Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Aug 11.
Published in final edited form as: Mol Microbiol. 2008 Mar 19;68(5):1107–1116. doi: 10.1111/j.1365-2958.2008.06205.x

Bacteriophage infection is targeted to cellular poles

Rotem Edgar 1,*, Assaf Rokney 2, Morgan Feeney 3, Szabolcs Semsey 1,4, Martin Kessel 5, Marcia B Goldberg 3, Sankar Adhya 1, Amos B Oppenheim 1,2,†
PMCID: PMC3740151  NIHMSID: NIHMS495747  PMID: 18363799

Summary

The poles of bacteria exhibit several specialized functions related to the mobilization of DNA and certain proteins. To monitor the infection of Escherichia coli cells by light microscopy, we developed procedures for the tagging of mature bacteriophages with quantum dots. Surprisingly, most of the infecting phages were found attached to the bacterial poles. This was true for a number of temperate and virulent phages of E. coli that use widely different receptors and for phages infecting Yersinia pseudotuberculosis and Vibrio cholerae. The infecting phages colocalized with the polar protein marker IcsA–GFP. ManY, an E. coli protein that is required for phage λ DNA injection, was found to localize to the bacterial poles as well. Furthermore, labelling of λ DNA during infection revealed that it is injected and replicated at the polar region of infection. The evolutionary benefits that lead to this remarkable preference for polar infections may be related to λ’s developmental decision as well as to the function of poles in the ability of bacterial cells to communicate with their environment and in gene regulation.

Introduction

Bacteriophages are among the smallest but most abundant organism on earth (~1031) (Suttle, 2005). For most phages, the tail mediates the anchoring of the phage to generally abundant bacterial outer membrane proteins that serve as specific receptors for their substrates. For example, the receptor for the temperate phage λ is the Escherichia coli maltoporin receptor LamB, which functions in amylomaltose uptake (Ryter et al., 1975; Szmelcman and Hofnung, 1975). The establishment of a stable phage–host interaction relays signals that allow injection of DNA from the phage capsid through the tail and into the host, leaving the empty capsid (head) attached to the cell surface. Following phage λ DNA injection, a decision between the lytic or lysogenic pathways of bacteriophage λ is made.

Our aim was to investigate the initial steps of binding (adsorption) and phage DNA injection. We designed a protocol that allows us to follow infection visually in living samples. We labelled different phages with quantum dots (QDots) and followed adsorption using a fluorescence microscope. The surprising results showed that at low multiplicities of infection (moi), phages preferentially adsorb, inject and replicate their DNA at the bacterial poles. This spatial preference was independent of host proteins, ManY and Pel, required for phage λ DNA injection. The significance of the pole, the binding of the phage to the pole and its implications in lytic-lysogenic decision are discussed.

Results and discussion

Labelling phage

Visualization of phages has been traditionally obtained by electron microscopy of stained preparations, following infections at high moi (see Ptashne, 2004). To this day, the small nanometer size of phages presents an obstacle for the direct observation of live infection. To visually monitor phage adsorption and follow the infection process, we developed a technique for phage labelling as described below. The method allows visualization of phage infection by light microscopy in real time, without interfering with the infection process. Phages are chemically decorated with biotin using a cross-linker (EZ-Link Sulpho-NHS-Biotin, PIERCE) that forms a stable amide bond with primary amines and has a long spacer arm to reduce steric hindrance. The biotinylated phage is then conjugated to streptavidin-coated QDots that are ~10–20 nm in size, smaller than tailed phage particles such as λ. We first confirmed cross-linking and binding of QDots to the λ phage using transmission electron microscopy (TEM) (Fig. 1A). QDots were detected on about 80% of the phages and the distribution of the QDots on the phage was not random. About 50% of the QDots were bound to the phage head, primarily on the half of the head that is closest to the tail. Additional 50% of QDots were bound to the tail.

Fig. 1.

Fig. 1

Open in a new tab

Various phages labelled with QDots bind to the polar regions of their host bacteria. A. Electron micrograph of QDots attached to phage that was chemically decorated with biotin using the cross-linker EZ-Link Sulpho-NHS-Biotin. QDots are marked with asterisks.

B. (i) λ phage labelled with QDots as in (A), infecting E. coli. (ii) Percentage of QDots-bound λ phage at the indicated relative distance on the bacteria cell, as observed in the fluorescence microscope.

C. Fluorescent micrographs of QDot-labelled (red) phages. The green colour shows GFP expressed from the late gene reporter fusion pR’tR’–GFP as a response to effective infection by QDot-labelled λ phage.

D. Electron micrographs of λ phage infecting cells at high, ~100 (top) and low, ~5 (bottom) moi after 1 h incubation at 37°C. Arrows mark phages at low moi.

E. Infection of E. coli cells with different phage labelled with QDots, as indicated. F. Infection of Y. pseudotuberculosis by QDot-labelled phage ϕA1122, and V. cholerae by QDot-labelled vibriophage KVP40.

