The bacterial CM is composed of lipids assembled in a bilayer containing integral and peripheral membrane proteins, carbohydrates, ions (K+, Na+), and water (Finean et al., 1984). There is a large difference of lipid composition among the CM of bacterial species. For most bacteria the predominant zwitterionic phospholipid is phosphatidylethanolamine (PE). In general, Gram-negative bacteria have a higher content of PE than Gram-positive bacteria. The predominant anionic lipids in bacterial membranes are phosphatidylglycerol and cardiolipin (Epand and Epand, 2009). The CM is hydrophobic and fluid in nature. It acts as a semi-permeable barrier to prevent leakage of the hydrophilic constituents from the cytoplasm and to protect the internal milieu of the cell from external environmental insults. Importantly, the CM is impermeable to protons allowing the cell to conserve the energy of electron-transfer as proton electrochemical gradient (ΔµH+) across the membrane. This form of energy is used to drive reactions such as adenosine triphosphate (ATP) synthesis, bacterial motility and active transport across the CM. The CM has many other important functions in cellular processes like cell-cell recognition, signal transduction, chemotaxis, chromosome replication, lipid synthesis, and cell division (for a review see Fyfe et al., 2001; Weiner and Rothery, 2007).
Outer membrane of Gram-negative bacteria
The OM is a distinguishing feature of Gram-negative bacteria. The primary function of the OM is to act as a permeability barrier for cytotoxic compounds and to control the access of solutes to the periplasm.
The OM is an asymmetric bilayer – the inner leaflet is composed of phospholipids, while the outer one consists mainly of lipopolysaccharides (LPS) (Fig. I.4A) (Nikaido, 1996). LPS, a toxic compound for animals, often serves as a phage receptor. LPS plays a critical role in the barrier function of the OM (Raetz and Whitfield, 2002). The LPS molecule can be structurally divided into three parts: a lipid moiety – lipid A, a nonrepeating “core” oligosaccharide and a hydrophilic polysaccharide called the O-antigen which differs from species to species (Madigan et al., 1997; Osborn et al., 1964). LPS molecules bind each other avidly, especially if cations like Ca2+ and Mg2+ are present to neutralize the negative charge of phosphate groups present on the molecule. The acyl chains are largely saturated, and this facilitates tight packing. The nonfluid continuum formed by the LPS molecules is a very effective barrier for hydrophobic molecules (Nikaido, 2003).
To ensure the influx of cell necessary nutrients from extracellular environment and for the extrusion of waste products from cell cytoplasm, the OM contains a number of transmembrane proteins: nonspecific porin channels and specific channels. These outer membrane proteins (OMPs) are formed essentially by β sheets that are wrapped into cylinders delimiting hydrophilic pores. Porins function as channels for traffic of hydrophilic low-molecular-weight substances (with an exclusion limit for molecules of ≥ 700 Da (Nikaido et al., 2003). Specific channels gate the diffusion of specific classes of nutrients.
OMPs are frequently phage receptors. Several OMPs have been identified by this feature (e.g. FhuA (or TonA receptor for phages T1, T5), Bfe (or BtuB; phage BF23), LamB (phage λ) and Tsx (Phage T6) (Heller, 1992).
Another class of OM proteins are lipoproteins. Lipoproteins contain lipid moieties that anchor them in the inner leaflet of the OM (Sankaran and Wu, 1994). As mentioned before, Braun’s lipoproteins link the periplasmic thin layer of PG to the OM, but the function of most of the 100 OM lipoproteins found in E. coli is not known (Miyadai et al., 2004).
Bacterial surface structures
Bacterial surface structures that protrude from the cell surface are key players in adhesion, motility, chemotaxis, bacterial conjugation, biofilm and S-layer formation. These structures include the glycocalyx (capsular polysaccharides and slime layer), various adhesins, flagella and S-layer proteins.
