1Department of Physics, University of Illinois at Urbana-Champaign, 2Center for the Physics of Living Cells, University of Illinois at Urbana-Champaign, 3Verna and Marrs McLean Department of Biochemistry and Molecular Biology, Baylor College of Medicine
Zeng, L., Golding, I. Following Cell-fate in E. coli After Infection by Phage Lambda. J. Vis. Exp. (56), e3363, doi:10.3791/3363 (2011).
The system comprising bacteriophage (phage) lambda and the bacterium E. coli has long served as a paradigm for cell-fate determination1,2. Following the simultaneous infection of the cell by a number of phages, one of two pathways is chosen: lytic (virulent) or lysogenic (dormant)3,4. We recently developed a method for fluorescently labeling individual phages, and were able to examine the post-infection decision in real-time under the microscope, at the level of individual phages and cells5. Here, we describe the full procedure for performing the infection experiments described in our earlier work5. This includes the creation of fluorescent phages, infection of the cells, imaging under the microscope and data analysis. The fluorescent phage is a "hybrid", co-expressing wild- type and YFP-fusion versions of the capsid gpD protein. A crude phage lysate is first obtained by inducing a lysogen of the gpD-EYFP (Enhanced Yellow Fluorescent Protein) phage, harboring a plasmid expressing wild type gpD. A series of purification steps are then performed, followed by DAPI-labeling and imaging under the microscope. This is done in order to verify the uniformity, DNA packaging efficiency, fluorescence signal and structural stability of the phage stock. The initial adsorption of phages to bacteria is performed on ice, then followed by a short incubation at 35°C to trigger viral DNA injection6. The phage/bacteria mixture is then moved to the surface of a thin nutrient agar slab, covered with a coverslip and imaged under an epifluorescence microscope. The post-infection process is followed for 4 hr, at 10 min interval. Multiple stage positions are tracked such that ~100 cell infections can be traced in a single experiment. At each position and time point, images are acquired in the phase-contrast and red and green fluorescent channels. The phase-contrast image is used later for automated cell recognition while the fluorescent channels are used to characterize the infection outcome: production of new fluorescent phages (green) followed by cell lysis, or expression of lysogeny factors (red) followed by resumed cell growth and division. The acquired time-lapse movies are processed using a combination of manual and automated methods. Data analysis results in the identification of infection parameters for each infection event (e.g. number and positions of infecting phages) as well as infection outcome (lysis/lysogeny). Additional parameters can be extracted if desired.
1. Creation of a crude phage lysate (Figure 1)
2. Phage purification (Figure 1)
3. Prepare one agarose gel slab (Figure 4)
4. Testing the purified phage stock
5. Infection (Figure 6)
6. Following cell fate under the microscope
7. Image analysis
8. Representative results:
The plaques of the fluorescently labeled phages (in Step 1.6 and Section 2) are significantly smaller than those of wild type (Figure 2). We therefore incubate the plates at least 12 hr in 37°C incubator for the plaques to be visible.
After ultracentrifugation of the phage sample with the CsCl step gradient (Step 2.10), two bands should be visible (Figure 3A). The top band, at the interface between the phage suspension and SM/CsCl 1.3 g/ml layer, contains cell debris and empty phage capsids. The bottom band, at the interface between SM/CsCl 1.3 g/ml and 1.5 g/ml layers, is the phage band. This band appears greenish for the fluorescent phage λLZ2. The band for wild type phage λIG2903 appears bluish5. After the ultracentrifugation of CsCl equilibrium gradient in Step 2.12, one phage band should be visible in the middle part of the tube (Figure 3B). Since the fluorescent phage λLZ2 contains a mixture of gpD-EYFP and gpD capsids, the ratio of protein-to-DNA is higher than that of wild type. Therefore, the band of the fluorescent phage λLZ2 is slightly lighter (appears to be at a higher location in the tube) than that of wild type λIG290310.
Figure 5 shows typical images obtained after labeling the phage with DAPI (Section 4). The YFP and DAPI signals of a successfully purified phage should have close to 100% correspondence. We typically observe that less than 1% of the YFP spots do not contain DAPI (representing capsids without the viral genome). Less than 1% of the DAPI spots do not contain YFP (corresponding to non-fluorescent phages)5.