Viability of labelled phage

We tested the ability of the labelled λ phage to adsorb to cells by visualizing live samples under the fluorescence microscope. Biotinylated phages were incubated with bacterial cells for 5 min at room temperature. After an additional short incubation with streptavidin-conjugated QDots, samples were washed and observed under the fluorescence microscope. Phage adsorption to cells could be readily visualized (Fig. 1B), and at moi of 1, for example, about 25% of the cells had QDots foci. We next tested viability of the QDot-conjugated λ phages as determined by their ability to form plaques. We optimized the cross-linking conditions to give maximum labelling (about 80% of the phage particles, as mentioned above) without affecting the number of plaque-forming units (PFUs) as compared with the PFUs of unlabelled phages (data not shown). Finally, we infected a bacterial host carrying the reporter fusion pR’tR’–GFP with the QDot-conjugated λ phage. Cells carrying the plasmid encoding GFP under the λ pR’tR’ promoter express GFP only in the presence of the λ phage late protein Q, an antiterminator that is produced following injection of the phage DNA (Kobiler et al., 2005). Incubation for 1 h following adsorption led to extensive activation of the reporter fusion as observed by the accumulation of GFP in the infected cells (Fig. 1C, green). Infected cells were identified by the presence of attached QDot-labelled phage (Fig. 1C, red). In agreement with the TEM estimation, 85% of the cells that expressed GFP had QDot-conjugated phage bound to them. The remainder of cells that express GFP most likely represent infection by unlabelled λ phage as GFP expression was observed only upon λ phage addition; no fluorescence was detected in pR’tR’–GFP carrying cells when λ phage was omitted (QDot-conjugated or unconjugated, data not shown). These results indicate that QDot-conjugated λ phages are both viable and able to carry out an effective infection.

During these tests we discovered that at lower moi (moi of 0.5–5.0), phage show a high preference for interacting with the polar regions of bacterial host cells, hereafter referred to as poles (Fig. 1B). About 70% of the QDot-labelled phages were found bound at the pole or at mid-cell, which represents a future pole of the cell (Fig. 1B; Table 1). The remainder of the QDot-labelled phages was distributed over the bacterial surface without apparent preference for a particular location. When phage infection was omitted or upon infection of an E. coli lamB::Cm mutant, which is resistant to λ infection because of the absence of the λ phage surface receptor LamB, no QDots were detected bound to cells (results not shown), indicating that the QDots observed on cells represent bound phage. Preferential binding of phage λ to the cellular poles was not due to the chemical modification of the phages or the presence of QDots, as electron micrographs of fixed samples showed phage λ, not labelled with QDots, at an moi of 5, also bound with high preference to the polar regions of the cells (Fig. 1D, bottom). The distribution of untreated phages was similar to that observed for QDot-conjugated phages at the same moi visualized by light microscopy.

Table 1.

Various phages labeled with QDots infecting their corresponding host were observed under the fluorescence microscope and scored for the localization on the cell surface.

Phage λ T7 KVP40 ϕA1122 P1 T4 λ λϕ80
Bacterial strain E. coli E. coli V. cholera Y. pseudotuberculosis E. coli E. coli E. coli pel E. coli
% foci at the pole and mid-cell 69% 71% 73% 68% 78% 95% 71% 68%
% foci in other locations 31% 29% 27% 32% 22% 5% 29% 32%
Open in a new tab

Percentage represents the number of cells with the indicated location, out of 100–500 cells with one foci, as observed in the fluorescence microscope.

These findings were surprising as a recent analysis of protein distribution in E. coli (GenoBase) found that only 16% of proteins demonstrate a specific localization within the E. coli cell, including polar and non-polar localization [given that GFP does not fold properly when exported from the cytoplasm, the number might be an underestimation (Feilmeier et al., 2000)]. These data are based on visualization of GFP fusion proteins in E. coli K-12 (W3110) using the fluorescence microscope (for construction details and related information: http://ecoli.naist.jp/gb5/WGB/intro.html). In this analysis, the distribution pattern of GFP expressed from in-frame gfp fusions to 4706 distinct E. coli ORFs was determined and categorized as cellular component (64.2%), membrane component (8.3%), membranous and foci (2.2%), foci (14.1%), ring (0.2%), nucleoid (1.3%), nucleoid and foci (0.6%), undefined (0.3%) or no signal (8.9%).

We next investigated whether the observed preference of λ phage for binding to the polar regions of E. coli is a unique behaviour of bacteriophage λ and E. coli or a more general behaviour of bacteriophage binding to Gram-negative bacteria.

Polar localization in other bacteria and phages

To test whether the preferential binding of λ phage particles to cellular poles is shared by other phages, we monitored the behaviour of additional temperate coliphages: H-19B, a lambdoid phage that carries the stxI genes, which encode the two toxin subunits of a Shiga-like toxin, ϕ80, P1, and the virulent coliphages T4 and T7. Each phage binds to distinct cellular receptors: the receptor for phage ϕ80 is the outer membrane iron transporter FhuA (Mangenot et al., 2005), the receptors for T4 phages are the outer membrane proteins OmpF and OmpC (Hashemolhosseini et al., 1994) and the receptor for phage T7 is lipopolysaccharide (Steven et al., 1988). Phage P1 has a wide host range (Kaiser and Dworkin, 1975). We used the same labelling method and confirmed viability for each of these phages by their infectivity. In each case, the bacterial poles were found to be the preferred site of binding (Fig. 1E and Table 1). The preference of phages for binding to the polar regions of the cell was also observed for bacteria other than E. coli K-12. QDot-conjugated phages were localized preferentially to the cell poles upon infection of Yersinia pseudotuberculosis by the T7-like Yersinia phage ϕA1122 (Garcia et al., 2003) and upon infection of Vibrio cholerae by the T4-like vibriophage KVP40 (Miller et al., 2003) (Fig. 1F and Table 1). These data indicate that the preference of infecting phages for polar sites on the bacterial surface is a general characteristic of phage infection. These results suggest the possibility of evolutionary conservation of certain host proteins or structures to the polar regions of Gram-negative bacteria.