Surface proteins called adhesins have the ability to adhere to host tissue, to solid surfaces or to other cells. Microbial adhesins are often assembled into complex polymeric organelle structures (fimbriae, curli, pili), although adhesins linked to the cell surface as monomers or simple oligomers also exist. Fimbriae are rod-shaped adhesive surface organelles (Klemm and Krogfelt, 1994). Curli are heteropolymeric proteinaceous filamentous appendages that play a role on the adherence properties of several biofilm-forming E. coli strains (Olsén et al., 1989). The F conjugative pilus is involved in horizontal gene transfer from a donor to a recipient cell during bacterial conjugation. In Gram-positive bacteria adhesins are usually attached covalently to stem peptides within the PG layers or sometimes via noncovalent ionic interactions to PG or teichoic acids (Silhavy et al., 2010). Another type of surface structures are those involved in cell motility – long rotating appendages anchored in CM called flagella. Some pili, designated type IV pili, can also generate motile forces (Mattick, 2002). Microbial adhesins and flagella are surface determinants of bacterial biofilm formation (van Houdt and Michiels, 2005).
The S-layer is a paracrystalline array found outside the PG layer in some Gram-positive or on top of the OM in several Gram-negative species. This self-assembling uniform lattice structure is usually composed of a single protein, but may also have attached carbohydrates. It completely envelops the bacterium. Each S-layer has a specific cut-off value and thus may act as molecular sieve to exclude agents that might damage the cell. Although present in many different bacteria and archaea, the functions of S-layers are poorly understood (Young, 2010 and references therein).
Bacterial pili, flagella, capsular and slime polysaccharides can serve as receptors for bacteriophages. Polymerization and depolymerization of pili is exploited by phages to approach the bacterial surface (Dreiseikelmann, 1994; Manchak et al., 2002).
Bacteriophage entry in the host cell
Like all other viruses, bacteriophages are obligate intracellular parasites. Therefore successful delivery of the virion genome to the bacterial cell is an essential early step for host infection. Phages use different strategies to deliver their genetic information across the bacterial envelope to the cell cytoplasm where their genome is expressed and replicated. The details of how viral DNA transits from the virion to the host cytoplasm and of how the cellular environment catalyzes and possibly organizes the entire process remain poorly understood. This process encompasses specific virion-host recognition, passage of the virion across the different cell envelope barriers and establishment of optimal conditions for virus multiplication in the host cytoplasm.
Phage infection is initiated by specific interaction of the virion with receptors at the host surface. The first contact with the bacterial surface usually leads to reversible binding of the virus particle that is not saturable. It allows for dissociation of viable phages from the bacterium. In a second step, phages attach irreversibly to cell envelope receptor committing to infection of the host. This process is normally saturable due to a limited number of active receptors accessible for the irreversible interaction at the cell surface. Reversible and irreversible adsorption can target the same receptor or involve different surface components. The most recent list of known receptors is described by Vinga et al., (2006a).
Structures employed as phage receptors likely include most if not all molecules or molecular assemblies exposed at the cell surface. However, each bacteriophage species is known to be able to infect just a narrow host range. Such specificity is due to the phage ability to interact only with specific receptor structures exposed on the bacterial cell surface. The distinct envelopes of Gram-negative and Gram-positive bacteria impose different barriers to virus attack (section I.4).
Phage receptor binding proteins
As mentioned before, reversible and irreversible adsorption can target the same receptor or involve different surface components. The two strategies correlate with distinct phage adsorption machineries whose complexity can vary from a single receptor binding protein (RBP) to complex baseplates with several RBPs as found in phages with long tails.
