Novel Burkholderia -Targeting Lytic Phage Rocko Preferentially Kills Stationary Phase Bacterial Cells

Abstract

Introduction:

Burkholderia cepacia complex (BCC) infections pose a significant risk to people with cystic fibrosis and can result in high mortality rates. High levels of antimicrobial resistance within the BCC have increased interest in using bacteriophages (phages) as alternative therapeutics. However, most currently available BCC-targeting phages are temperate or lysogeny-capable, limiting their therapeutic use.

Methods:

To expand the pool of lytic Burkholderia phages, a phage targeting Burkholderia cenocepacia isolated from lake water was characterised by genome sequencing, electron microscopy, host range analysis, and bacterial killing assays performed at different temperatures.

Results:

The phage, designated as Burkholderia phage vB_BceS_Rocko (φRocko), possessed a 47 kb dsDNA circularly permuted genome and siphovirus morphology. At 37°C, antibacterial activity was modest and limited to actively growing bacterial cells. However, at higher phage densities and reduced temperatures (27°C and 30°C), pronounced bacterial killing was observed, particularly against stationary-phase cells. φRocko also exhibited activity against three additional BCC species, including both clinical and environmental isolates. Turbid plaques were produced by φRocko at cooler temperatures (23°C and 30°C), whereas plaques were largely absent at 37°C. Despite this, no lysogenic activity was detected at any temperature, suggesting that plaque morphology was not associated with lysogeny.

Conclusion:

The discovery of this novel phage adds another option to the limited arsenal of lytic phages active against BCC. Notably, φRocko’s ability to target stationary-phase bacteria may prove useful for therapeutic applications.

Keywords

lytic phage phage therapy bacteriophage

Introduction

The Burkholderia cepacia complex (BCC) is a group of >20 closely related Burkholderia species, that are ubiquitous in the environment and are causative agents of opportunistic infections, primarily in people with cystic fibrosis (CF; pwCF).^1,2 In the United Kingdom, BCC is responsible for 1–4% of CF lung infections annually³ and can cause severe and recurrent infections that may develop into the invasive “cepacia syndrome.” In particular, Burkholderia cenocepacia infections can result in disqualification from lung transplants due to increased mortality.^4,5

Treatment of Burkholderia infections is complicated by high levels of intrinsic resistance to many antibiotics, including β-lactams, polymyxins, and aminoglycosides,⁶ severely limiting the availability of therapeutic options. Bacteriophages (phages) and phage therapy represent an alternative approach for difficult-to-treat infections.⁷ While these host-specific viruses have great potential, their application to Burkholderia infections is limited by the relatively small number of Burkholderia-targeting phages. Despite extensive environmental sampling efforts, just 10 lytic phages targeting the BCC have been reported,^8,9 with the majority of phages being temperate phages,⁹ capable of undergoing lysogeny through integration into the bacterial genome.^8,9 Initial studies of lytic BCC-targeting phages noted that of the four lytic phages isolated, three had highly similar genomes, with one additional diverse phage,¹⁰ suggesting a restricted but diverse phage population infecting BCC species. Similarly, a recently described lytic, flagellotropic jumbo phage exhibited a broad host range across multiple Burkholderia species,¹¹ reinforcing the notion that although the diversity of lytic Burkholderia-targeting phages is increasing, they are still rare. Temperate phages are considered less suitable for therapeutic use, as they may integrate into host chromosomes rather than entering lytic replication and can additionally transduce foreign DNA, that can increase bacterial resistance or virulence.¹² Advances in phage genetics have enabled the modification of temperate phages via the inactivation of integration-associated genes,¹³ allowing their successful therapeutic use, including the treatment of Mycobacterium abscessus infections in pwCF.^14,15 Because temperate phage-derived virions can also exhibit lytic activity against noncarrier strains without engineering, they are being investigated for therapeutic use, particularly against species like Burkholderia, for which lytic phages are scarce.^9,16 Prophage-derived phages have demonstrated efficacy in preventing Burkholderia gladioli plant infections¹⁷ and in controlling Clostridium difficile infections in vivo.¹⁸ Additionally, studies propose inducing lysis in pathogenic strains through the direct activation of endogenous prophages as a therapeutic strategy.^19,20 Nevertheless, obligately lytic phages currently remain the gold standard for therapeutics.

The successful use of phages to treat other CF lung pathogens has not yet translated into success against BCC lung infections, as the two reported attempts to treat Burkholderia infections in pwCF proved unsuccessful.^21,22 In both instances, the patients had severe disease and had already undergone lung transplantation, and phage therapy was administered as a last resort. In each case, only a single lytic phage was available, and treatment was likely administered too late to enable patient recovery from advanced infection.^21,22 Given the limited number of lytic Burkholderia phages and the challenges posed by slow-growing bacterial populations in chronic BCC infections, there remains a need to identify lytic phages capable of acting across all stages of infection. In this study, we isolated and characterized Burkholderia phage vB_BceS_φRocko, a novel lytic phage targeting B. cenocepacia K56-2, and examined its genomic features, host range, lifestyle, and behavior under different temperature conditions. In particular, we identified its unexpected activity against stationary-phase bacterial cells. Together, this study uncovers a rare example of a Burkholderia-targeting lytic phage with activity against metabolically restricted cells, which may prove useful for the treatment of chronic infections or biofilm control.

Materials and Methods

Microorganisms and culture conditions

All bacterial strains used in this study are listed in Supplementary Table S1. Burkholderia species were routinely cultured at 37°C, at 180 rpm in lysogeny broth (LB; Invitrogen, USA). For solid agar, 1% agar (VWR Chemicals, USA) was added to LB, and strains were grown at 37°C for 48 h. For phage propagation, experiments were conducted at 30°C, as is routine for many Burkholderia phages,^10,23–25 and plates were incubated for 36 h. Plaque assays used 100 µL of stationary-phase bacterial cultures (∼5.6 × 10⁹ colony forming units [CFU]/mL) and 100 µL of phage solution, with a 15-min adsorption period before 3 mL of 0.5% LB agar + 2 mM MgSO₄ (Sigma, Japan) + 2 mM CaCl₂ (Sigma, Japan) was added and poured over a solid agar base.

Phage isolation

For the isolation of a novel phage, 50 mL of water was taken from the lake bordering the Tocil Wood and Nature Reserve on the University of Warwick campus (GPS coordinates: 52.376148, −1.559658), and bacteria were removed using a 0.45 µM filter (Cytiva, USA). The water was combined with an equal volume of mid-exponential (OD_{600 nm} 0.2) B. cenocepacia K56-2 wildtype or the aflagellated deletion mutant (ΔfliF) in 2× LB supplemented with 2 mM CaCl₂ and 2 mM MgSO₄, and the samples were incubated for 20 h at 30°C, at 170 rpm. Following incubation, bacterial cells were pelleted, and the supernatant was filtered through a 0.22 µM filter (Merck, Germany), before being plated in soft agar overlay plates containing B. cenocepacia K56-2 wildtype or ΔfliF. Following the initial isolation, all phages were routinely cultured on the wildtype strain. Plaques were cored using a P1000 tip, diluted in SM buffer (100 mM NaCl, 8 mM MgSO₄, 10 mM Tris pH 7.5), and plated in soft agar overlay plates. Phages were isolated sequentially five times from the initial lysate. For the generation of phage stocks, completely lysed phage overlay plates (containing <1 × 10⁵ plaque forming units [PFU/mL]) were flooded with 5 mL SM buffer and were incubated at room temperature for 30 min. The soft agar and SM buffer were collected and agitated by vortexing to release any additional phage, and solids were pelleted at 9,000 g before the supernatant was filtered through a 0.22 µM filter (Cytiva, USA). Phages were routinely stored at 4°C.