Colocalization of QDot-conjugated λ phages with the polar protein IcsA

The process of cell division leads to the development of new cell poles that are derived from the septum at mid-cell. Several lines of evidence indicate that molecular markers of the future cell pole are present at mid-cell prior to septation. Indeed, certain polar proteins are able to recognize mid-cell, as well as the sites of future cell poles in filamentous cells. IcsA, a polar outer membrane protein present in Shigella, is secreted at the bacterial pole after it is targeted to the pole in the bacterial cytoplasm (Charles et al., 2001; Brandon et al., 2003). Localization of IcsA to the pole depends on two independent peptides, residues 1–104 and residues 507–620; each of these sequences is able to direct a GFP fusion to the pole in the cytoplasm of Shigella flexneri, E. coli and other enterobacteriaceae (Charles et al., 2001). In bacteria in which cell division is blocked, a GFP fusion to IcsA residues 507–620 (IcsA507–620–GFP) localizes at or near potential cell division sites, which correspond to the future poles of the cell (Janakiraman and Goldberg, 2004). IcsA507–620–GFP can therefore be used as a marker for sites that correspond to poles and future poles of the cell. E. coli and Y. pseudotuberculosis cells expressing IcsA507–620–GFP were infected with QDot-labelled phage λ and QDot-labelled Yersinia phage ϕA1122 respectively. The labelled phage colocalized with IcsA507–620–GFP (Fig. 2A and B), which was localized to the polar regions of the cell and/or to mid-cell. This result supports the results presented above and leads to the conclusion that the sites to which the QDot-conjugated phages attach on the bacterial surface correspond to present or future polar regions of the cell.

Fig. 2.

Fig. 2

Open in a new tab

Colocalization of IcsA507–620–GFP and adsorbed phage in live bacteria.

A and B. Shown from left to right, IcsA-labelled poles (green), QDot-labelled phages (red), and overlay of the two: (A) E. coli or (B) Y. pseudotuberculosis expressing IcsA–GFP and infected with QDot-labelled phage λ or ϕA1122, respectively, at low moi.

C. QDot-labelled phage λ colocalized with IcsA507-620–GFP at intervals along the lengths of filamented cells.

To test whether the observed preference of phage for binding to the cellular poles is determined by the hemispherical configuration of the poles, we examined the distribution of phage on cells that had been filamented. As mentioned above, in filamented cells, potential cell division sites that correspond to future poles are marked by IcsA507-620–GFP (Janakiraman and Goldberg, 2004). To generate filamentous cells, cultures were treated with the antibiotic aztreonam, which inhibits the septum-specific enzyme PBP 3 (FtsI) (Adam et al., 1997). At low moi, QDot-conjugated phage λ colocalized with at least one of the IcsA507-620–GFP foci, which were present at regular intervals along the lengths of filamented cells that had been filamented (Fig. 2C), as previously described (Janakiraman and Goldberg, 2004), indicating that phage bind to sites corresponding to potential poles as well as to mature poles. Moreover, preferential binding to the poles does not depend on the hemispherical configuration of the binding site.

Involvement of the host proteins LamB and ManY, which are required for λ DNA injection

LamB, the bacterial receptor for phage λ, is an abundant outer membrane protein, with approximately 30,000 copies of LamB per cell (Boos and Shuman, 1998). As expected, we observed that binding of phage λ was dependent on LamB, as no bacteria-bound QDots were detected when an E. coli lamB mutant was infected with QDot-conjugated phage (data not shown). Using fluorescently labelled bacteriophage lambda tails, LamB has been shown to be distributed in spirals that extend from pole to pole along the length of the cell and to move laterally in the outer membrane along these spirals (Gibbs et al., 2004). LamB thereby provides many sites for phage adsorption that are distributed over the entire cell surface and would allow bound phage to move along the cell surface. Given the abundance of LamB, we tested whether the distribution of bound phage differs upon infection at high moi.At an moi of ~100, phage λ was bound over the entire surface of the cell (Fig. 1D, top), as has been frequently seen in classical electron micrographs (Ryter et al., 1975). The polar binding that was observed by TEM at low moi, but not at high moi (Fig. 1D, bottom and top respectively), is consistent with a process in which phages bind LamB at many locations on the cell surface. We postulate that when phages are limiting, LamB-bound phages move laterally along the cell surface until they reach the pole, while at high moi, phages move laterally reaching other locations in addition to the one that reached the pole.