The vast majority of known bacterial viruses (96%; Ackermann, 2009) use a tail device for specific recognition of the host, binding to its surface and delivery of the genome from their icosahedral capsid to the bacterial cytoplasm (order Caudovirales). The tail adsorption apparatus can be composed of the distal tail protein (Dit), combined with a tail spike or more complex structures featuring tail fibers, and/or elaborated baseplates. Curiously, some tailed phages encode two alternate tail fiber genes and can switch them to extend their host range (Kamp et al., 1978; Scholl et al., 2001). When phages feature distal fibers, these are frequently used to sense the bacterial cell surface and to interact in a reversible way with host receptors (phages T5, T4, T7). Besides recognition, an activity frequently associated to tail spikes is the ability to attack the host cell receptor or other surface structures facilitating penetration of the phage through the cell wall (Steinbacher et al., 1996).
Phages belonging to the Podoviridae family (P22, φ29, ε15; section I.3) bind frequently to host cells through stubby homotrimeric tail spikes that are attached to their short tail. In phage φ29, the neck appendages (gp12) of the virion are responsible for the reversible interaction with the B. subtilis cell wall (Guo et al., 2003; Villanueva and Salas, 1981). Another tail component, gp13, is proposed to function as a cell-wall degrading enzyme (Xiang et al., 2008).
Adsorption of the Myoviridae family phages T4, P2 or Mu to their bacterial hosts is normally mediated by long and short tail fibers or tail-spike proteins attached to the complex baseplate found at the tip of the contractile tail. Phage T4 host recognition occurs through a reversible adsorption of at least three long tail fibers to the LPS or OmpC found at the bacterial surface. This interaction leads to transmission of a signal to the baseplate, causing the extension of the homotrimeric short fibers (protein gp12) and irreversible interaction with LPS (Kanamaru et al., 2002). Recently, Bartual et al., (2010) presented the crystal structure of the receptor-binding tip of the bacteriophage T4 long tail fiber. This protein is highly homologous to the tip of bacteriophage λ side tail fibers as revealed by sequence similarity analysis.
Siphoviridae (e.g. phages p2, TP901-1, SPP1, λ, T5) use baseplates, straight tail fibers or tail spikes for host receptor recognition. Phage λ uses the tip of its straight tail fiber, formed by protein gpJ, to bind to LamB (Berkane et al., 2006; Wang et al., 2000) while phage T5 protein pb5, located in the tip of the tail conical region, mediates irreversible binding to FhuA (Heller and Schwarz, 1985). All known L. lactis siphophages possess a baseplate at the tip of their tail involved in host recognition and attachment (Sciara et al., 2010 and references therein). Recently Sciara et al., 2010 reported the structure of the lactococcal phage p2 baseplate and proposed a mechanism for baseplate activation during attachment to the host cell. The baseplate is composed of three protein species, including six trimers of the RBP. In the presence of Ca2+, the RBPs rotate, presenting their binding sites to the host. Subsequently a channel opens at the bottom of the baseplate for DNA passage. Phage SPP1 uses an adsorption device located at the tail fiber (Plisson et al., 2007). Interestingly, several tail components of the SPP1 show significant sequence and fold similarity to equivalent proteins from lactococcal phages which possess a baseplate at the tail tip (Veesler et al., 2010a,b). Recently, Goulet et al., (2011) determined the cryo-EM structure of a complex formed by two SPP1 proteins, distal tail cap gp19.1 (Dit) and N-terminus of the gp21 (Tal). Remarkably, in Dit-Tal complexes a Dit hexamer associates with Tal trimers that can be display a “closed” conformation or an “open” conformation delineating a central channel. The two conformational states dock nicely into the EM map of the SPP1 cap domain, respectively before and after DNA release (Goulet et al., 2007; Plisson et al., 2007). Moreover, the open/closed conformations of Dit-Tal are consistent with the structures of the corresponding proteins in the siphophage p2 baseplate, where the Tal protein (ORF16) attached to the ring of Dit (ORF15), was also found to adopt these two conformations (Sciara et al., 2010). Therefore, Goulet and co-workers proposed that this is a prevalent mechanism for the tail opening in siphophages infecting Gram-positive bacteria.
