Case Study 2 Bacteriophage T4 Short Tail Fiber

Bacteriophage T4 acts like a self-powered nanoscopic syringe. It is also one of the largest and most complex viruses known and contains more than 40 different structural proteins (31). It is very efficient, since one particle is generally sufficient to productively infect its E. coli host and several hundred daughter phages can be produced in 30 min at 37°C. During infection, first the long tail fibers bind to the bacterial lipopolysaccharide or OmpC (outer membrane protein C). Once at least three long tail fibers have bound, the phage base-plate changes conformation from the hexagon form to the star form (32). This leads the short tail fibers to extend and bind to the bacterial lipopolysaccharide core region (Fig. 3). The next step is contraction of the outer tail tube, driving the inner tail tube through the bacterial membrane (Fig. 3). The inner tail tube is capped by a needle (33) with lysozyme activity, helping the puncturing of the bacterial cell wall. The phage DNA passes through the tail tube into the bacterium and directs the production of daughter phage.

The long tail fibers are a complex of four different proteins (34). The proximal half of the fiber (or "thigh") is formed by a gp34 trimer (1289 amino acids per subunit), the hinge (or "knee") contains a monomer of gp35 (372 amino acids), and the distal half of the fiber (or "shin") is made up of trimers of gp36 and gp37. gp37 (1026 residues per monomer) is responsible for receptor binding and makes up the bulk of the distal fiber half, which gp36 (221 amino acids per monomer) connects the gp37 trimer to the hinge. The crystal structures of the long tail fiber proteins have not yet been reported.

Short tail fibers are composed of a single protein, gp12, of 527 residues, which forms parallel homotrimers. Coexpression of gp12 with the T4 chaper-one protein gp57 (22) in E. coli and subsequent purification led to the availability of material of suitable quantity, purity, and solubility. However, despite extensive trials, crystals could not be obtained from this material. Like the adenovirus fiber, its N-terminus is flexible and/or unstable, and 37 to 41 amino acids can be digested away already at 37°C. This common pattern may be owing to the fact that the N-termini of both proteins are attached to the viral capsid or the base plate in their native context and, therefore, may become destabilized in the isolated, purified form. On further increase in the temperature at 56°C followed by protease digestion, the first 83 amino acids are

N-terminal Triple Collar Receptor-

repeat beta-helix binding domain

Fig. 3. Schematic drawing of uncontracted bacteriophage T4 (top left) and contracted T4 (top right). Only four of the six fibritin and long tail fibers are shown for clarity. Uncontracted T4 particles are about 300 nm high. At the bottom is a composite structure of the known domains of the bacteriophage T4 short tail fiber (combined from PDB codes 1H6W and 1OCY). One of the chains is shown in black and the other two in gray. The receptor-binding domain, collar domain, triple P-helix, and resolved N-terminal repeat are labeled. Five more N-terminal repeats are present in the full-length protein. In the schematics, the part of the short tail fiber of which the structure is known is boxed in gray. Full-length short tail fibers are about 35 nm long, and the part of the structure shown is about 15 nm.

N-terminal Triple Collar Receptor-

repeat beta-helix binding domain

Fig. 3. Schematic drawing of uncontracted bacteriophage T4 (top left) and contracted T4 (top right). Only four of the six fibritin and long tail fibers are shown for clarity. Uncontracted T4 particles are about 300 nm high. At the bottom is a composite structure of the known domains of the bacteriophage T4 short tail fiber (combined from PDB codes 1H6W and 1OCY). One of the chains is shown in black and the other two in gray. The receptor-binding domain, collar domain, triple P-helix, and resolved N-terminal repeat are labeled. Five more N-terminal repeats are present in the full-length protein. In the schematics, the part of the short tail fiber of which the structure is known is boxed in gray. Full-length short tail fibers are about 35 nm long, and the part of the structure shown is about 15 nm.

removed and a second cleavage site is uncovered after Arg-395. This heat- and protease-stable fragment crystallized successfully (12). Another fragment, prepared in the same way except in the presence of divalent zinc cations, led to a fragment in which only the N-terminal 83 amino acids are removed. The resulting crystal structures (15,16) allowed the identification of several new folds, although parts of the crystallized protein were not visible in the crystallographic electron density because of static disorder. This is true for amino acids 85 to

245 in the first fragment, and for residues 85 to 329 in the second. Nevertheless, the combined crystal structures allowed us to describe the structure of amino acids 246 to 527 (Fig. 3).

The N-terminal domain contains six 17-residue sequence repeats comprising conserved hydrophobic residues such as alanines and threonines, and conserved glutamic acid residues. Of these repeats, one has been resolved in the crystal structure (residues 255-271); it basically comprises two P strands connected by a solvent-exposed type 1 P turn. Eight of these sequence repeats are also present in gp34 and one in gp37. Because the residues between the repeats are not conserved and the spacing between the repeats is variable, it appears that the T4 fiber shaft structures are not as regular as the structure of the aden-ovirus fiber shaft.

The region comprising amino acids 290 to 327 forms a right-handed triple-stranded P-helix, in which each monomer contributes six, six-residue P strands in two 360° turns (Fig. 3). Two amino acids of each P strand contribute hydrophobic side chains to a central hydrophobic core, and the framework of the triple P-helix is held together by interchain hydrogen bonds between the P strands. In the P-helix domain, each monomer has 57% of its surface buried in the complex. Interestingly, the needle at the bottom of the tail tube, which is a trimer of gp5, also contains a triple P-helix, but with longer P strands of eight residues and comprising seven 360° turns (33).

The arrow-shaped C-terminal head domain has a globular collar domain (15) and a strongly intertwined zinc-containing receptor-binding domain (16). The collar domain has weak structural homology to a part of bacteriophage T4 gp11, suggesting a possible distant evolutionary relationship. The receptor-binding domain has yet another, unique fold. It contains few regular secondary structure elements, although the normal requirements for a stable protein fold are fulfilled: it has a hydrophobic core and hydrogen-bonding requirements are met. In the center of the trimeric receptor-binding domain, we found a single zinc ion, octahedrally coordinated by the NE2 atoms of two histidine residues from each monomer. In the absence of zinc ions, this domain unfolds at 56°C, whereas in the presence of zinc it does not.

Was this article helpful?

0 0

Post a comment