An extension of this model is that phage DNA injection might occur most efficiently after the phage reaches the cell pole. To identify the site of phage DNA injection, we first determined the location of ManY (PtsM), an inner membrane protein that is known to be required for phage λ DNA injection (Scandella and Arber, 1974). ManY is the mannose-specific enzyme IICMan component of the sugar phosphotransferase system (PTS); a polytopic inner membrane component of the mannose-specific ABC transporter. It lies within the pel locus, which was shown in early work to be required for injection of phage λ DNA into E. coli (Williams et al., 1986). If phage injection occurred preferentially at the pole, one would expect ManY to be preferentially at the pole as well. To determine the localization of ManY, we generated a GFP–ManY protein fusion, expressed under the control of the arabinose promoter PBAD. Expression of GFP–ManY rescued the ability of λ to produce plaques on a lawn of a manY mutant of E. coli (data not shown), indicating that the GFP–ManY fusion is functional for λ DNA injection. GFP–ManY localized to the poles and/or to the cell division sites of manY E. coli cells in 87% of cells (Fig. 3B, left), whereas the expression of GFP alone showed uniform diffused fluorescence (foci at the pole and/or mid-cell was observed in < 1% of the cells expressing GFP alone). These results and the observation that Enzyme I, which carries the first enzymatic step in the PTS pathway, is also localized to the cell poles (Patel et al., 2004) suggest that the observed polar localization of GFP–ManY is specific and functional. As expected from our model, localization of QDot-conjugated phage to the cell poles was unperturbed in cells carrying a deletion in manY or pel (Fig. 3A), as ManY/Pel is not required for phage adsorption, but for a later step of phage DNA injection.

Fig. 3.

Fig. 3

Open in a new tab

Localization of QDot-labelled λ phage and GFP–ManY in a manY strain.

A. Infection of E. coli manY mutant cells with QDot-labelled phage λ.

B. Distribution of GFP signal from E. coli BW25113 ΔmanY::Kan expressing GFP–ManY (left) or GFP alone (right). Direct fluorescence microscopy of GFP signal. Images are representative of those obtained in three experiments. Arrows, foci at the bacterial pole; arrowheads, foci at mid-cell.

Localization of injected phage λ DNA

To test whether DNA injection takes place at the poles, we constructed a λ phage carrying an array of 64 lac operators (Lau et al., 2003; Fekete and Chattoraj, 2005). The expression of LacI–GFP was induced for 1 h prior to phage infection. Following infection, phage genomes carrying the cluster of LacO appeared near the pole and remained there during replication, as indicated by increasing fluorescence in time-lapse images (Fig. 4). This result suggests that, as for phage particles adsorption, DNA injection occurs preferentially at the poles. In contrast, we found that cells with stable lysogens, in which the phage genome is integrated at the attλ of the bacterial genome, contained two LacI–EYFP foci that appear, as expected, in the nucleoid region (data not shown). Unlike the cellular replication machinery, which has been suggested to localize to mid-cell (Li et al., 2002), phage DNA-associated replisomes appear to be replicated at the subcellular position of the injected DNA, at the pole, as has also been shown for the replisomes associated with a multicopy plasmid in Bacillus subtilis (Wang et al. 2004).

Fig. 4.

Fig. 4

Open in a new tab

Monitoring the localization of phage DNA in lytic cycle. Cells carrying a plasmid expressing LacI–GFP were infected by phage λ that contain an array of lac operators in its genome.

A. Snapshot gallery.

B. Time-laps picture taken at 5, 10, 15, 27 and 57 min after incubation at 37°C following infection at room temperature. Cells after λ phage DNA injection at the pole (left) or mid-cell (right).

FtsH, a protein involved in λ phage lysis-lysogeny determination, is localized to the pole

The protease FtsH (HflB) degrades phage λ CII and CIIII proteins and thereby affects whether infecting λ phage will follow the lytic or the lysogenic pathway. ftsH mutations stabilize the CII and CIII proteins, resulting in higher levels of the CI repressor synthesis and thus shifting the lyticlysogenic decision in favour of lysogeny (Herman et al., 1997). In B. subtilis, during vegetative growth, FtsH accumulates in the mid-cell septum, whereas at the onset of sporulation, at positions near the cell poles that appears to coincide with future division sites (Wehrl et al., 2000). To determine the localization of FtsH in E. coli, we constructed a protein fusion of GFP to FtsH. The C-terminus of FtsH undergoes proteolytic cleavage, a self-catalysed reaction that requires ATP hydrolysis (Akiyama, 1999). While complementation experiments with genetically constructed variants suggested that both the processed and the unprocessed forms of FtsH are functional (Akiyama, 1999), for the localization studies we fused GFP to the uncleaved N-terminus of FtsH. The fusion protein was functional, as determined by its ability to complement λ lytic growth on an ftsH-deleted strain, which dose not support the lytic cycle. The efficiency of plating (PFU) increased by 103 on the ftsH-deleted host (data not shown). Fluorescence microscopy showed that the GFP–FtsH fusion forms foci that are predominantly polar in the cell (Fig. 5), consistent with FtsH being localized mainly to the bacterial pole. The concentration of FtsH at the pole suggests that CII expressed from λ phage that infects at the pole may be more susceptible to degradation by FtsH than CII expressed from λ phage that infects at other sites.

Fig. 5.

Fig. 5

Open in a new tab

FtsH, a protein involved in λ phage lysis-lysogeny determination, is localized to the pole. Localization of the GFP–FtsH fusion protein expressed in an ftsH strain.