A striking structural and functional similarity was shown between the Type VI secretion system (T6SS) and bacteriophage tail components (Bönemann et al., 2010; Leiman et al., 2009; Pell et al., 2009; Veesler et al., 2010a). Secretion is fundamental to bacterial virulence. Bacterial pathogens use at least 7 distinct extracellular protein secretion systems to export proteins through their cell envelope and in some cases into the host eukaryotic cell or another bacterium (Hood et al., 2010; Schwarz et al., 2010). Several components of the of the T6SS are structurally very similar to those that make up the long tail of bacteriophages. Using crystallographic, biochemical, and bioinformatic analyses, there 3 T6SS components, were shown to be homologous to bacteriophage T4 tail proteins. These include the tail tube protein, the membrane-penetrating needle, and a protein associated with the needle and tube (Leiman et al., 2009). A remarkable structural similarity was also shown between bacteriophage λ major tail protein and a component of the bacterial T6SS (Pell et al., 2009). The structural similarity observed between phage tail and the T6SS proteins, strongly supports the hypothesis that phage tails and T6SS are evolutionarily related. These components of a membrane-penetrating nanomachines most possibly evolved first in the context of a phage tail and bacteria evolving to use parts of phages for their own advantage. The question, how these structurally related machines perform similar but clearly distinct tasks: phage DNA ejection into the cell cytoplasm and delivery of bacterial virulence factors, remain to understand.
Filamentous bacteriophages (fd, M13, f1, Ike, CTXΦ, IF1) use different domains (N1 and N2) of their pIII protein to bind to cellular receptors (Stengele et al., 1990). The initial binding of the pIII N2 domain to a pilus is followed by the irreversible interaction with the C-terminal domain of host protein TolA. The latter interaction is mediated through the domain N1 of pIII (Lorenz and Schmid, 2011; Lubkowski et al., 1998; Riechmann and Holliger, 1997).
Membrane containing bacteriophages belonging to the Cystoviridae (e.g. phage φ6), Tectiviridae (e.g. phage PRD1) and Corticoviridae (e.g. phage PM2) families (Table I.1) and icosahedral tailless ssDNA phage φX174 use spike proteins as receptor-binding proteins (Abrescia et al., 2008; Bamford et al., 1987; Kawaura et al., 2000; Mindich et al., 1982; Romantschuk and Bamford, 1985; Suzuki et al., 1999).
The ssRNA containing phages of Leviviridae family bind to the host cell receptor via a maturation protein (van Duin and Tsareva, 2006).
Cell wall passage
Bacteriophages have to cross the cell wall of their host bacterium for genome delivery to the host cytoplasm. The major classes of PG hydrolases contain unrelated enzymes as lysozymes, amidases, transglycosylases or endopeptidases. Several phages possess genes encoding PG hydrolysing enzymes with lytic domains which are conserved between bacteriophages and plasmid genomes (Koonin and Rudd, 1994; Lehnherr et al., 1998). Bacteriophages are known to have lytic cell wall hydrolysis proteins for phage entry and lytic enzymes for phage progeny escape (endolysins), but their function and roles in virus cycle are different. Contrary to endolysins which are synthesized at the late stage of the lytic cycle, virions often harbor PG hydrolases that locally degrade the PG in order to facilitate the entry of phage DNA during infection. The action of PG hydrolyzing enzymes must be well controlled and cause only transient openings in the PG sacculus to maintain the bacterial cell envelope integrity. In all cases phage entry enzymes are components of the phage particles (Moak and Molineux, 2004). These can be either associated with the tail structure or be an internal head protein, as phage T4 tail protein gp5 and T7 protein gp16, respectively, suggesting a role in cell wall penetration during cell wall passage (Kanamaru et al., 2002; Kao and McClain, 1980; Moak and Molineux, 2000; Nakagawa et al., 1985). Bacteriophage T5 Pb2 is a multifunctional protein that carries fusogenic and PG hydrolysing activities in addition to its tape measure role (Boulanger et al., 2008). Protein P17 of Staphylococcus aureus phage P68 was shown to have both receptor binding and PG hydrolysing activities (Takác and Bläsi, 2005). The tail tip protein gp13 of bacteriophage φ29 has also two PG hydrolyzing activities: the N-terminus domain is a lysozyme, while the C-terminus is an endopeptidase (Xiang et al., 2008). Interestingly, it was shown that residues in the active site and the structural pattern are conserved among bacterial lysostaphin homologs (autolysins) and bacteriophage φ29 gp13 C-terminal endopeptidase domain (Cohen et al., 2009).
