Single-Cell Approach Reveals Intercellular Heterogeneity in Phage-Producing Capacities

ABSTRACT Bacteriophage burst size is the average number of phage virions released from infected bacterial cells, and its magnitude depends on the duration of an intracellular progeny accumulation phase. Burst size is often measured at the population level, not the single-cell level, and consequently, statistical moments are not commonly available. In this study, we estimated the bacteriophage lambda (λ) single-cell burst size mean and variance following different intracellular accumulation period durations by employing Escherichia coli lysogens bearing lysis-deficient λ prophages. Single lysogens can be isolated and chemically lysed at desired times following prophage induction to quantify progeny intracellular accumulation within individual cells. Our data showed that λ phage burst size initially increased exponentially with increased lysis time (i.e., period between induction and chemical lysis) and then saturated at longer lysis times. We also demonstrated that cell-to-cell variation, or “noise,” in lysis timing did not contribute significantly to burst size noise. The burst size noise remained constant with increasing mean burst size. The most likely explanation for the experimentally observed constant burst size noise was that cell-to-cell differences in burst size originated from intercellular heterogeneity in cellular capacities to produce phages. The mean burst size measured at different lysis times was positively correlated to cell volume, which may determine the cellular phage production capacity. However, experiments controlling for cell size indicated that there are other factors in addition to cell size that determine this cellular capacity. IMPORTANCE Phages produce offspring by hijacking a cell's replicative machinery. Previously, it was noted that the variation in the number of phages produced by single infected cells far exceeded cell size variation. It was hypothesized that this variation is a consequence of variation in the timing of host cell lysis. Here, we show that cell-to-cell variation in lysis timing does not significantly contribute to the burst size variation. We suggest that the constant burst size variation across different host lysis times results from cell-to-cell differences in capacity to produce phages. We found that the mean burst size measured at different lysis times was positively correlated to cell volume, which may determine the cellular phage production capacity. However, experiments controlling for cell size indicated that there are other factors in addition to cell size that determine this cellular capacity.


Effect of experimental conditions on phage stability
Aliquots of phage λSam7 lysate (200 µl) were dispensed into the wells of a 96-well plate to simulate the conditions used in the burst size assay (for details see Methods). The treatment involved exposing the lysate to chloroform for 20 minutes, which included 10 min of shaking at 37° C inside the plate reader followed by 10 min incubation at room temperature without shaking. A 100 µl aliquot from each well was thereafter removed for plaque assays and the estimated plaque forming units were compared to a control group that was plated without the treatment. Figure S1. Chloroform treatment in 96-well plates reduces phage titer. Experimental conditions used in the burst size assay reduced phage titers by 19% when compared to a control (ctrl) group (p < 0.001; t test; n = 32). Error bars, mean ± SEM.

Cell volume measurements
We measured cell dimensions using a simple microscopic setup described previously (Dennehy and Wang, 2011). Briefly, the lysis-deficient lysogen was grown to A600 = 0.3-0.4 at the permissive temperature of 30°C. The cells were then immobilized to a 22 mm glass coverslip that was pretreated with 0.01% poly-L-lysine (mol. wt. 150 K-300 K; Millipore Sigma, St. Louis, MO) for 10 min. The coverslip was assembled as a perfusion chamber (RC-21B, Warner Instruments, New Haven, CT) and attached to a heated platform (PM2; Warner Instruments, New Haven, CT). The platform was then mounted onto an inverted microscope stage (TS100, Nikon, Melville, NY). The perfusion chamber was infused with heated LB at 30°C (Inline heater: SH-27B, dual channel heating controller: TC-344B; Warner Instruments, New Haven, CT). The lytic cycle was induced by raising the chamber temperature to 42°C for 20 min. After heat shock, the temperature was maintained at 37°C. The cell growth was captured as a video using an eyepiece camera (10× MiniVID TM ; LW Scientific, Norcross, GA, 10 fps). Video screenshots taken at different times were used as input files for ImageJ software (Schneider, Rasband and Eliceiri, 2012), which was used to measure cell lengths and widths. A micrometer slide was used to set the scale (5.0887 pixels/µm) in ImageJ. The volume (V) of each cell was approximated as the volume of a capsule with cylindrical radius equal to the radii of the hemispheres on either ends: = .

Single-cell burst size determined using FACSAria cell sorting
To prepare cells for fluorescence assisted cell sorting (FACS), single colonies of E. coli lysogen MC4100 (λ cI857 S105) were placed in 10 mL LB medium and cultured overnight at 30° C in a water bath shaker rotating at 250 rpm. Stationary phase cultures were spun at 3,000 rpm for 10 min, and the pellets were resuspended in 3 mL fresh LB. Subsequently, 30 μL fluorescein isothiocyanate conjugate (FITC; Virostat) was added and the cultures were returned to the water bath shaker. After 1 hr, free FITC was removed by centrifuging the cultures at 3,000 rpm for 5 min and adding 3 mL phosphate-buffered saline (PBS) to the pellet. The wash step was repeated for a total of three washes.
Cell sorting was performed using a FACSAria instrument (BD Biosciences) equipped with a 70 μm nozzle and a blue 488 nm laser. PBS was used as a sheath fluid. A preliminary analysis indicated that cells fell into two subpopulations based on the amount of laser excitation energy released and detected by photomultiplier tubes (PMT2). Since fluorescence intensity is proportional to cell size, gating was used to select cells that fell within a narrow range within the subpopulation expressing the lower amount of fluorescence intensity to minimize variation according to cell size.
This strategy was used to sort single-cells into wells of a round bottom 96-well plate (Corning) where each well contained 100 μL LB medium. The plates were placed on ice until being placed in a 42° C incubator for 15 min for induction of lysis. The plate was then incubated in a 37° C incubator for 1.5 hrs. Subsequently, 100 μL of an exponential culture (OD600 = 0.2) of E. coli MC4100 was added to each well and the cultures were allowed to rest for 15 min for phage adsorption. The entire contents of the well were then removed and added to 3 mL LB top agar, which was then poured on 35 mL LB bottom agar and incubated overnight at 37° C to visualize plaques. Figure S5. Burst size distribution of wild type phage λ. Following thermal induction, FACS ARIA was used to sort single FITC-labeled lysogens into wells of a 96-well plate. Cell detection was tuned to a very narrow expression profile to ensure homogeneity in cell size. Phages released from lysed cells were enumerated using plaque assays.