Model for λ infection and the lytic-lysogenic decision

The results presented above lead us to propose a working model for phage infection that is illustrated in Fig. 6. An infecting phage λ first encounters one of the many LamB receptors on the cell surface and then the LamB bound λ phage migrates along the cell surface. At the poles, it encounters a ‘slot’ or interacts with additional proteins, which stops its movement. The DNA is injected at the poles, with the help of ManY, where phage replication occurs. As FtsH localizes preferentially to the cell pole, the CII protein, the major player in the lytic-lysogenic cycle decision, is subject to degradation. Consequently, the phage that infects at the pole (at low moi) is more likely to end up in the lytic cycle. At high moi, where it has been shown that lysogenization is preferable (Kourilsky, 1973), phages stop their movement at sites other than the pole. DNAinjection at these sites might be less efficient as ManY localization is preferentially to the pole and mid-cell. Because FtsH is preferentially localized to the pole, CII expressed by phage that inject DNA at these other sites may be less apt to be degraded, favouring lysogenization. In addition, CII and CIII expressed by phage that infects at the pole may titrate the FtsH locally, thereby leading to accumulation of CII expressed by phages that infect at non-polar sites. CII protein, made at latter sites, can then diffuse through the cell to its other targets, as supported by a diffuse fluorescent signal from the functional CII–GFP fusion protein expressed from a plasmid (data not shown).

Fig. 6.

Fig. 6

Open in a new tab

A schematic model for phage infection (not to scale). An infecting phage λ first encounters one of the many LamB receptors on the cell surface. It moves while bound to LamB and migrates along the cell to the poles where interaction with additional host proteins takes place. The DNA is injected at the poles, a process that depends on the polar protein ManY. Following DNA injection, phage replication occurs at the pole; host and phage proteins required for phage development are colocalized at the poles. The protease FtsH, which is localized at the pole, degrades CII, leading to the lytic cycle. If more then one phage binds a cell (right), CII from the phage infecting in a location other than the pole is not degraded. Due to the effect of CII, which diffuses throughout the cytoplasm, including to phage DNA that was injected at the pole, lysogenization is established. IM, inner membrane; OM, outer membrane.

Why then do phages (λ, T4, T7, P1 Yersinia phage ϕA1122 and vibriophage KVP40 tested in the present study) attach to the polar regions of the cell? It is possible that the polar region is the preferred place for DNA injection to take place, as uptake of DNA by naturally competent bacteria also occurs at the pole (Chen et al., 2005). The polar localization of ManY and FtsH supports this notion, as does the reported polar localization of FhuD (see http://ecoli.naist.jp). The identification of host components essential for DNA injection of other phages and the localization of these host components remains to be investigated. The nucleoid-free cytoplasm at the cell poles could also provide a space for rapid phage DNA replication following infection. Regardless of the evolutionary forces, the data presented here indicate that phages have evolved mechanisms to utilize the asymmetry that is present in Gram-negative bacteria.

Experimental procedures

Bacteria strains plasmid and phage

Standard microbiological techniques were used for growth and manipulation of bacteria and phage (Miller, 1972; Maniatis et al., 1982; Sambrook et al., 1989). The phages λ cI857S7, ϕ80, H-19B, P1vir, T4, T7 and Yersinia phage ϕA1122 are from our collection. Phage KVP40 was kindly supplied by the D. Hinton laboratory. Bacterial strains E. coli K-12 C600 and W3110, Y. pseudotuberculosis, are from our collection. V. cholera was kindly provided by the D. Chattoraj laboratory. Other strains include the ftsH deletion strain, which is W3110 sfhC ZAD220::Tn10 ftsH3::Kan (Ogura et al., 1999). To generate a lysate of λ I29 CIts 64 lacO Kan, an overnight culture of the lysogen MG1655 gal::64 λ cIO cm, attB::λ I29 CIts 64 lacO Kan was grown at 30°C in LB medium supplemented with the appropriate antibiotics. It was then back-diluted 1:100 into the same medium, grown for 4 h at 32°C, and then shifted to 42°C until lysis was visible.

For the localization of λ phage DNA, we used E. coli strain BW27750, which allows uniform expression of genes under the control of the arabinose promoter (PBAD) among cells in the population (Khlebnikov et al., 2001). As a source of LacI–EYFP (an enhanced yellow-green variant of the Aequorea Victoria GFP) we have used the low-copy-number plasmids pRFY001 in which the fusion protein is cloned under the control of the arabinose promoter (Lau et al., 2003; Fekete and Chattoraj, 2005).

The pR’tR’–GFP expression plasmid used for viability tests of the labelled λ phage carries GFP under the control of the λ promoter-terminator (λ coordinates 5′-44300 and 44831-3′) (Marr et al., 2001).

pBAD24–gfp::manY and pBAD24–gfp::ftsH were obtained by cloning the coding sequences of manY or ftsH under the control of the arabinose promoter, downstream of gfp in pBAD–gfp (Nilsen et al., 2004) as a translational fusion.

For localization of GFP–ManY, strain BW25113 ΔmanY::Kan and its isogenic strain BW25113 (Baba et al., 2006) were transformed with plasmids pBAD24–gfp or pBAD24–gfp::manY.