Table of contents :
I.1. General properties of phages
I.2. Bacteriophage properties and diversity
I.3. Tailed viruses, order Caudovirales
I.4. Bacterial cell envelope: the barrier to host cell infection
I.4.1. Cell wall
I.4.2. Cytoplasmic membrane
I.4.3. Outer membrane of Gram-negative bacteria
I.4.4. Bacterial surface structures
I.5. Bacteriophage entry in the host cell
I.5.1. Phage adsorption
I.5.1.1. Bacterial receptors
YueB-related membrane proteins
I.5.1.2. Phage receptor binding proteins
I.5.2. Cell wall passage
I.5.3. CM passage and DNA entry in the cytoplasm
I.5.3.1. Genome ejection through a specialized vertex
Phages of Gram-negative bacteria
Phages of Gram-positive bacteria
I.5.3.2. Capsid dissociation at the cell envelope
I.5.3.3. Fusion and endocytosis-like penetration
I.5.3.4. Forces driving tailed phages dsDNA entry in the cytoplasm
I.5.3.5. Ion fluxes accompanying phage infection
I.6. Bacteriophage SPP1
II. MATERIALS AND METHODS
II.1. Bacterial strains and bacteriophages
II.2. Microbiology methods
II.2.1. General methods
II.2.2. Measurement of SPP1 adsorption to B. subtilis cells
II.3. Molecular biology and genetics methods
II.3.1. General recombinant DNA techniques
II.3.2. DNA separation by gel electrophoresis
II.3.3. Bacterial strains construction
II.3.3.1. Construction of YB886 isogenic strains carrying the yueB 6 deletion and yueB conditional mutants
II.3.3.2. Expression of yueB-gfp fusions
II.3.3.3. Expression of lacI-cfp fusions
II.3.4. Construction of bacteriophage SPP1delX110lacO64
II.4. Biochemical methods
II.4.1. Protein analysis by SDS-PAGE and Western blotting
II.4.2. Fractionation of B. subtilis extracts and Western blot of YueB engineered versions
II.4.3. SPP1 phages disruption
II.4.4. Chemical modification of bacteriophages
II.5. Fluorescence microscopy
II.5.1. Phage binding localization
II.5.2. Internal YueB localization
II.5.3. External YueB localization
II.5.4. Phage DNA detection
II.5.5. Image acquisition
II.6. Measurements of ion fluxes and determination of the membrane voltage
II.7. Electron microscopy
III.1. SPP1 – induced changes in the host cell CM
III.1.1. Optimization of the conditions for SPP1-induced CM depolarization
III.1.2. Comparison to other B. subtilis phages
III.1.3. CM depolarization caused by SPP1 infection requires the phage receptor YueB and depends on its abundance at the bacterial surface
III.2. Effects of Ca2+ ions on SPP1 infection
III.3. Analysis of SPP1 infection in space and time
III.3.1. Localization of SPP1 particles on the surface of B. subtilis
III.3.2. Cellular localization of SPP1 receptor YueB
III.3.3. Localization of SPP1 DNA in infected cells
III.4. Ca2+ ions effect on SPP1 DNA entry
IV.1. The interaction of SPP1 with YueB triggers a very fast depolarization of the B. subtilis CM.
VI.2. Effect of Ca2+ ions in SPP1 infection
VI.3. Spatio-temporal program of SPP1 entry in B. subtilis
VI.4. The working model