Genome assembly and analysis

Phage DNA was purified from >1 × 10⁹ PFU/mL by phenol:chloroform extraction as outlined in the PCI/SDS DNA Extraction protocol at PhagesDB.²⁶ The DNA was submitted to MicrobesNG (United Kingdom) for Illumina Short Read paired-end sequencing. Reads were trimmed with Cutadapt v0.6.10,²⁷ down-sampled to 150× coverage with rasusa v2.0.0²⁸ and assembled with SPAdes genome assembler v3.15.5.²⁹ Genomes were annotated with Pharokka v1.7.1,³⁰ encompassing tools PHANOTATE for coding sequence (CDS) prediction,³¹ tRNAscan-SE 2.0 for tRNA prediction,³² and CRT for CRISPR prediction.³³ Functional annotation, including the detection of toxin, virulence, or antimicrobial resistance genes, was achieved through matching each CDS to the VFDB,³⁴ PHROG,³⁵ and CARD³⁶ databases using MMseqs2³⁷ and PyHMMER.³⁸ Contigs were matched to their closest relatives in the INPHARED database³⁹ using mash.⁴⁰ Genome plots were generated with pyCirclize.⁴¹ Taxonomy was assigned using TaxMyPhage.⁴² Genomes were scrutinized for lysogeny genes using PhageLeads.⁴³ Sequencing reads were indexed and mapped to the reference genome using SAMtools,⁴⁴ and the nucleic acid read mapping was visualized using Integrative Genome Viewer (IGV).⁴⁵ Single-nucleotide polymorphisms (SNPs) were identified using Snippy.⁴⁶ Genome comparisons of the amino acid sequences from annotated genes were made to the following phage genomes: vB_BceS_KL1,²³ vB_PaeS_SCH_Ab26,⁴⁷ vB_Pae-Kakheti25,⁴⁸ and vB_Pae_PA73⁴⁹ using Clinker.^23,47–50

Plaque morphology

Images of plaques following incubation at 30°C for 36 h were acquired using a CanoScan LiDE 400 scanner (Canon, Japan). Plaque size was determined using ImageJ.⁵¹ For comparison of activity across different temperatures, 5 µL of phage lysate at 1 × 10⁷, 1 × 10⁵ and 1 × 10³ PFU/mL was plated onto a wildtype B. cenocepacia K56-2 lawn, and the plates were incubated at 23°C, 30°C, and 37°C for 48 h.

Transmission electron microscopy

Phages were purified from bacterial lysate using Vivaspin 20 centrifugal concentrator 300 kDa MWCO columns (Cytiva, USA), and the buffer was exchanged into SM buffer. A 5 µL drop of purified phage was applied to a glow-discharged carbon/formvar-coated 300 mesh copper TEM grid (EM Resolutions, United Kingdom). The grids were stained with 2% uranyl acetate for 1 min and were imaged on the JEOL 2100 Plus electron microscope (JEOL, Japan) with a Gatan ultrascan camera (Gatan, USA).

Lysogeny assay

Lysogeny was screened following the protocol outlined in Gordillo Altamirano and Barr⁵² and using the Lysogeny Experiments protocol at PhagesDB.²⁶ Briefly, φRocko at concentrations of 10¹⁰, 10⁹, 10⁸, and 10⁷ PFU/mL was spotted onto a soft agar overlay with wildtype B. cenocepacia K56-2 and was incubated for 48 h at 23°C, 30°C, and 37°C, respectively. These temperatures were maintained during subsequent culturing, except that all plaque soft agar overlay plates were all cultured at 30°C. Following incubation, bacteria were collected from the mesa and streaked onto a new plate. Sixteen single colonies per temperature were collected and were streaked onto a fresh LB agar plate and onto a soft agar overlay with wildtype B. cenocepacia K56-2. To check for dormant phages, all colonies, regardless of activity on the overlay plate, were streaked onto a fresh LB agar plate, and a single colony subsequently was tested for phage activity on a new overlay plate, and lysate was spotted onto another overlay plate. Bacterial cells from the culture that were used to make the lysate were tested for the presence of φRocko by polymerase chain reaction (PCR) with the 2× PCRBIO Taq Mix Red (PCR Biosystems, England) as per manufacturer’s instructions and with primers that bound to the endolysin gene (Fwd GAACCAATCAAAACCCCGACGCCG; Rev GATTGCCCACATTATGCCGCCCC). The products were analyzed on a 1% agarose gel. The test for lysogeny was repeated three times at 30°C and once at 23°C and 37°C.

Phage virion thermostability assay

Phage virion thermostability was tested by storing 8 × 10⁹ PFU/mL of phage in 100 µL SM buffer in triplicate at 4°C, 23°C, 30°C, and 37°C for 48 h. Following the incubation period, the phages were diluted in SM buffer, and 10 µL was plated in technical triplicate onto soft agar overlay plate containing stationary phase culture, and was incubated at 30°C for 36 h. Recovered phages were quantified and analyzed for differences relative to the 4°C control sample.

Phage killing kinetics assay

Overnight cultures of wildtype B. cenocepacia K56-2 (prepared in biological triplicate) were diluted 1/100 into prewarmed media and were grown to mid-exponential (OD_{600 nm} = 0.3–0.4) phase. Cultures were then diluted to an OD_{600 nm} of 0.001 (4 × 10⁶ CFU/mL) in LB + 2 mM MgSO₄ + 2 mM CaCl₂ and were combined with ɸRocko at an initial multiplicity of infection (MOI) of 250 and subsequent 10-fold dilutions, each in technical triplicate, to a final volume of 200 µL in a 96-well plate covered with a Breathe-Easy^® membrane (Sigma, Japan). Plates were incubated in a Spark Multimode Microplate reader (Tecan, Switzerland) for 48 h at 27°C, 30°C, and 37°C with constant aeration (180 rpm). Measurements were taken every 20 min at OD_{600 nm}. The area under the curve (AUC) was calculated in R⁵³ using gcplyr.⁵⁴ The virulence index (V_i) and growth reduction coefficient (GRC) were determined using the equation 1– $[\frac{AUC (infected)}{AUC (uninfected)}]$ . For the V_i, only the lag and exponential growth phases were included to determine the AUC, whereas the entire curve was used to calculate the AUC for the GRC.