Construction of (lacO64) phage

The λ (lacO64) phage was constructed by in vivo recombineering (Thomason et al., 2005). First, the R6Kγ/zeo plasmid was constructed by ligating the 814 bp XmnI–EcoRI fragment of pGEM-4Z (Promega) with the 324 bp HindIII–PvuII fragment of plasmid pMOD™-3<R6Kγori/MCS> (Epicentre) zeocin resistance cassette of pEM7/zeo (Invitrogen), which was amplified by PCR using primers containing EcoRI and HindIII restriction sites, and digested with the cognate enzymes. The R6Kγ/zeo plasmid was maintained in E. coli EC100 pir+ host expressing the P protein required for replication of plasmids containing the R6Kγ origin of replication. The regions of 46368–46465 and 47981–48387 of bacteriophage λ (GenBank Accession No. J02459) were amplified by PCR using the primer pairs of λ64upNdeI (GAAAACATATGATATCACTATGCAAAAACAACTGGAAGG AAC) and λ64upPvuII (AAAAGGATCCAGCTGCTCGGTTTT ATTACTTTAGGC), or λ64downRI (AAAAGAATTCGATATC GTATTAATTGATCTGCATCAACTTAAC) and λ64downPvuII (AAAAGGATCCTTACTGCAATGCCCTCGTAATTAAGT) respectively. The PCR product of the 46368–46465 region was digested with NdeI and BamHI, and the PCR product of 47981–48387 region was digested with EcoRI and BamHI. The fragments were inserted between the EcoRI and NdeI ends of plasmid R6Kγ/zeo in a three-piece ligation reaction, resulting in R6Kγ/λ. The ends of the 4070 bp NdeI fragment of plasmid pRFB110 (Fekete and Chattoraj, 2005), containing an array of 64 lac operator sites and a kanamycin resistance cassette, were filled in using the Klenow fragment of DNA polymerase. The fragment was then inserted at the PvuII site located between the two λ phage-derived sequences of R6Kγ/λ. Finally, the resulting plasmid was digested with EcoRV, and the linear fragment containing the array of 64 lac operator sites and the kanamycin resistance cassette flanked by the λ phage-derived sequences was recombined onto a λI21CIts prophage by recombineering, and recombinant phages were isolated as kanamycin-resistant colonies.

Chemical modification of bacteriophages

CsCl-purified or crude bacteriophage lysates were incubated with 4 μM EZ-LinkTM Sulpho-NHS-LC-LC-Biotin [sulphosuccinimidyl-6-(biotinamido)-6-hexanamido hexanoate] (PIERCE). A 10 mM solution was made in 1× PBS to which 10 mM MgSO4 was added and the solution was diluted 10- or 100-fold. After 30 min at room temperature the reaction was stopped by the addition of glycine to a final concentration of 0.025%. Conditions were optimized to have minimal effect on phage viability as tested by direct plaque assay.

Preparation of samples for fluorescence and TEM

For phage binding localization, cells were grown to exponential phase (with 0.2% maltose for λ phage, with 2 μg ml−1 aztreonam for experiments in filamented cells) in LB medium and mixed with appropriate chemically modified bacteriophage. After incubation at room temperature, samples were centrifuged for 5 min at 2300 g and re-suspended in PBS buffer. Alternatively, infected cells were incubated at 37°C for additional 5 min to induce bacteriophage DNA injection. For fluorescence microscopy, Qdot® 655 streptavidin conjugates (Invitrogen) were added in excess, followed by a 5 min incubation at room temperature and spin as before.

For λ phage DNA localization, an overnight culture of BW27750 strain carrying pRFY001 was diluted 1:100 in minimal medium + maltose and grown to OD600 0.2. Arabinose at a concentration of 0.00002% was added to induce the expression of LacI–EYFP for 1 h. Cells were concentrated to OD600 10 and phage λ was added followed by incubation for 20 min on ice and additional 5 min at 37°C. About 3 μl of cells were then placed on a slide and overlaid with a coverslip treated with poly-l-lysine (Sigma Diagnostics).

For time-lapse microscopy, slides with 1.5% agarose (Seakem GTG Agarose, Cambrex Bio Science Rockland) in minimal medium were prepared. About 3 μl of cells were then placed on a slide and overlaid with a coverslip. The microscope stage was maintained at 37°C.

The expression of GFP–ManY or GFP–FtsH was induced by adding arabinose to exponential-phase cells to a concentration of 0.2% for 30–60 min at 37°C. Cells were then harvested, washed and observed using a fluorescent microscope.

Fluorescence microscopy

Most microscopy was performed on an Eclipse E1000 microscope (Nikon), using a Sensicam QE CCD camera (Cooke Corporation) controlled by IP Laboratories software (Scanalytics). Approximately 2 μl of samples were placed on a slide and overlaid with a coverslip or coverslip treated with poly-l-lysine (Sigma Diagnostics).

GFP–ManY fluorescence microscopy was performed using a 100× oil immersion objective on a TE300 microscope (Nikon) with Chroma Technology filters. Images were captured using a Photometrics CoolSnap HQ charge-coupled device camera and IP Laboratory software (Scanalytics).

Transmission electron microscopy

QDots-conjugated CsCl-purified λcI857susS7 or CsCl-purified λcI857susS7 phage that had been incubated with late exponential-phase W3110 cells grown in LB with 0.25% maltose at the indicated moi was used as samples. Four microlitres of the sample to be examined was applied as a drop to a carbon-coated, glow-discharged specimen grid. After 30 s the excess liquid was wicked away by filter paper and the grid was negatively stained by allowing several drops of 1% aqueous uranyl acetate to flow over the grid. Specimens were examined in a FEI Tecnai12 operating at 120 kV and images were recorded on a Gatan CCD Camera.

Acknowledgements

We thank Debbie Hinton for the vibriophage KVP40, and Ilan Rosenshine for fruitful discussions. This research was supported in part by The Israel Science Foundation (Grant No. 489/01-1 and 340/04), by the Intramural Research Program of the National Institutes of Health, National Cancer Institute, Center for Cancer Research, and by NIH Grants AI35817 and AI071240 (to M.B.G.).

