The Linear Double-Stranded DNA of Phage Bam35 Enters Lysogenic Host Cells, but the Late Phage Functions Are Suppressed

Bacteria and phage cultivation.The bacterial strains used in this study are listed in Table 1. B. thuringiensis cells were cultivated in Luria-Bertani broth (LB) (35). Lysogenic or resistant cell lines were obtained by isolating microcolonies from the centers of plaques and by isolating colonies from confluently lysed plates of B. thuringiensis serovar israelensis HER1410 infected with a clear-plaque variant of Bam35 (29). Remaining free phage particles were eliminated through eight consecutive single-colony purification steps. The presence of Bam35 DNA in the cells was verified both by isolating total DNA from bacterial cells (Wizard genomic DNA purification kit; Promega) and by PCR on DNA from single bacterial colonies (20) with specific primers hybridizing to the ends of Bam35 ORF18 encoding the major coat protein (29).

Lysogenization and resistance frequencies.HER1410 cells (2 × 10⁸) in a liquid culture were infected (multiplicity of infection [MOI], 3 or 30), incubated with aeration at 37°C for 10 min, and plated. Survivors (from 1/40 of the infection mixture) were isolated and tested for their ability to form plaques and to release virions spontaneously and for the presence of the 15-kb DNA Bam35 genome.

Cell flocculation assay.To test cell flocculation, cells from a single colony were grown overnight at 37°C with aeration, followed by incubation at room temperature for approximately 20 h without aeration. The flocculation level was estimated visually.

Detection of spontaneously released virions.To detect spontaneously released viruses, cells were diluted to 5 × 10⁷ CFU/ml and grown at 37°C for 3 h. The supernatants of the cell cultures were plated on HER1410 lawns.

Phage adsorption test.The phage adsorption test was performed by mixing 300 to 600 phage particles with 1.2 × 10⁷ cells grown to a cell density of ∼2 × 10⁸ CFU/ml. The infection mixture was incubated for 10 min at 37°C with aeration, and the cells were removed by centrifugation (microcentrifuge at 12,000 rpm for 3 min at 22°C), followed by washing with LB. The number of nonadsorbed phage particles was determined by plating the supernatants on HER1410 lawns.

Evaluation of premature cell lysis.To test the premature lysis of infected cells, the cells were grown at 37°C to ∼2 × 10⁸ CFU/ml and infected with Bam35 using an MOI of ∼20. At different time points postinfection, a metabolic inhibitor (sodium azide, sodium fluoride, sodium arsenate, or potassium cyanide) was added to a final concentration of 20 mM. The culture turbidity was followed using a Klett-Summerson colorimeter (A₅₄₀).

Measurements of ion fluxes.Ion flux measurements were performed as described previously (13), with the following modifications. The experiments were carried out simultaneously in two 5-ml thermostat-regulated (37°C) vessels with aeration. The concentrations of K⁺ and TPP⁺ ions in the medium were monitored by selective electrodes. To describe the early changes in PM permeability of Bam35-infected cells, the cells were grown to 2 × 10⁸ CFU/ml, collected by centrifugation, resuspended in LB (pH 8.0; pH adjusted with NaOH) to obtain 1/50 of the original volume, and kept on ice until used (maximum, 4 h). The concentrated cell suspension (200 μl) was added to 5 ml LB (for TPP⁺ measurements, the LB contained 3 μM TPP⁺), incubated for 5 min, and infected with freshly made Bam35 stock. The electrodes were calibrated by adding an appropriate amount of TPP⁺ or K⁺ at the beginning (TPP⁺) or at the end (K⁺) of every experiment. To determine the amount of intracellular K⁺ and TPP⁺, the PM was permeabilized by the addition of gramicidin D (GD) (4 μg/ml).

To describe the late changes in PM permeability of Bam35-infected cells, 400 ml of the bacterial culture (grown to ∼2 × 10⁸ CFU/ml at 37°C) was infected with Bam35 (MOI, ∼10) and incubated at 37°C with aeration. At each time point, two 5-ml samples were taken and transferred into the vessels. TPP⁺ was added to a final concentration of 3 μM (in TPP⁺ measurements only) and, after stabilization of the electrode potential, GD was added. Both K⁺ and TPP⁺ electrodes were calibrated at the end of every experiment. The K⁺ concentration in LB was estimated from calibration curves.