Lytic cells are recognized by the accumulation of YFP fluorescence (green channel) inside the cell, followed by cell lysis. Lysogenic cells are recognized by the accumulation of uniform mCherry fluorescence (red) inside the cell and the resumption of normal cell growth and division. Uninfected cells (or cells where infection has failed) will not display any of the phenotypes above and will grow and divide normally. Figure 7 shows a few image-sets of phase-contrast, YFP and mCherry channels, and the corresponding overlaid images of these three channels, from a typical time-lapse movie (Section 6). The individual phages (green spots) are clearly visible at the initial time frame (Figure 7A). Typically, a number of phages are seen on the cell surface (presumably infecting those cells) while other phages are unadsorbed, as shown in Figure 7B (left panel). The infection outcome becomes distinguishable over time. The lytic cycle is indicated by the intracellular production of new phages (green, Figure 7C) followed by cell lysis (exploded cells with released green phages, Figure 7D). Lysogeny is indicated by the production of mCherry from the PRE promoter (red, Figure 7C) and the resumption of cell growth and division (red, Figure 7D).
Figure 1. Flow chart describing the creation of fluorescent phages. A crude phage lysate is first obtained by inducing a lysogen of the gpD-EYFP phage, harboring a plasmid expressing wild type gpD protein (panels A-B). The phage is purified through a series of steps (panels C-L).
Figure 2. Phage plaques. Plaques of the fluorescent phage (left) are smaller than those of wild type (right) after incubating plates for 12 hr at 37°C.
Figure 3. Phage bands after ultracentrifugation. A) Two bands are visible after ultracentrifugation in a CsCl step gradient. The top one corresponds to cell debris and empty phage capsids; the bottom band contains the desired phage. Left: fluorescent phage, right: wild type. B) A single phage band is visible after ultracentrifugation in a CsCl equilibrium gradient. The fluorescent phage band (left) is greenish, compared to a bluish band for wild type phage (right).
Figure 4. The procedure of preparing agarose gel slabs.
Figure 5. Fluorescent images of phages after DAPI staining. Individual phages are easily distinguishable, and YFP and DAPI signals co-localize very well.
Figure 6. Schematic description of phage infection and imaging setup. Click here to view a full-sized version of this image.
Figure 7. Typical images from a time-lapse movie of phage infection. Shown are the phase-contrast, YFP and mCherry channels, as well as an overlay of the three channels. (A) YFP-channel images from the initial time frame. Left, the sum of YFP images at different z-positions. The three right images are sample YFP images at different z-positions, corresponding to different areas of the cell surface. (B), (C) and (D) Overlaid images (left) of the phase-contrast (middle-left), YFP (middle-right) and mCherry (right) channels at different time frames. (B) At t = 0, two cells are seen, each infected by a single phage (green spots), and one cell is infected by 3 phages. Also observed are some unadsorbed phages. (C) At t = 80 min, the two cells infected by single phages have each gone into the lytic pathway, as indicated by the intracellular production of new phages (green). The cell infected by 3 phages has gone into the lysogenic pathway, as indicated by the production of mCherry from the PRE promoter (red). (D) At t = 2 hr, the lytic pathway has resulted in cell lysis (cell exploded), while the lysogenic cell has divided§.
§Left panels of Figure 7(C) and (D) are reprinted from Cell, 141, Lanying Zeng, Samuel O. Skinner, Chenghang Zong, Jean Sippy, Michael Feiss, and Ido Golding, Decision Making at a Subcellular Level Determines the Outcome of Bacteriophage Infection, 682-691, Copyright (2010), with permission from Elsevier.
|Strain name||Relevant genotype||Source/reference|
|LE392||supF||John Cronan, University of Illinois|
|λLZ1||gpD-EYFP, cI857 Sam7 D-eyfp b::kanR||Zeng et al.5|
|λLZ2||gpD-mosaic, same genotype as λLZ1||Zeng et al.5|
|pPRE-mCherry||mCherry under the control of PRE, ampR||Zeng et al.5|
|pPLate*D||gpD under the control of λ late promoter, ampR||Zeng et al.5|
Table 1. Bacterial strains, phages and plasmids used in this work.
|Density ρ (g/ml)||CsCl (g)||SM (ml)||Refractive Index η|
Table 3. CsCl solutions prepared in SM buffer (100 ml) for step gradients.
Bacterial Strains, Phage and Plasmids:
Strain LE392 is supF. It was chosen to suppress the Sam7 mutation in the phage genome (see Table 1 for details). Thus, induced lysogens will eventually lyse and release phage particles, as will infected cells that have chosen the lytic pathway. Lysogenic cells are grown at 30°C due to the presence of the temperature sensitive cI857 allele in the phage genome. After heat induction, gpD-EYFP and wild type gpD are co-expressed from the genome of λLZ1 and the plasmid pPlate*D respectively. As a result, the capsid of the newly created phage λLZ2 contains a mixture of gpD-EYFP and gpD proteins. This mosaic phage is structurally stable and sufficiently fluorescent to allow detection of individual phages5. pPRE-mCherry is a reporter plasmid used to detect choice of the lysogenic pathway. The promoter PRE is activated by CII during the establishment of lysogeny1,11. pPRE-mCherry5 was derived from pE-gfp11 by replacing gfp with mCherry12. For more details see our earlier work5.
Growth Condition Parameters:
During lysogen induction (Section 1), mild shaking at 180 rpm gives a good virus yield13. Use of glucose in the growth medium should be avoided as glucose metabolism generates acidic metabolic products, and mature lambda particles are unstable at acidic pH13. The addition of MgSO4 is aimed at stabilizing the phage capsid3. For phages carrying wild type cI (instead of cI857), the lysogen can be induced using the DNA-damaging agent Mitomycin C3. In Step 1.3, the incubation at 37°C should normally not exceed 90 minutes. It is useful to check the cell density by OD600 every 30 min. For a good lysate, OD600 drops to around 0.2 or less, and the remaining OD600 is a result of cell debris. Incubating too long may result in a lower phage yield since the newly created phage may start to inject their DNA into cell debris. To obtain a visible phage band (at least 1 x 1011 phage particles) in Steps 2.11 and 2.13, grow at least 500 ml culture in Step 1.2. The addition of 0.2% Maltose into the growth medium in Steps 5.1 and 5.2 is aimed at inducing expression of LamB, the receptor for phage lambda adsorption3,14. The 1000-fold dilution instead of 100-fold in Step 5.2 is aimed at reducing the mCherry background level from the reporter plasmid pPRE-mCherry. In Step 5.5 for phage DNA injection triggering, 35°C is chosen to avoid induction of the temperature-sensitive cI857 allele.
The phage purification steps (Steps 2.1 through 2.11) can be replaced with other purification protocols5, but the final ultracentrifugation through CsCl equilibrium gradient (Steps 2.12 and 2.13) is unavoidable. Swinging bucket rotors are needed in Steps 2.10 and 2.12 to ensure sharp visible phage bands. Obtaining a pure phage stock can easily take up to a week, so it is necessary to check the phage titer along the way to make sure nothing goes wrong during the intermediate steps.
During all purification procedures in Section 2, it is critical to handle phage lysate gently to avoid shearing phage tails from phage heads. During cell infection in Section 5 (e.g., Steps 5.5 through 5.7), it is also critical to avoid the shearing of phage particles from the infected cell. Note that if the phage is sheared from the infected cell after injecting its DNA, the result is a "dark" infection, i.e. the infection outcome will be observed in experiment but the infecting phage will not. To minimize such problems, we use a wide pipette tip whenever handling phages or the phage/cell mixture.