References

  1. Adam M, Fraipont C, Rhazi N, Nguyen-Distèche M, Lakaye B, Frère J-M, et al. The bimodular G57–V577 polypeptide chain of the class B penicillin-binding protein 3 of Escherichia coli catalyzes peptide bond formation from thiolesters and does not catalyze glycan chain polymerization from the lipid II intermediate. J Bacteriol. 1997;179:6005–6009. doi: 10.1128/jb.179.19.6005-6009.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Akiyama Y. Self-processing of FtsH and its implication for the cleavage specificity of this protease. Biochemistry. 1999;38:11693–11699. doi: 10.1021/bi991177c. [DOI] [PubMed] [Google Scholar]
  3. Baba T, Ara T, Hasegawa M, Takai Y, Okumura Y, Baba M, et al. Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Mol Syst Biol. 2006;2:1–11. doi: 10.1038/msb4100050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Boos W, Shuman H. Maltose/maltodextrin system of Escherichia coli: transport, metabolism, and regulation [Review] Microbiol Mol Biol Rev. 1998;62:204–229. doi: 10.1128/mmbr.62.1.204-229.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Brandon LD, Goehring N, Janakiraman A, Yan AW, Wu T, Beckwith J, Goldberg MB. IcsA, a polarly-localized autotransporter with an atypical signal peptide, uses the Sec apparatus for secretion, although the Sec apparatus is circumferentially distributed. Mol Microbiol. 2003;50:45–60. doi: 10.1046/j.1365-2958.2003.03674.x. [DOI] [PubMed] [Google Scholar]
  6. Charles M, Pérez M, Kobil JH, Goldberg MB. Polar targeting of Shigella virulence factor IcsA in Enterobacteriacae and Vibrio. Proc Natl Acad Sci USA. 2001;98:9871–9876. doi: 10.1073/pnas.171310498. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chen I, Christie PJ, Dubnau D. The ins and outs of DNA transfer in bacteria [Review] Science. 2005;310:1456–1460. doi: 10.1126/science.1114021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Feilmeier BJ, Iseminger G, Schroeder D, Webber H, Phillips GJ. Green fluorescent protein functions as a reporter for protein localization in Escherichia coli. J Bacteriol. 2000;182:4068–4076. doi: 10.1128/jb.182.14.4068-4076.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Fekete RA, Chattoraj DK. A cis-acting sequence involved in chromosome segregation in Escherichia coli. Mol Microbiol. 2005;55:175–183. doi: 10.1111/j.1365-2958.2004.04392.x. [DOI] [PubMed] [Google Scholar]
  10. Garcia E, Elliott JM, Ramanculov E, Chain PS, Chu MC, Molineux IJ. The genome sequence of Yersinia pestis bacteriophage phiA1122 reveals an intimate history with the coliphage T3 and T7 genomes. J Bacteriol. 2003;185:5248–5262. doi: 10.1128/JB.185.17.5248-5262.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gibbs KA, Isaac DD, Xu J, Hendrix RW, Silhavy TJ, Theriot JA. Complex spatial distribution and dynamics of an abundant Escherichia coli outer membrane protein, LamB. Mol Microbiol. 2004;53:1771–1783. doi: 10.1111/j.1365-2958.2004.04242.x. [DOI] [PubMed] [Google Scholar]
  12. Hashemolhosseini S, Holmes Z, Mutschler B, Henning U. Alterations of receptor specificities of coliphages of the T2 family. J Mol Biol. 1994;240:105–110. doi: 10.1006/jmbi.1994.1424. [DOI] [PubMed] [Google Scholar]
  13. Herman C, Thevenet D, D’Ari R, Bouloc P. The HflB protease of Escherichia coli degrades its inhibitor λ cIII. J Bacteriol. 1997;179:358–363. doi: 10.1128/jb.179.2.358-363.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Janakiraman A, Goldberg MB. Evidence for polar positional information independent of cell division and nucleoid occlusion. Proc Natl Acad Sci USA. 2004;101:835–840. doi: 10.1073/pnas.0305747101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Kaiser D, Dworkin M. Gene transfer to myxobacterium by Escherichia coli phage P1. Science. 1975;187:653–654. doi: 10.1126/science.803710. [DOI] [PubMed] [Google Scholar]
  16. Khlebnikov A, Datsenko KA, Skaug T, Wanner BL, Keasling JD. Homogeneous expression of the P (BAD) promoter in Escherichia coli by constitutive expression of the low-affinity high-capacity AraE transporter. Microbiology. 2001;147:3241–3247. doi: 10.1099/00221287-147-12-3241. [DOI] [PubMed] [Google Scholar]
  17. Kobiler O, Rokney A, Friedman N, Court DL, Stavans J, Oppenheim AB. Quantitative kinetic analysis of the bacteriophage lambda genetic network. Proc Natl Acad Sci USA. 2005;102:4470–4475. doi: 10.1073/pnas.0500670102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kourilsky P. Lysogenization by bacteriophage lambda. I. Multiple infection and the lysogenic response. Mol Gen Genet. 1973;122:183–195. doi: 10.1007/BF00435190. [DOI] [PubMed] [Google Scholar]
  19. Lau IF, Filipe SR, Soballe B, Okstad OA, Barre FX, Sherratt DJ. Spatial and temporal organization of replicating Escherichia coli chromosomes. Mol Microbiol. 2003;49:731–743. doi: 10.1046/j.1365-2958.2003.03640.x. [DOI] [PubMed] [Google Scholar]