Characteristics of TPP⁺-selective electrodes have been described previously (13, 17). The electrodes were connected to the electrode potential-amplifying system, based on an ultralow input bias current operational amplifier (AD549JH; Analog Devices). The amplifying system was connected to a computer through the data acquisition board AD302 (Data Translation, Inc., Marlboro, Mass.). TPP⁺ chloride and GD were purchased from Sigma.

Electron microscopy.For thin-section electron microscopy, B. thuringiensis cells were grown in LB at 37°C to a density of ∼2 × 10⁸ CFU/ml and infected with Bam35 using an MOI of 12. Samples were collected at different time points after infection and fixed with 3% (vol/vol) glutaraldehyde. As a control, noninfected bacteria were fixed at a cell density of 4 × 10⁸ CFU/ml. After 20 min of fixation at room temperature, cells were collected, washed twice with 20 mM potassium phosphate buffer (pH 7.2), and prepared for transmission electron microscopy as previously described (3). The electron micrographs were taken with a JEOL 1200 EXII electron microscope (at the EM unit, Institute of Biotechnology, University of Helsinki) operating at 60 kV.

DNA sequencing.The DNA for sequencing was amplified by PCR using Bam35-specific primers and native Pfu polymerase (MBI Fermentas). The DNA templates for amplification reactions were prepared from colonies and plaques as described elsewhere (18, 20). PCR products (about 1.8 kb) were purified using microcentrifuge spin columns (QIAquick PCR Purification Kit; QIAGEN), and both strands of the DNA were sequenced (Sequencing Laboratory, Institute of Biotechnology, University of Helsinki).

Alternative hosts for Bam35.Bam35 is a Bacillus-specific phage and can lyse Bacillus megaterium strain Tiberius and some strains of Bacillus cereus and B. thuringiensis (2). In order to test the sensitivity of B. thuringiensis strains other than HER1410 to Bam35, we used a plaque assay to screen 14 strains obtained from the Bacillus Genetic Stock Center. Only one alternative host strain in addition to HER1410 was found. B. thuringiensis serovar kurstaki 4D22 was able to support Bam35 propagation with the same plating efficiency as HER1410 but gave more turbid plaques. In liquid cultures, the titer of Bam35-infected 4D22 cell lysates was about 35% of that of HER1410. In addition, there were differences in Bam35 adsorption to HER1410 and 4D22 hosts (Table 2).

Bam35-resistant and lysogenic HER1410 cells.Liquid culture cells surviving Bam35 infection were isolated. Plaque assay, phage adsorption, and spontaneous plaque production tests were carried out. Bam35-resistant and lysogenic HER1410 colonies appeared with approximately similar frequencies (Table 3). Three cell lines (HER1410_R19, HER1410_L3, and HER1410_L5) were chosen for further analyses (Table 2).HER1410_R19 did not support virus propagation or produce virus particles spontaneously and did not adsorb Bam35 particles. PCR from single colonies showed no Bam35-specific DNA. Consequently, HER1410_R19 is a genuine Bam35-resistant strain. HER1410_L3 and HER1410_L5 did not support virus propagation (plaque assays) but adsorbed Bam35 with high efficiency (Table 2) and released small amounts of virus into culture supernatants. PCR from single colonies and the sequence analysis revealed Bam35-specific DNA in these cells. Purified DNA from HER1410_L3 and HER1410_L5 strains contained ∼15-kb DNA elements that were not found in the original (HER1410) or resistant strains (not shown, but see previous work, where this phenomenon is described for HER1410_L5 strain [37]). Consequently, HER1410_L3 and HER1410_L5 are lysogenic strains carrying Bam35 prophages.

Bam35 entry induces depolarization of the PM.Concentration of the Δψ indicator TPP⁺ in the medium was registered during the first 10 min of Bam35 infection (Fig. 1). Rapid release of HER1410 cell-accumulated TPP⁺ (PM depolarization) began 1.5 to 2 min after phage addition and was followed by a slow uptake of TPP⁺ (PM repolarization) starting 4 to 5 min postinfection (Fig. 1A). Interestingly, the same time course of TPP⁺ ion movements was observed when lysogenic HER1410_L3 cells were infected. In the case of serovar kurstaki (4D22), Bam35 induced slow leakage of accumulated TPP⁺ that began about 1.5 to 2 min postinfection but subsequent PM repolarization was not observed (Fig. 1B). The time course of TPP⁺ accumulation in nonadsorbing derivative cells (HER1410_R19) infected with Bam35 was the same as in the case of noninfected HER1410 cells. The initial drop in TPP⁺ concentration in this experiment was due to TPP⁺ binding by the input virus and the increased volume of infection mixture (insert in Fig. 1A). The phage-induced efflux of accumulated TPP⁺ was MOI dependent (Fig. 1C).