Host range analysis and efficiency of plating

Concentrated phage lysates (>1 × 10¹⁰ PFU/mL) were spotted (5 µL) onto soft agar overlay plates containing each of the Burkholderia species outlined in Supplementary Table S1 and activity was observed after 48 h at 30°C. Phages were quantified by spotting 10 µL of phage lysate diluted in SM buffer in technical triplicate onto a soft agar overlay plate containing stationary phase bacterial culture and incubated at 30°C for 48 h. Plaques were quantified and the EOP relative to the B. cenocepacia K56-2 wildtype strain was determined using the following equation: $\frac{PFU / mL test strain}{average PFU / mL K 56 - 2 host strain}$ . The EOP was conducted in biological triplicate.

Visualization and statistical analysis

Plots were generated in R⁵³ using ggplot2.⁵⁵ For comparison between two samples in the virion thermostability and host range EOP assays, a Mann–Whitney U test was performed using R to determine any statistically significant differences. For comparisons of changes in the phage kill curve, a Kruskal–Wallis test was performed using R.

Results

Phage isolation and characterisation

Six phages were isolated from the lake bordering the Tocil Wood and Nature Reserve on the University of Warwick campus against two B. cenocepacia K56-2 host strains (wildtype and ΔfliF): four using wildtype B. cenocepacia K56-2 (phages 1–4) and two using B. cenocepacia ΔfliF (phages 7 and 8). Both strains recovered similar numbers of plaques that had a shared morphology.

Short read sequencing of the dsDNA genomes confirmed that the phages isolated on the same hosts were 100% identical at the nucleotide level, while phages isolated on different hosts shared 99.18% nucleic acid identity.

A representative phage was selected from each of the identical sets for further analysis (phages 1 and 8 were isolated against wildtype and ΔfliF, respectively). Transmission electron microscopy revealed that both phages displayed siphovirus morphology with long flexible tails (184 ± 4 nm) and heads of 49 ± 4 nm (Fig. 1).

FIG. 1.

Transmission electron micrographs of φRocko (A) and φRocko.8 (B). The phages were stained with 2% uranyl acetate and were imaged at 60,000-fold magnification. The phages appear to have small proteins attached around the tail tip complex. Scale bar in bottom right represents 100 nm.

Phage genome analysis, annotation, and organisation

Genome analysis showed that phage 1 and phage 8 had genome sizes of 43,738 bp and 43,737 bp respectively and shared an average GC content of 55%. A total of 355 SNPs were detected between the genomes (Supplementary Table S2). These SNPs were likely present during the initial isolation, as it is unlikely that they arose independently during the five rounds of purification. Overall, the genomic similarity confirms that these isolates represent two variants of the same phage: Burkholderia phage vB_BceS_Rocko (φRocko; accession ERS25458009) represented by phage 1 and the closely related variant Burkholderia phage vB_BceS_Rocko.8 (φRocko.8; accession ERS25458010) represented by phage 8.

The SNPs were distributed across 41 of the predicted 58 ORFs in φRocko (Fig. 2), resulting in one less ORF in φRocko than in φRocko.8. Five ORFs did not align between the two genomes (Supplementary Table S3) and φRocko contained three fewer structural genes than φRocko.8. Notably, the major capsid protein was annotated in φRocko.8 (RCKO8_CDS_10 Supplementary Fig. S1), but not in the φRocko sequence. Given the presence of seven missense mutations and one stop mutation in the φRocko.8 region relative to φRocko, the absence of the major capsid protein in φRocko is likely due to an annotation error, rather than a true biological difference. Both genomes encoded a MazG protein, a pyrophosphohydrolase, which has been implicated in viral metabolism⁵⁶ and neither genome encoded any antimicrobial resistance (AMR) or toxin genes as determined using Pharokka. Furthermore, both genomes lacked any lysogeny markers, including transposases, excisionases and recombinases and PhageLeads⁴³ analysis did not identify any temperate markers or virulence-associated genes. As φRocko and φRocko.8 showed no detectable differences in activity, all subsequent analyses focussed exclusively on φRocko.

FIG. 2.

Graphical representation of the φRocko genome. The outer ring shows the CDSs, with colors corresponding to their predicted functions (legend underneath). Genes outlined in black highlight the CDSs that differ between the two variants of φRocko. The middle ring shows the GC content: black represents positive GC content and gray represents negative GC content. The inner ring denotes the GC skew: where purple represents positive GC skew and green represents negative GC skew. CDSs, coding sequences.

To determine the DNA packing strategy employed by the phages, the read alignment patterns of all six phages were scrutinized in IGV. For all phages, coverage was consistent, with no sudden drops or peaks. Furthermore, each phage displayed a different origin and wrap-around ends were identified in all sequences, suggesting that the φRocko genome is circularly permuted⁵⁷ with random termini. As such, all φRocko sequences were reordered to align with the holin gene. This provided a consistent start position and facilitated comparison with related phages identified through TaxMyPhage,⁴² which classified φRocko as a new phage species within the Kilunavirus genus, based on the ICTV taxonomy and the average nucleotide identity (ANI) threshold matching this genus. φRocko shared 70.6% similarity with the other Burkholderia Kilunavirus phage vB_BceS_KL1 (KL1)²³ and similarity with three Pseudomonas aeruginosa targeting phages (Fig. 3): Ab26 (55.5%), 73 (56.2%) and Kakheti25 (57%). Although PhageLeads did not identify any temperate lifestyle genes in the φRocko genome, the software detected in the KL1 phage genome predicted a cro gene, (Fig. 3: red gene), which regulates phage genome integration.^58,59 φRocko and the Pseudomonas phages each contain homologs of this gene, but PhageLeads did not identify these homologs as cro genes in any of the other phage genomes.

FIG. 3.

Genomic comparison of the novel φRocko and its closest relatives (Burkholderia phage KL1 and Pseudomonas phages Ab26, Kakheti25 and PA73). Coding sequences are color coded based on shared homology, and links between phages are colored to reflect their shared identity (legend on bottom left). The cro gene identified in KL1 and its homologs in the other phages are colored in red. Scale bar represents 2.5 kb.

Phage plaque morphology, host receptor analysis, and host range

φRocko formed plaques ranging from 0.3 mm to 3 mm in diameter with the larger plaques having an uneven shape and jagged edges relative to the smaller plaques that lacked this uneven shape (Fig. 4A). Plaques began to appear after 24 h and were fully formed after 36 h. As φRocko produces turbid plaques at 30°C, we looked at the formation of lytic zones at 23°C and 37°C (Fig. 4B). The plaques at 23°C were virtually indistinguishable from those at 30°C, except for a slightly larger hazy ring at the periphery of the zone. At 37°C, the lytic zones were substantially reduced with activity only visible at very high phage concentrations (>10⁸ PFU/mL spots) and individual plaques could not be distinguished (data not shown). This reduction was independent of the phage virion thermostability at 37°C, as the concentration of φRocko did not significantly reduce over the same time period compared to the 4°C control (Fig. 4C).

FIG. 4.

Plaque morphology and host-range analysis of φRocko. (A) Plaque morphology of φRocko on a wildtype B. cenocepacia K56-2 lawn incubated at 30°C for 36 h. (B) Lytic areas produced by φRocko under high concentrations (PFU/mL concentration shown at the top of the image) following 48 h at 23°C, 30°C, and 37°C, with phage concentration present at the top. (C) Thermostability of the φRocko virion following incubation at various temperatures for 48 h. Data represent the mean percentage of recovered phage relative to the input control (white circle) ± SD relative to the input control, n = 3. Individual data points are plotted in color. A Mann–Whitney test was conducted comparing samples to the 4°C control and any differences were not significant. (D) Efficiency of plating of φRocko against B. cenocepacia K56-2 mutant strains (ΔfliF, ΔmotB, ΔpglL) relative to the wildtype. Data represents mean ± SD, n = 3. All differences were not significant. (E) Heat map of BCC host range. Nonsusceptible strains are shown in grey. ND denotes a strain where phage activity was observed but could not be quantified. The scale for susceptible strains (blue—less susceptible, green—most susceptible) is shown under the plot.

We screened φRocko against a small selection of B. cenocepacia K56-2 mutant strains (ΔfliF—lacks a flagellum, ΔmotB—paralysed flagellum and ΔpglL—glycosylation deficient; Fig. 4D) in order to identify the phage receptor. No significant differences in infectivity were observed, suggesting that φRocko does not rely on either the flagellum or glycoproteins for infection.

To understand the host range, φRocko was screened against a panel of BCC strains and species (Supplementary Table S1; 5 B. cenocepacia strains and 14 different BCC species including type strains). φRocko showed quantifiable activity against two B. cenocepacia isolates, an environmental isolate of B. lata, the single representative of this species in the test panel and one of the two B. orbicola clinical CF isolates tested (Fig. 4E). Additional activity was observed against the B. seminalis isolate (a clinical CF isolate) at high phage titres (10⁹–10¹⁰ PFU/mL), but individual plaques were not observed under the conditions tested. φRocko infected B. cenocepacia H111 at a rate comparable to the isolation strain B. cenocepacia K56-2 and the EOP of B. lata BCC400 was 1.5-fold higher, although this was not statistically significant (Fig. 4E). The EOP for B. orbicola BCC19 was very low compared to the host strain. For all species, the φRocko plaques were turbid and consistent in size and shape to the plaques formed on B. cenocepacia K56-2. Overall, the analysis from this panel suggests that the host range for φRocko may be limited.

Temperature-dependent phage killing dynamics and assessment of lysogeny

To understand how temperature influences φRocko activity, we analyzed the effect of the phage on bacterial growth in liquid culture across six MOIs at 27°C (Fig. 5A), 30°C (Fig. 5B), and 37°C (Fig. 5C). The strongest effect was observed at 27°C, particularly at the highest phage density (MOI 250), which caused a slight delay in the stationary phase. This corresponded to a ∼0.38 local virulence index score (V_i, Fig. 5D; purple). The most striking effect at this temperature occurred during the stationary phase, where high MOIs (250 and 25) resulted in a decrease in bacterial density after 20 h. Although the growth was partially recovered, these samples never reached the density of the uninfected control, with the overall differences in growth contributing to significant p values (1.8 × 10⁻⁴⁴ for MOI 250 and 6.1 × 10⁻³² for MOI 25). This is reflected in the elevated growth reduction coefficient (GRC, Fig. 5E, purple) with scores of 0.52 for MOI 250 and 0.23 for MOI 25. At lower MOIs (2.5–0.0025), φRocko caused statistically significant reductions in growth, which are reflected in the V_i and GRC values of <0.1.

FIG. 5.

Effect of temperature on φRocko activity. Growth of B. cenocepacia K56-2 over 48 h at (A) 27°C, (B) 30°C, and (C) 37°C infected with varying concentrations of φRocko. Statistical significance was determined using a Kruskal–Wallis test and is denoted next to the sample name. ****p value < 0.001 and *p value < 0.05. For simplicity, only significant differences are shown in A–C. The point at which stationary phase was reached is denoted by a dotted line. The area under the curve (AUC) before this point was used to calculate the V_i (D) relative to the uninfected sample. The AUC for the entire graph was used to calculate the GRC (E) relative to the uninfected sample. Samples in D and E are color coded to correspond with the temperature (27°C—purple, 30°C—green, and 37°C—pink) Data: mean ± SD, n = 3.

Raising the temperature to 30°C attenuated the overall effect of φRocko on bacterial growth. At the highest phage density (MOI = 250), effect was observed on both the exponential and stationary phases (p value 6.5 × 10⁻⁴⁷), but the magnitude was reduced relative to 27°C, with intermediate V_i and GRC scores (V_i = 0.21, GRC = 0.32; Fig. 5D–E, green). For lower density samples at 30°C, the V_i and GRC values were stabilized at ∼0.1, indicating a minimal biological impact despite statistically significant differences in growth. At 37°C, φRocko showed very little activity across all MOIs, with the growth curves closely aligning with the uninfected control. The only significant difference was identified at the MOI 2.5 sample (p value 0.01), which may reflect differences in the lag phase rather than phage-mediated activity. Consistent with this, the V_i and GRC scores were < 0.06, dropping to <0 for the GRC at MOIs = 25–0.025, reflecting lack of phage activity (Fig. 5D–E, pink). Notably, at MOI 250, the V_i was 0.1, despite the GRC being 0.03, showing that φRocko had a small effect on exponential phase bacteria at 37°C, but had no effect on stationary phase cells. Collectively, these data suggest that φRocko activity is strongly temperature dependent, with higher phage densities having a more profound effect on stationary phase cells.

As the phage killing kinetic assays suggest, φRocko was never able to clear cultures, yet was capable of producing high titres in liquid media (>1 × 10¹⁰ PFU/mL). This pattern of minimal bacterial killing aligns with the turbid plaques formed by the phage, yet contrasts with the observation of high phage productivity, suggesting that φRocko may use a passive phage release mechanism to largely avoid cell death. Such passive phage release mechanisms are characteristic of filamentous phages and have additionally been observed in at least one prophage infecting Lactobacillus lactis⁶⁰ but these have not been previously shown for any dsDNA phages infecting Burkholderia. Although the investigation of such a mechanism was beyond the scope of this study, the concept merits exploration in future work. To determine whether this behavior could reflect a temperate lifestyle, we performed lysogeny assays at 23°C, 30°C, and 37°C. No phage activity was detected in the subsequent bacterial isolates in the first or third generations at any temperature and PCR screening did not identify φRocko in the bacterial genome in any sample at the conclusion of the assay (data not shown). Together, these results suggest that φRocko is not temperate and cannot undergo lysogeny.

Discussion

Burkholderia infections remain amongst the most difficult to treat infections in pwCF due to their intrinsic drug resistance. While phage therapy offers a therapeutic alternative, a lack of lytic phages active against BCC hinders this advance. We have identified two variants of a novel lytic phage, φRocko and φRocko.8, adding to the 10 lytic phages previously characterized against this pathogen.⁸ φRocko exhibited temperature-sensitive infection dynamics with activity favored at cooler temperatures (23–30°C) and could infect stationary phase cells at high phage densities. φRocko also displayed a limited host range, infecting <20% of the strains tested. This restricted range is consistent with the host ranges reported for other lytic phages targeting the BCC^8,9 and may represent an intrinsic feature of phages infecting this genus. However, it is also likely influenced by the limited number of characterized lytic phages and the incomplete assessment of their host ranges. These features highlight the constraints and potential of φRocko as a therapeutic agent.

To assess the host-driven differences between the two φRocko variants, we compared the φRocko and φRocko.8 genomes and identified 355 SNPs that likely arose during the initial isolation on the B. cenocepacia wildtype and ΔfliF strains. Despite these genomic changes altering the predicted ORF content and many of the SNPs occurring in structural genes, φRocko.8 displayed indistinguishable morphology. These alterations may have contributed to functional differences between the variants and this warrants further investigation. Similarly, efforts to identify the phage host receptor were unsuccessful. A targeted screen of mutants was unable to identify any statistically significant differences between the mutants and the wildtype strain. Therefore, a more extensive mutant library is required to identify the receptor required for φRocko infection.

φRocko was most closely related to another B. cenocepacia-targeting phage KL1, a lysogen-capable phage isolated from sewage.²³ KL1 produces turbid plaques and exhibits delayed plaque formation,²³ characteristics shared with φRocko and also encodes a predicted cro gene. Although φRocko encodes a protein sharing 76% amino acid identity with KL1 Cro, this homolog was not flagged as cro in φRocko, nor in related Pseudomonas phages, including Kakheti25, a lytic phage that forms clear plaques.⁴⁸ Consistent with a strictly lytic lifestyle, φRocko lacks genes associated with lysogeny, including an integrase protein, which is typically >250 amino acids,⁶¹ substantially larger than the 20+ hypothetical proteins encoded in the φRocko genome. KL1 has previously been shown to be retained in 22% of first-generation bacterial cells, following infection with the phage.⁶² However, retention of KL1 was strain specific and while the presence of the phage genome was confirmed by PCR, its integration into the genome was not tested and could represent pseudolysogeny—a form of episomal replication—rather than true integration. While φRocko was screened for lysogeny in the third generation, it is possible that a small proportion of φRocko was transiently maintained in a limited number of cells and escaped detection. Such temporary maintenance of φRocko via pseudolysogeny could contribute to the delayed killing of stationary phase cells. Nevertheless, the absence of lysogeny-associated genes and the lack of detected lysogens, supports the conclusion that φRocko is functionally lytic.

The ability of φRocko to impact stationary phase bacterial cells was most pronounced at 27°C, where high phage densities (MOIs 250 and 25) resulted in a measurable reduction in cell density during the stationary phase. The GRC captures this impact best, integrating the effects across the full curve, whereas the V_i only captures the disruption to exponential growth. Although all φRocko-infected samples at 27°C temperature differed significantly from the uninfected control, the effects at MOIs ≤ 2.5 were biologically minimal, as reflected by the V_i and GRC values below 0.1. At 30°C, a similar but attenuated pattern of phage activity was observed, indicating that the modest temperature increase reduced the efficacy of φRocko infection. Although impact to both the exponential and stationary growth phase were observed at the highest phage density, the magnitude of this effect was reduced relative to the 27°C across both the V_i and GRC. Once the phage density dropped below MOI 25 at 30°C, the V_i and GRC values remained constant, sitting around 0.1, regardless of the reducing phage concentration. Given the small p-values across these samples, it is likely that this very slight impact on growth is phage mediated, which may lend the phage to being useful in low densities in the presence of additional phages. At 37°C, a slight dose-dependent response can be observed in the V_i values between MOIs 250 and 25. However, this was not shared in the GRC, where all samples returned values close to 0, showing the lack of impact of φRocko at 37°C outside a minor effect in the exponential phase. While the overall effect of φRocko is still modest compared to many phages that can decimate a population, the killing of stationary phase cells is an important and less frequently observed feature.^63–65 Stationary phase cells are metabolically distinct from actively growing cells and often feature less receptors compared to their exponential counterparts, which for many phages makes them unsuitable hosts. However, the stationary phase may be more reflective of how the bacteria exist in the general environment, where the ideal growth conditions are not always met.⁶⁶ The metabolic restriction of stationary phase is also biologically similar in a number of ways to a biofilm^67,68 and it might therefore be possible that these stationary phase active phages like φRocko, could have antibiofilm properties. Another factor that may influence the differences in infectivity across temperatures and growth phases, is φRocko’s ability to adsorb to host cells. Cell membrane composition changes between growth phases⁶⁹ and in different temperatures,⁷⁰ altering phage receptor availability and affecting a phage’s ability to bind to host cells. Investigation into φRocko’s ability to adsorb under these different conditions would allow a deeper understanding of its behavior across temperatures and growth phases and will provide key information for future studies.

The increased activity at lower temperatures observed in φRocko has been previously observed in numerous low-temperature active phages,^71–73 including those isolated against Pseudomonas and Escherichia.^74–77 These phages, which were isolated from a range of sources, demonstrate significantly reduced or completely ablated activity at higher temperatures and have been of interest for their applications in food manufacturing^71,74,78 and biofilm control,⁷³ due to their activity in these environmental conditions.^79,80 The molecular mechanisms that control for low temperature activity are unknown and thus, we cannot discriminate any genes in the φRocko genome that would corroborate the low-temperature preference we have observed, although the bacterial metabolism and overall phage virion thermostability is likely to be involved.⁷⁵ As some studies have shown that certain BCC clinical isolates are indistinguishable from environmental isolates,^81,82 it is possible that φRocko is adapted for activity on environmental isolates and that there is enough similarity to the clinical isolates that we were able to observe some activity. Given its lack of activity at 37°C, φRocko in its current state would not be suitable for phage therapy of clinical BCC infections in the absence of other phages. However, in vitro evolution could be applied to increase φRocko’s temperature range^83,84 and if achieved, the phage’s predisposition to target stationary phase bacteria could be a powerful tool in the treatment of biofilms^85,86 or chronic infections.

Conclusion

This study describes the isolation and characterisation of φRocko, a novel phage with activity across four BCC species and a genomic profile consistent with therapeutic safety requirements. Its capacity to infect stationary phase cells at 27°C and 30°C is a particularly noteworthy finding, as this phenotype is relatively uncommon and may reflect adaptation to biofilm or environmental settings, where bacteria exist predominantly in a slow-growing state.

The principal limitation of φRocko is its failure to demonstrate significant activity at 37°C, which prevents its use as a standalone therapeutic for human infections. The mechanistic basis of this temperature sensitivity remains unclear and merits further investigation, along with broader testing against a wider panel of clinical BCC isolates to fully define its host range and resistance potential.

Despite this, φRocko’s stationary phase activity offers functional complementarity to phages that target actively dividing cells, making it a strong candidate for inclusion in phage cocktails targeting chronic BCC infections, such as those seen in cystic fibrosis patients. Future work focused on a phage cocktail design and the molecular basis of temperature-dependent phage infectivity, will be key to translating these findings towards clinical applications.

Data Availability

All genome sequencing data generated in this study are available under the project PRJEB95892 accessions ERS25458009 for Burkholderia phage vB_BceS_Rocko and ERS25458010 for Burkholderia phage vB_BceS_Rocko.8.

Author Disclosure Statement

No competing financial interests exist.

Supplemental Material

sj-docx-1-pta-10.1177_26416549261465080 — Supplemental material for Novel Burkholderia-Targeting Lytic Phage Rocko Preferentially Kills Stationary Phase Bacterial Cells

Supplemental material, sj-docx-1-pta-10.1177_26416549261465080 for Novel Burkholderia-Targeting Lytic Phage Rocko Preferentially Kills Stationary Phase Bacterial Cells by Jessica M. Lewis, Olivia Dutton, Tanaz Mistry, and Antonia P. Sagona

Supplemental Material

sj-csv-2-pta-10.1177_26416549261465080 — Supplemental material for Novel Burkholderia-Targeting Lytic Phage Rocko Preferentially Kills Stationary Phase Bacterial Cells

Supplemental material, sj-csv-2-pta-10.1177_26416549261465080 for Novel Burkholderia-Targeting Lytic Phage Rocko Preferentially Kills Stationary Phase Bacterial Cells by Jessica M. Lewis, Olivia Dutton, Tanaz Mistry, and Antonia P. Sagona

Footnotes

Funding Information

The authors acknowledge funding from the Monash Warwick Alliance AMR training programme to A.P.S. that funded the work of J.M.L. The authors also acknowledge support from the BBSCR TRDF (Grant BB/V017217/1) to A.P.S. and from the School of Life Sciences, University of Warwick to T.M. (funded by the MBio programme), O.D., J.M.L., and A.P.S. Burkholderia BCC strains were provided by Prof. Eshwar Mahenthiralingam, Cardiff University, as part of the United Kingdom CF Infection Biorepository (). The authors would like to thank the University of Warwick, School of Life Sciences Media Preparation team for their assistance preparing the media used in this study and Dr. Saskia Bakker from the Advanced Bioimaging RTP at University of Warwick for her assistance with the transmission electron microscopy.

References

1. Jin

, Zhou

, et al. Genome-based classification of Burkholderia cepacia complex provides new insight into its taxonomic status. Biol Direct 2020;15(1):6; doi: 10.1186/s13062-020-0258-5

2. Gutiérrez Santana

, Coria Jiménez

. Burkholderia cepacia complex in cystic fibrosis: Critical gaps in diagnosis and therapy. Ann Med 2024;56(1):2307503; doi: 10.1080/07853890.2024.2307503

3.Trust CF. UK Cystic Fibrosis Registry 2024 Annual Data Report. Cystic Fibrosis Trust: London; 2024.

4.NHS Blood and Transport. POL231/7—Lung candidate selection criteria. NHS Blood and Transplant (NHSBT): Bristol, UK; 2023.

5. Sfeir

. Burkholderia cepacia complex infections: More complex than the bacterium name suggest. J Infect 2018;77(3):166–170; doi: 10.1016/j.jinf.2018.07.006

6. Rhodes

, Schweizer

. Antibiotic resistance in Burkholderia species. Drug Resist Updat 2016;28:82–90; doi: 10.1016/j.drup.2016.07.003

7. Zalewska-Piątek

. Phage therapy-challenges, opportunities and future prospects. Pharmaceuticals (Basel) 2023;16(12):1638; doi: 10.3390/ph16121638

8. Wang

, Zhang

, Chen

, et al. Burkholderia phages and control of Burkholderia-associated human, animal, and plant diseases. Microorganisms 2025;13(8):1873

9. Lauman

, Dennis

. Advances in thage therapy: Targeting the Burkholderia cepacia complex. Viruses 2021;13(7):1331; doi: 10.3390/v13071331

10.

10. Summer

, Gonzalez

, Bomer

, et al. Divergence and mosaicism among virulent soil phages of the Burkholderia cepacia complex. J Bacteriol 2006;188(1):255–268; doi: 10.1128/jb.188.1.255-268.2006

11.

11. Supina

BSI

, McCutcheon

, Peskett

, et al. A flagella-dependent Burkholderia jumbo phage controls rice seedling rot and steers Burkholderia glumae toward reduced virulence in rice seedlings. mBio 2025;16(3):e0281424; doi: 10.1128/mbio.02814-24

12.

12. Monteiro

, Pires

, Costa

, et al. Phage therapy: Going temperate? Trends Microbiol 2019;27(4):368–378; doi: 10.1016/j.tim.2018.10.008

13.

13. Dedrick

, Guerrero Bustamante

, Garlena

, et al. Mycobacteriophage ZoeJ: A broad host-range close relative of mycobacteriophage TM4. Tuberculosis (Edinb) 2019;115:14–23; doi: 10.1016/j.tube.2019.01.002

14.

14. Dedrick

, Guerrero-Bustamante

, Garlena

, et al. Engineered bacteriophages for treatment of a patient with a disseminated drug-resistant Mycobacterium abscessus. Nat Med 2019;25(5):730–733; doi: 10.1038/s41591-019-0437-z

15.

15. Dedrick

, Smith

, Cristinziano

, et al. Phage therapy of mycobacterium infections: Compassionate use of phages in 20 patients with drug-resistant mycobacterial disease. Clin Infect Dis 2023;76(1):103–112; doi: 10.1093/cid/ciac453

16.

16. Nordstrom

, Griffith

, Rangachar Srinivasa

, et al. Harnessing the diversity of Burkholderia spp. Prophages for therapeutic potential. Cells 2024;13(5):428; doi: 10.3390/cells13050428

17.

17. Lauman

, Dennis

. Prophylactic phage biocontrol prevents Burkholderia gladioli infection in a quantitative ex planta model of bacterial virulence. Appl Environ Microbiol 2024;90(10):e0131724; doi: 10.1128/aem.01317-24

18.

18. Nale

, Spencer

, Hargreaves

, et al. Bacteriophage combinations significantly reduce Clostridium difficile growth in vitro and proliferation in vivo. Antimicrob Agents Chemother 2016;60(2):968–981; doi: 10.1128/aac.01774-15

19.

19. Al-Anany

, Fatima

, Hynes

. Temperate phage-antibiotic synergy eradicates bacteria through depletion of lysogens. Cell Rep 2021;35(8):109172; doi: 10.1016/j.celrep.2021.109172

20.

20. Lauman

, Cassell

TRH

, Dennis

. Synergy between ciprofloxacin and temperate phages overcomes therapeutic limitations caused by lysogen formation. Sci Rep 2025;16(1):850; doi: 10.1038/s41598-025-30556-7

21.

21. Aslam

, Courtwright

, Koval

, et al. Early clinical experience of bacteriophage therapy in 3 lung transplant recipients. Am J Transplant 2019;19(9):2631–2639; doi: 10.1111/ajt.15503

22.

22. Haidar

, Chan

, Cho

, et al. Phage therapy in a lung transplant recipient with cystic fibrosis infected with multidrug-resistant Burkholderia multivorans. Transpl Infect Dis 2023;25(2):e14041; doi: 10.1111/tid.14041

23.

23. Lynch

, Stothard

, Dennis

. Comparative analysis of two phenotypically-similar but genomically-distinct Burkholderia cenocepacia-specific bacteriophages. BMC Genomics 2012;13:223; doi: 10.1186/1471-2164-13-223

24.