  20. Li Y, Sergueev K, Austin S. The segregation of the Escherichia coli origin and terminus of replication. Mol Microbiol. 2002;46:985–996. doi: 10.1046/j.1365-2958.2002.03234.x. [DOI] [PubMed] [Google Scholar]
  21. Mangenot S, Hochrein M, Radler J, Letellier L. Real-time imaging of DNA ejection from single phage particles. Curr Biol. 2005;15:430–435. doi: 10.1016/j.cub.2004.12.080. [DOI] [PubMed] [Google Scholar]
  22. Maniatis T, Fritsch EF, Sambrook JF. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press; Cold Spring Harbor Laboratory, NY: 1982. [Google Scholar]
  23. Marr MT, Datwyler SA, Meares CF, Roberts JW. Restructuring of an RNA polymerase holoenzyme elongation complex by lambdoid phage Q proteins. Proc Natl Acad Sci USA. 2001;98:8972–8978. doi: 10.1073/pnas.161253298. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Miller JH. Experiments in Molecular Genetics. Cold Spring Harbor Laboratory Press; Cold Spring Harbor, NY: 1972. [Google Scholar]
  25. Miller ES, Heidelberg JF, Eisen JA, Nelson WC, Durkin AS, Ciecko A, et al. Complete genome sequence of the broad-host-range vibriophage KVP40: comparative genomics of a T4-related bacteriophage. J Bacteriol. 2003;185:5220–5233. doi: 10.1128/JB.185.17.5220-5233.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Nilsen T, Ghosh AS, Goldberg MB, Young KD. Branching sites and morphological abnormalities behave as ectopic poles in shape-defective Escherichia coli. Mol Microbiol. 2004;52:1045–1054. doi: 10.1111/j.1365-2958.2004.04050.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Ogura T, Inoue K, Tatsuta T, Suzaki T, Karata K, Young K, et al. Balanced biosynthesis of major membrane components through regulated degradation of the committed enzyme of lipid A biosynthesis by the AAA protease FtsH (HflB) in Escherichia coli. Mol Microbiol. 1999;31:833–844. doi: 10.1046/j.1365-2958.1999.01221.x. [DOI] [PubMed] [Google Scholar]
  28. Patel HV, Vyas KA, Li X, Savtchenko R, Roseman S. Subcellular distribution of enzyme I of the Escherichia coli phosphoenolpyruvate: glycose phosphotransferase system depends on growth conditions. Proc Natl Acad Sci USA. 2004;101:17486–17491. doi: 10.1073/pnas.0407865101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Ptashne M. A Genetic Switch: Phage Lambda Revisited. 3rd edn Cold Spring Harbor Laboratory Press; Cold Spring Harbor, NY: 2004. [Google Scholar]
  30. Ryter A, Shuman H, Schwartz M. Intergration of the receptor for bacteriophage lambda in the outer membrane of Escherichia coli: coupling with cell division. J Bacteriol. 1975;122:295–301. doi: 10.1128/jb.122.1.295-301.1975. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd edn Cold Spring Harbor Laboratory Press; Cold Spring Harbor, NY: 1989. [Google Scholar]
  32. Scandella D, Arber W. An Escherichia coli mutant which inhibits the injection of phage lambda DNA. Virology. 1974;58:504–513. doi: 10.1016/0042-6822(74)90084-1. [DOI] [PubMed] [Google Scholar]
  33. Steven AC, Trus BL, Maizel JV, Unser M, Parry DA, Wall JS, et al. Molecular substructure of a viral receptor-recognition protein. The gp17 tail-fiber of bacteriophage T7. J Mol Biol. 1988;200:351–365. doi: 10.1016/0022-2836(88)90246-x. [DOI] [PubMed] [Google Scholar]
  34. Suttle CA. Viruses in the sea [Review] Nature. 2005;437:356–361. doi: 10.1038/nature04160. [DOI] [PubMed] [Google Scholar]
  35. Szmelcman S, Hofnung M. Maltose transport in Escherichia coli K-12: involvement of the bacteriophage lambda receptor. J Bacteriol. 1975;124:112–118. doi: 10.1128/jb.124.1.112-118.1975. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Thomason LC, Bubunenko M, Costantino N, Wilson HR, Oppenheim A, Datta S, Court DL. Recombineering: genetic engineering in bacteria using homologous recombination. In: Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K, editors. Current Protocols in Molecular Biology. Wiley; New York, N.Y.: 2005. pp. 1.16.11–1.16.21. [DOI] [PubMed] [Google Scholar]
  37. Wang JD, Rokop ME, Barker MM, Hanson NR, Grossman AD. Multicopy plasmids affect replisome positioning in Bacillus subtilis. J Bacteriol. 2004;186:7084–7090. doi: 10.1128/JB.186.21.7084-7090.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Wehrl W, Niederweis M, Schumann W. The FtsH protein accumulates at the septum of Bacillus subtilis during cell division and sporulation. J Bacteriol. 2000;182:3870–3873. doi: 10.1128/jb.182.13.3870-3873.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Williams N, Fox DK, Shea C, Roseman S. Pel, the protein that permits lambda DNA penetration of Escherichia coli, is encoded by a gene in ptsM and is required for mannose utilization by the phosphotransferase system. Proc Natl Acad Sci USA. 1986;83:8934–8938. doi: 10.1073/pnas.83.23.8934. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES