Crystal structures of eukaryotic translation initiation factor 5A from Methanococcus jannaschii at 1.8 A resolution

Abstract

Eukaryotic translation initiation factor 5A (eIF-5A) is a ubiquitous protein found in all eukaryotic cells. The protein is closely associated with cell proliferation in the G1-S stage of the cell cycle. Recent findings show that the eIF-5A proteins are highly expressed in tumor cells and act as a cofactor of the Rev protein in HIV-1-infected cells. The mature eIF is the only protein known to have the unusual amino acid hypusine, a post-translationally modified lysine. The crystal structure of eIF-5A from Methanococcus jannaschii (MJ eIF-5A) has been determined at 1.9 A and 1.8 A resolution in two crystal forms by using the multiple isomorphous replacement method and the multiwavelength anomalous diffraction method for the first crystal form and the molecular replacement method for the second crystal form. The structure consists of two folding domains, one of which is similar to the oligonucleotide-binding domain found in the prokaryotic cold shock protein and the translation initiation factor IF1 despite the absence of any significant sequence similarities. The 12 highly conserved amino acid residues found among eIF-5As include the hypusine site and form a long protruding loop at one end of the elongated molecule.

Figure 1

Typical regions of the solvent-flattened experimental electron density map.

Figure 2

(A) Topology diagram of the MJ eIF-5A structure. The arrows represent β-strands and the short cylinder represents a 3₁₀ helix. The lysine modification site is represented by a gray circle. (B) Ribbon diagram of MJ eIF-5A structure in C2 crystal form. The arrows represent β-strands. The secondary structures were assigned by the method of Kabsch and Sander (31). Two domains are colored magenta and blue and connected by a green linker. The side chain of Lys-40 is shown as a ball-and-stick model. This figure was made with molscript (32).

Figure 3

The α-carbon traces of domain I (thick line, residues 1–73) are overlapped with those of domain II (thin line, residues 74–132) in a stereo diagram. The β-strands β3, β4, and β5 of domain I are superimposed on β7, β8, and β9 of domain II, respectively. The first and the last residues in each domain are labeled.

Figure 4

Ribbon diagram showing the twofold dimer present in the P4₁22 crystal form (crystal form I). The arrows represent β-strands. The side chain of Lys-40 is shown as a ball-and-stick model. Two domains are colored magenta and blue with different shadings for each subunit. The two-fold symmetry axis is marked in the center of the dimer.

Figure 5

Stereo diagram of superimposed α-carbon traces of MJ eIF-5A structure in crystal form I (thick line) and those in crystal form II (thin line). The superposition is based on residues 5–73 of domain I.

Figure 6

Two different views of the surface charge distribution of MJ eIF-5A as calculated by program grasp (39). The red and blue colors represent negatively and positively charged surfaces, respectively. The lysine modification site (Lys-40) is labeled.