24. Weiser

, Yap

, Otter

, et al. A novel inducible prophage from Burkholderia vietnamiensis g4 is widely distributed across the species and has lytic activity against pathogenic Burkholderia. Viruses 2020;12(6):601; doi: 10.3390/v12060601

25.

25. Lynch

, Abdu

, Schobert

, et al. Genomic characterization of JG068, a novel virulent podovirus active against Burkholderia cenocepacia. BMC Genomics 2013;14:574; doi: 10.1186/1471-2164-14-574

26.

26. Russell

, Hatfull

. PhagesDB: The actinobacteriophage database. Bioinformatics 2017;33(5):784–786; doi: 10.1093/bioinformatics/btw711

27.

27. Martin

. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet j 2011;17(1):10–12.

28.

28. Hall

. Rasusa: Randomly subsample sequencing reads to a specified coverage. JOSS 2022;7(69):3941.

29.

29. Prjibelski

, Antipov

, Meleshko

, et al. Using SPAdes de novo assembler. Curr Protoc Bioinformatics 2020;70(1):e102; doi: 10.1002/cpbi.102

30.

30. Bouras

, Nepal

, Houtak

, et al. Pharokka: A fast scalable bacteriophage annotation tool. Bioinformatics 2023;39(1):btac776; doi: 10.1093/bioinformatics/btac776

31.

31. McNair

, Zhou

, Dinsdale

, et al. PHANOTATE: A novel approach to gene identification in phage genomes. Bioinformatics 2019;35(22):4537–4542; doi: 10.1093/bioinformatics/btz265

32.

32. Chan

, Lin

, Mak

, et al. tRNAscan-SE 2.0: Improved detection and functional classification of transfer RNA genes. Nucleic Acids Res 2021;49(16):9077–9096; doi: 10.1093/nar/gkab688

33.

33. Bland

, Ramsey

, Sabree

, et al. CRISPR Recognition Tool (CRT): A tool for automatic detection of clustered regularly interspaced palindromic repeats. BMC Bioinformatics 2007;8(1):209; doi: 10.1186/1471-2105-8-209

34.

34. Chen

, Yang

, Yu

, et al. VFDB: A reference database for bacterial virulence factors. Nucleic Acids Res 2005;33:D325–D328; doi: 10.1093/nar/gki008

35.

35. Terzian

, Olo Ndela

, Galiez

, et al. PHROG: Families of prokaryotic virus proteins clustered using remote homology. NAR Genom Bioinform 2021;3(3):lqab067; doi: 10.1093/nargab/lqab067

36.

36. Alcock

, Raphenya

, Lau

TTY

, et al. CARD 2020: Antibiotic resistome surveillance with the comprehensive antibiotic resistance database. Nucleic Acids Res 2020;48(D1):D517–D525; doi: 10.1093/nar/gkz935

37.

37. Steinegger

, Söding

. MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nat Biotechnol 2017;35(11):1026–1028; doi: 10.1038/nbt.3988

38.

38. Larralde

, Zeller

. PyHMMER: A Python library binding to HMMER for efficient sequence analysis. Bioinformatics 2023;39(5):btad214; doi: 10.1093/bioinformatics/btad214

39.

39. Cook

, Brown

, Redgwell

, et al. INfrastructure for a PHAge REference Database: Identification of large-scale biases in the current collection of cultured phage genomes. Phage (New Rochelle) 2021;2(4):214–223; doi: 10.1089/phage.2021.0007

40.

40. Ondov

, Treangen

, Melsted

, et al. Mash: Fast genome and metagenome distance estimation using MinHash. Genome Biol 2016;17(1):132; doi: 10.1186/s13059-016-0997-x

41.

41. Shimoyama

. pyCirclize: Circular visualization in Python. GitHub, Inc: San Francisco, CA; 2022. Available from: https://github.com/moshi4/pyCirclize

42.

42. Millard

, Denise

, Lestido

, et al. taxMyPhage: Automated taxonomy of dsDNA phage genomes at the genus and species level. Phage (New Rochelle) 2025;6(1):5–11; doi: 10.1089/phage.2024.0050

43.

43. Yukgehnaish

, Rajandas

, Parimannan

, et al. PhageLeads: Rapid assessment of phage therapeutic suitability using an ensemble machine learning approach. Viruses 2022;14(2):342; doi: 10.3390/v14020342

44.

44. Li

, Handsaker

, Wysoker

, et al.; 1000 Genome Project Data Processing Subgroup. The sequence alignment/map format and SAMtools. Bioinformatics 2009;25(16):2078–2079; doi: 10.1093/bioinformatics/btp352

45.

45. Robinson

, Thorvaldsdóttir

, Winckler

, et al. Integrative genomics viewer. Nat Biotechnol 2011;29(1):24–26; doi: 10.1038/nbt.1754

46.

46. Seemann

. snippy: Fast bacterial variant calling from NGS reads. 2015.

47.

47. Essoh

, Latino

, Midoux

, et al. Investigation of a large collection of pseudomonas aeruginosa bacteriophages collected from a single environmental source in Abidjan, Côte d’Ivoire. PLoS One 2015;10(6):e0130548; doi: 10.1371/journal.pone.0130548

48.

48. Karumidze

, Thomas

, Kvatadze

, et al. Characterization of lytic Pseudomonas aeruginosa bacteriophages via biological properties and genomic sequences. Appl Microbiol Biotechnol 2012;94(6):1609–1617; doi: 10.1007/s00253-012-4119-8

49.

49. Kwan

, Liu

, Dubow

, et al. Comparative genomic analysis of 18 Pseudomonas aeruginosa bacteriophages. J Bacteriol 2006;188(3):1184–1187; doi: 10.1128/jb.188.3.1184-1187.2006

50.

50. Gilchrist

CLM

, Chooi

Y-H

. Clinker clustermap: Automatic generation of gene cluster comparison figures. Bioinformatics 2021;37(16):2473–2475; doi: 10.1093/bioinformatics/btab007

51.

51. Schindelin

, Arganda-Carreras

, Frise

, et al. Fiji: An open-source platform for biological-image analysis. Nat Methods 2012;9(7):676–682; doi: 10.1038/nmeth.2019

52.

52. Altamirano

FLG

, Barr

. Screening for lysogen activity in therapeutically relevant bacteriophages. Bio Protoc 2021;11(8):e3997; doi: 10.21769/BioProtoc.3997

53.

53.R Core Team R. R: A language and environment for statistical computing. Citeseer: 2020.

54.

54. Blazanin

. gcplyr: an R package for microbial growth curve data analysis. BMC Bioinformatics 2024;25(1):232; doi: 10.1186/s12859-024-05817-3

55.

55. Villanueva

RAM

, Chen

. ggplot2: elegant graphics for data analysis. Taylor & Francis: 2019.

56.

56. Rihtman

, Bowman-Grahl

, Millard

, et al. Cyanophage MazG is a pyrophosphohydrolase but unable to hydrolyse magic spot nucleotides. Environ Microbiol Rep 2019;11(3):448–455; doi: 10.1111/1758-2229.12741

57.