• Open in new tab
• Download powerpoint

FIG. 1.

Effects of Bam35 infection on TPP⁺ (A and C) and K⁺ (B and D) ion fluxes across the PM of different B. thuringiensis strains and cell lines during the first 10 min of infection. The experiments were carried out at 37°C in LB (pH 8.0) containing 3 μM TPP⁺ (in panels A and C). Approximately 4 × 10⁸ CFU/ml cells were infected with Bam35 using an MOI of 20 (A and B) or as indicated below (C and D). GD was added to a final concentration of 4 μg/ml. Experiments were done with B. thuringiensis strains HER1410 (black, infected; yellow, noninfected) and 4D22 (blue) and cell lines HER1410_L3 (red) and HER1410_R19 (green) (A and B). HER1410 cells were infected with Bam35 using MOIs of 16 (red), 4 (blue), and 2 (green) (C and D). The insert in panel A indicates the decrease in TPP⁺ concentration due to TPP⁺ binding and dilution by the added phage suspension.

Bam35 entry induces K⁺ leakage.The addition of Bam35 particles to the wild-type (HER1410) and lysogenic (HER1410_L3) cell suspensions resulted in the rapid efflux of K⁺ ions starting approximately 2 min postinfection and lasting approximately 2 min (Fig. 1B). The magnitude of phage-induced K⁺ leakage from 4D22 cells was about one-third of that observed with HER1410 cells, and the time course was longer (approximately 3.5 min). There was neither a significant change in K⁺ concentration in the medium after addition of phage particles to the resistant (HER1410_R19) cells nor a spontaneous leakage of K⁺ from noninfected HER1410 cells detected. The magnitude of the phage-induced K⁺ efflux was consistent with the phage effect on Δψ of each host strain. As in the case of TPP⁺ accumulation, the magnitude of K⁺ leakage was MOI dependent (Fig. 1D).

Late physiological changes in Bam35-infected cells.Bam35-induced changes in PM permeability late in the infection cycle were followed by sampling throughout one-step growth experiments. An increase of TPP⁺ accumulation by infected HER1410 cells was detected during the first half of the infection cycle. Accumulation of K⁺ by infected cells also was observed during the first 28 min of infection (Fig. 2A). Strong efflux of K⁺ from infected cells started about 20 min before any decrease in the turbidity of the cell suspension was detected (50 min postinfection), and there was no measurable K⁺ gradient on the PM at the time of lysis. In addition, when the K⁺ leakage began, Δψ (or TPP⁺ accumulation) dropped rapidly and remained at approximately half of the initial level at the time of cell lysis. As a control, increases of TPP⁺ and K⁺ accumulation by noninfected HER1410 cells were observed due to the growth of the culture (Fig. 2B). The addition of Bam35 particles to lysogenic (HER1410_L3) cells did not result in cell lysis or cause measurable late K⁺ leakage or a decrease in TPP⁺ accumulation (Fig. 2C). However, Bam35 entry induced early depolarization of the PM and the early efflux of K⁺ (Fig. 1A and B). Electrochemical parameters of noninfected lysogenic cells (Fig. 2D) did not differ considerably from those of the infected cells. This result shows that the majority of infected lysogenic cells do not lyse and consequently do not release progeny viruses.

• Open in new tab
• Download powerpoint

FIG. 2.

Changes in B. thuringiensis HER1410 and HER1410_L3 cell energetics during Bam35 infection. The cells were grown in a large volume (400 ml) and infected at time zero (MOI, ∼10). Two 5-ml samples of cell suspension were withdrawn at each time point, and the amounts of intracellular K⁺ and accumulated TPP⁺ were measured (see Materials and Methods). Infected HER1410 cells, A; noninfected HER1410 cells, B; infected HER1410_L3 cells, C; noninfected HER1410_L3 cells, D. Symbols: K⁺, closed circles; TPP⁺, open triangles; and OD, open circles.