Staining the phage stock with DAPI (Section 4) is a quick and efficient method to check the purity of the phage stock. It can also be used to test for possible degradation of an existing phage stock over time. For a pure stock, the co-localization of YFP and DAPI signals under the fluorescence microscope should be close to 100%. We typically observe that less than 1% of the YFP spots do not contain DAPI (representing capsids without the viral genome), which indicates that these particles did not successfully package the viral DNA or had already injected their DNA elsewhere. Less than 1% of the DAPI spots do not contain YFP (corresponding to non-fluorescent phages). If this is not the case, Steps 2.12 through 2.14 need to be repeated in order to purify again. With regards to imaging parameters, the microscope setup in Step 4.3 is not as critical as in Section 5 because no long-term live-cell imaging is required here. However, keeping the same microscopy settings as in Section 5 is useful if one wishes to calibrate the fluorescence intensity of a single phage particle. If the PBS-agarose slab is not very clean, or too much DAPI dye is used, some DAPI spots corresponding to phage DNA may be surrounded with a "halo". If too little DAPI dye is used, the signal from the DAPI channel may be very weak.
For the imaging in Section 6, we use a commercial inverted epifluorescence microscope (Eclipse TE2000-E, Nikon) with a 100x objective (Plan Fluo, numerical aperture 1.40, oil immersion) and standard filter sets (Nikon). The fluorescence light source is an Arc lamp with control of light intensity. The following features are computer controlled: x, y and z position; bright field and fluorescence shutters; and fluorescence filter choice. An auto-focus feature is required. Otherwise, the focus may easily drift away during the time-lapse movie (normally 4 hours long). The ability to acquire multiple (x,y) positions at each time point is useful, as it allows to follow multiple infection events in parallel. We typically acquire 8 stage positions in each movie, following up to 100 infection events. The camera we use is a cooled 512x512 CCD with 16x16 μm pixel camera with a dynamic range of 16 bits (Cascade512, Photometrics). Acquisition is performed using MetaMorph software (Molecular Devices). The microscope should be placed in a temperature-controlled room; alternatively, the microscope stage should be surrounded by a temperature-controlled chamber.
For live-cell imaging, it is critical to avoid unnecessary exposure of the sample, which could lead to bleaching and phototoxicity. Therefore, it is best to first characterize your system to find an optimal light exposure which allows for fluorescence detection while not leading to excessive bleaching or inhibiting cell growth. To obtain a good fluorescence image, play with the exciting light intensity, exposure time and camera gain. In Steps 6.2-6.3, the 10 min frame interval is chosen for the purpose of minimizing light exposure. In each frame, only a single in-focus image is needed in phase-contrast (for cell recognition) and fluorescent channels (for determining cell fate). In the first time point, however, multiple z-position images through the YFP channel are required to capture all infecting phages on the cell surface. The YFP exposure time in the initial frame may also need to be higher than that used for the time-lapse movie in the later time frames.
Very carefully count phage particles around the cell surface in Step 7.1. As noted above, we take a series of z-stacks through YFP channel in Step 6.2. However, this may still leave some fluorescent phage particles out-of-focus, which challenges the counting. The cell length in the initial time frame is measured using the Metamorph software. The cell length can also be measured by ImageJ or other software tools. Additionally, an automated home built Matlab program can be very useful in obtaining information such as fluorescence change over time along cell lineages.
No conflicts of interest declared.
We are grateful to Michael Feiss and Jean Sippy for the guidance on phage creation and purification. We thank Michael Elowitz for providing the cell recognition software, Schnitzcell. Work in the Golding lab is supported by grants from the National Institutes of Health (R01GM082837), the National Science Foundation (082265, PFC: Center for the Physics of Living Cells), the Welch Foundation (Grant Q-1759) and Human Frontier Science Program (RGY 70/2008).
|SM buffer||TEKnova, Inc.||S0249|
|Dialysis cassette||Thermo Fisher Scientific, Inc.||66333|
|SW40Ti ultra-clear tube||Beckman Coulter Inc.||344060|
|SW60Ti ultra-clear tube||Beckman Coulter Inc.||344062|
|SW40Ti rotor||Beckman Coulter Inc.||331302|
|SW60Ti rotor||Beckman Coulter Inc.||335649|
|Epifluorescence microscope||Nikon Instruments||Eclipse TE2000-E|
|Table 2. Reagents and equipment.|