaging, showing branched DNA substrates block the portal channel, and the addition of resolvase allows the process to resume (Dasgupta et al. 2024). Furthermore, the binding ability of endonuclease VII with the portal protein has been elucidated, indicating the potential formation of a ‘packasome’ complex (Golz and Rries Kemper 1999). Both endonuclease VII and sk1 HrdP share the ββα‐metal fold in their active sites, thus exhibiting structural similarity which may reflect shared functionality (Biertümpfel et al. 2007; Sokolowska et al. 2009). Also, these structure‐specific resolvases do not exhibit sequence‐specificity, as also observed for sk1 HrdP (Golz and Rries Kemper 1999). Furthermore, the absence of HrdP did not appear to affect the initiation of DNA packaging yet seems to prevent the completion of the process after progressing to a certain point, resulting in an expanded prohead structure. The halted packaging process and resulting prohead expansion in sk1ΔhrdP mutants indicate that HrdP may play a crucial role in clearing such DNA obstructions, facilitating the final stages of phage maturation. However, the observed head expansion may also reflect an alternative, non‐productive assembly outcome rather than a DNA packaging intermediate. Further studies are therefore needed to directly assess resolvase activity of HrdP and its potential interaction with the portal protein. Additionally, given that sk1 already encodes a Holliday‐junction endonuclease, the biological rationale for the presence of dual resolvases remains unclear and requires further investigation whether they play distinct roles during the phage life cycle (e.g., Holliday‐junction endonuclease functioning in DNA replication and HrdP in DNA packaging) (Chandry et al. 1997).

In this study, we demonstrated the essential role of Skunavirus HrdP in DNA packaging. These findings extend current understanding of DNA packaging in Skunavirus by functionally characterising HrdP and associating it with the packaging process. However, its detailed functional and structural characteristics remain unclear, and experimental evidence is currently limited to a single phage strain, sk1, highlighting the need for further studies. A deeper understanding of HrdP in Skunavirus may facilitate the development of anti‐phage agents targeting its specific vulnerability at the DNA packaging step, thereby providing a potential control point for intervention and strengthening starter culture robustness.

Conclusion

5

The highly conserved hrdP, encoding an HNH endonuclease, is essential for the maturation of infective Skunavirus phage sk1. Deletion of hrdP prevents the production of intact virions, even though DNA replication and capsid expansion initiate. Given the significant impact of Skunavirus on the dairy industry, HrdP represents a unique and specific target for developing novel anti‐phage strategies to enhance starter culture robustness.

Author Contributions

Arjen Nauta: funding acquisition, resources, supervision, writing – review and editing. Natalia Biere: visualization, writing – review and editing. Frank Hille: visualization, writing – review and editing. Jennifer Mahony: conceptualization, funding acquisition, supervision, writing – review and editing. Christian Cambillau: investigation, visualization, writing – review and editing. Jun‐Hyeok Yu: conceptualization, funding acquisition, investigation, visualization, writing – original draft, writing – review and editing. Charles M. A. P. Franz: writing – review and editing. Douwe van Sinderen: conceptualization, funding acquisition, resources, supervision, writing – review and editing.

Funding

This work was supported by grants from Irish Research Council (https://research.ie/) and FrieslandCampina (https://www.frieslandcampina.com/) under award number EPSPG/2020/42 to J.‐H.Y. Science Foundation Ireland (https://www.sfi.ie/) 12/RC/2273‐P2 to D.S. Science Foundation Ireland 20/FFP‐P/8664 to J.M.

Conflicts of Interest

J.‐H.Y. was affiliated with University College Cork (UCC) at the time of manuscript preparation, but he is currently employed at Singapore‐MIT Alliance for Research and Technology (https://smart.mit.edu/). A.N. was an employee of FrieslandCampina (https://www.frieslandcampina.com/) at the time of manuscript preparation, but he is currently employed by DSM‐Firmenich (https://www.dsm‐firmenich.com/en/home.html).

Supporting information

Table S1: Primers used in this study. Figure S1: Schematic overview of bacteriophage genome engineering using CRISPR‐Cas9. Cmr, chloramphenicol resistance gene; Emr, erythromycin resistance gene.