57. Casjens

, Gilcrease

. Determining DNA packaging strategy by analysis of the termini of the chromosomes in tailed-bacteriophage virions. Methods Mol Biol 2009;502:91–111; doi: 10.1007/978-1-60327-565-1_7

58.

58. Schubert

, Dodd

, Egan

, et al. Cro’s role in the CI Cro bistable switch is critical for {lambda}’s transition from lysogeny to lytic development. Genes Dev 2007;21(19):2461–2472; doi: 10.1101/gad.1584907

59.

59. Little

, Shepley

, Wert

. Robustness of a gene regulatory circuit. Embo J 1999;18(15):4299–4307; doi: 10.1093/emboj/18.15.4299

60.

60. Liu

, Alexeeva

, Bachmann

, et al. Chronic release of tailless phage particles from Lactococcus lactis. Appl Environ Microbiol 2022;88(1):e01483-21; doi: 10.1128/AEM.01483-21

61.

61. Groth

, Calos

. Phage integrases: Biology and applications. J Mol Biol 2004;335(3):667–678; doi: 10.1016/j.jmb.2003.09.082

62.

62. Lauman

, Dennis

. Synergistic interactions among Burkholderia cepacia complex-targeting phages reveal a novel therapeutic role for lysogenization-capable phages. Microbiol Spectr 2023;11(3):e0443022; doi: 10.1128/spectrum.04430-22

63.

63. Silva

, Pinto

, França

, et al. Phage SEP1 hijacks Staphylococcus epidermidis stationary cells’ metabolism to replicate. mSystems 2024;9(7):e0026324; doi: 10.1128/msystems.00263-24

64.

64. Bryan

, El-Shibiny

, Hobbs

, et al. Bacteriophage T4 infection of stationary phase E. coli: Life after log from a phage perspective. Front Microbiol 2016;7:1391.

65.

65. Robb

, Woods

, Robb

. Phage growth characteristics on stationary phase Achromobacter cells. J Gen Virol 1978;41(2):265–272; doi: 10.1099/0022-1317-41-2-265

66.

66. Velásquez

, Huguet-Tapia

, He

. Shared in planta population and transcriptomic features of nonpathogenic members of endophytic phyllosphere microbiota. Proc Natl Acad Sci U S A 2022;119(14):e2114460119; doi: 10.1073/pnas.2114460119

67.

67. Shree

, Singh

, Sodhi

, et al. Biofilms: Understanding the structure and contribution towards bacterial resistance in antibiotics. Med Microecol 2023;16:100084; doi: 10.1016/j.medmic.2023.100084

68.

68. Pires

, Melo

LDR

, Azeredo

. Understanding the complex phage-host interactions in biofilm communities. Annu Rev Virol 2021;8(1):73–94; doi: 10.1146/annurev-virology-091919-074222

69.

69. Lee

T-H

, Charchar

, Separovic

, et al. The intricate link between membrane lipid structure and composition and membrane structural properties in bacterial membranes. Chem Sci 2024;15(10):3408–3427; doi: 10.1039/d3sc04523d

70.

70. Peano

, Chiaramonte

, Motta

, et al. Gene and protein expression in response to different growth temperatures and oxygen availability in Burkholderia thailandensis. PLoS One 2014;9(3):e93009; doi: 10.1371/journal.pone.0093009

71.

71. Abhisingha

, Dumnil

, Pitaksutheepong

. Efficiency of phage cocktail to reduce Salmonella Typhimurium on chicken meat during low temperature storage. LWT 2020;129:109580; doi: 10.1016/j.lwt.2020.109580

72.

72. Qin

, Cheng

, Zhang

, et al. Complete genome sequence of the cold-active bacteriophage VMY22 from Bacillus cereus. Virus Genes 2016;52(3):432–435; doi: 10.1007/s11262-016-1300-7

73.

73. Fernández

, Duarte

, Jurado

, et al. Multipronged impact of environmental temperature on Staphylococcus aureus infection by phage Kayvirus rodi: Implications for biofilm control. Biofilm 2025;9:100248; doi: 10.1016/j.bioflm.2024.100248

74.

74. Qin

, Gong

, Li

, et al. Psychrophilic phage phiGM22-3 efficiently controls Pseudomonas fluorescens contamination in cold-stored milk. Int J Food Microbiol 2024;411:110525; doi: 10.1016/j.ijfoodmicro.2023.110525

75.

75. Wei

, Wang

, Tang

, et al. Characterization of the cold-adapted broad-spectrum bacteriophage PJNS034 and its application in controlling pathogenic bacteria in dairy products. LWT 2025;223:117685; doi: 10.1016/j.lwt.2025.117685

76.

76. Zajančkauskaitė

, Noreika

, Rutkienė

, et al. Low-temperature virus vB_EcoM_VR26 shows potential in biocontrol of STEC O26:H11. Foods 2021;10(7):1500.

77.

77. Kaliniene

, Truncaitė

, Šimoliūnas

, et al. Molecular analysis of the low-temperature Escherichia coli phage vB_EcoS_NBD2. Arch Virol 2018;163(1):105–114; doi: 10.1007/s00705-017-3589-5

78.

78. Tanaka

, Yamada

, Takeuchi

, et al. A lytic bacteriophage for controlling Pseudomonas lactis in raw cow’s milk. Appl Environ Microbiol 2018;84(18):e00111-18–e00118; doi: 10.1128/AEM.00111-18

79.

79. Hwang

, Cho

, Kang

, et al. Genomic analysis of two cold-active pseudoalteromonas phages isolated from the continental shelf in the Arctic Ocean. Viruses 2023;15(10):2061; doi: 10.3390/v15102061

80.

80. Senčilo

, Luhtanen

A-M

, Saarijärvi

, et al. Cold-active bacteriophages from the Baltic Sea ice have diverse genomes and virus–host interactions. Environ Microbiol 2015;17(10):3628–3641; doi: 10.1111/1462-2920.12611

81.

81. Baldwin

, Mahenthiralingam

, Drevinek

, et al. Environmental Burkholderia cepacia complex isolates in human infections. Emerg Infect Dis 2007;13(3):458–461; doi: 10.3201/eid1303.060403

82.

82. LiPuma

, Spilker

, Coenye

, et al. An epidemic Burkholderia cepacia complex strain identified in soil. Lancet 2002;359(9322):2002–2003; doi: 10.1016/s0140-6736(02)08836-0

83.

83. Kering

, Zhang

, Nyaruaba

, et al. Application of adaptive evolution to improve the stability of bacteriophages during storage. Viruses 2020;12(4):423; doi: 10.3390/v12040423

84.

84. Burrowes

, Molineux

, Fralick

. Directed in vitro evolution of therapeutic bacteriophages. The Appelmans Protocol 2019.

85.

85. Jaishankar

, Srivastava

. Molecular basis of stationary phase survival and applications. Front Microbiol 2017;8:2000; doi: 10.3389/fmicb.2017.02000

86.

86. Ghosh

, Barman

, Mandal

. Phosphate deficiency induced biofilm formation of Burkholderia on insoluble phosphate granules plays a pivotal role for maximum release of soluble phosphate. Sci Rep 2019;9(1):5477; doi: 10.1038/s41598-019-41726-9

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

4.12 MB

0.04 MB