Lysogenic cells produce viruses sporadically.Bam35-infected and noninfected lysogenic cells were subjected to thin-section electron microscopy. Electron micrographs (Fig. 3) of Bam35-infected lysogenic (HER1410_L3) cells revealed both empty and filled virus particles associated with the cell surface, confirming the unaltered adsorption (Table 2). No virus particles were observed on the surface of HER1410_R19 cells. This indicates that there are no active receptors on the resistant cell surface. Occasional lysing cells were observed when infected lysogenic bacteria (HER1410_L3) were examined. The titer of infected lysogenic culture supernatants was low. However, the number of virus particles produced per induced cell did not deviate considerably from normal infection (Fig. 3 C and E), as observed using electron microscopy. Interestingly, rare lysis events of noninfected lysogenic (HER1410_L3) cells producing only a few virus particles were also captured. These observations were consistent with the results obtained with plaque assays.

• Open in new tab
• Download powerpoint

FIG. 3.

Thin-section electron micrographs of Bam35-infected (A, C, D, and E) or noninfected (B) B. thuringiensis cells. Lysogenic cell HER1410_L3 with surface-associated virus particles (A). Partially lysed noninfected HER1410_L3 cell; a few virus particles are visible in the cell interior (B). Progeny viruses observed in HER1410_L3 cell at 55 min postinfection (C). Resistant cell HER1410_R19 after Bam35 addition; no bound viruses are seen (D). Partially lysed HER1410 cell with progeny viruses 40 min postinfection (E). Black arrows mark DNA-containing virus particles, and white arrow points to an empty particle. Bar represents 200 nm in all panels.

Bam35 infection induces holin-type effects in the host cell.Premature lysis of infected cells caused by the addition of metabolic inhibitors has been proven to be a hallmark of the endolysin-holin lysis systems operating in many bacteriophage infections (Escherichia coli phage λ [4, 30], gram-negative bacterium-infecting phage PRD1 [33, 46], and Bacillus subtilis phage φ29 [36]). We tested whether the addition of metabolic inhibitors to Bam35-infected cells could trigger premature lysis. Sodium azide (inhibitor of cytochrome oxidase and membrane H⁺-ATPase), potassium cyanide (inhibitor of cytochrome oxidase), or sodium arsenate (reduces the concentration of intracellular ATP by arsenolyzation of acetyl phosphates) inhibited growth but did not induce lysis of noninfected cells (shown for azide in Fig. 4). These poisons, added at the proper time, induced premature lysis of infected HER1410cells (Fig. 4A). Normally, Bam35-infected HER1410 cells lysed at approximately 50 min postinfection. Sodium fluoride, an inhibitor of the enolase reaction, did not inhibit growth of noninfected cells or induce premature lysis of infected cells. Timing experiments with sodium azide, sodium arsenate, or potassium cyanide revealed that these poisons induce premature lysis if added at 35 min postinfection or later (not shown). The addition of arsenate or azide to 4D22 cells at 40 min postinfection also resulted in premature lysis (not shown). Premature lysis inducers had no effect on both infected and noninfected resistant cells (HER1410_R19, shown for azide in Fig. 4B) other than growth inhibition. Growth inhibition of infected lysogenic (HER1410_L3 or HER1410_L5) cells could be detected at approximately 60 to 110 min postinfection. However, infected (as well as noninfected) HER1410_L3 and HER1410_L5 cells did not lyse after the addition of the metabolic inhibitor (shown for HER1410_L3 and azide in Fig. 4C).

• Open in new tab
• Download powerpoint

FIG. 4.

Effect of metabolic inhibitors on infected or noninfected B. thuringiensis strains HER1410 (A), HER1410_R19 (B), and HER1410_L3 (C). Cells were grown to a cell density of ∼2 × 10⁸ CFU/ml (37°C) and infected at time zero using an MOI of ∼20. Noninfected cells: no metabolic inhibitors added (closed circles) or NaN₃ added (open circles). Bam35-infected cells: no metabolic inhibitors added (closed triangles), NaN₃ added (open triangles), KCN added (closed squares), arsenate added (open squares), or NaF added (closed diamonds). The arrows indicate the time point of addition of inhibitors.