PDBsum entry: 1gp9

Hormone/growth factor

PDB id: 1gp9

Contents

Protein chains

170 a.a. *

Ligands

EPE ×5

Waters ×156

* Residue conservation analysis

PDB id:

1gp9

Name:

Hormone/growth factor

Title:

A new crystal form of the nk1 splice variant of hgf/sf demonstrates extensive hinge movement and suggests that the nk1 dimer originates by domain swapping

Structure:

Hepatocyte growth factor. Chain: a, b, c, d. Fragment: nk1, residues 40-210. Synonym: scatter factor, sf, hepatopoeitin a. Engineered: yes

Source:

Homo sapiens. Human. Organism_taxid: 9606. Cell: fibroblast. Expressed in: yeast. Expression_system_taxid: 4932. Expression_system_variant: gs115.

Biol. unit:

Dimer (from PDB file)

Resolution:

2.5Å R-factor: 0.237 R-free: 0.288

Authors:

K.Watanabe,D.Y.Chirgadze,D.Lietha,E.Gherardi,T.L.Blundell

Key ref:

K.Watanabe et al. (2002). A new crystal form of the NK1 splice variant of HGF/SF demonstrates extensive hinge movement and suggests that the NK1 dimer originates by domain swapping. J Mol Biol, 319, 283-288. PubMed id: 12051906 DOI: 10.1016/S0022-2836(02)00199-7

Date:

31-Oct-01 Release date: 19-Nov-01

Protein chains

P14210  (HGF_HUMAN) -  Hepatocyte growth factor

Seq: 728 a.a.

Struc: 170 a.a.*

* PDB and UniProt seqs differ at 1 residue position (black cross)

DOI no: 10.1016/S0022-2836(02)00199-7

J Mol Biol 319:283-288 (2002)

PubMed id: 12051906

A new crystal form of the NK1 splice variant of HGF/SF demonstrates extensive hinge movement and suggests that the NK1 dimer originates by domain swapping.

K.Watanabe, D.Y.Chirgadze, D.Lietha, H.de Jonge, T.L.Blundell, E.Gherardi.

ABSTRACT

NK1 is a splice variant of the polypeptide growth factor HGF/SF that consists of the N terminal (N) and first kringle (K) domains and retains receptor binding and signalling. While NK1 behaves as a monomer in solution, two independent crystallographic structures have previously shown an identical, tightly packed dimer. Here we report a novel orthorhombic crystal form of NK1 at 2.5 A resolution in which four NK1 protomers are packed in two distinct dimers in the asymmetric unit. Although the basic architecture of the new NK1 dimers is similar to the two described earlier, the new crystal form demonstrates extensive hinge movement between the N and K domain that leads to re-orientation of the receptor-binding sites. The hinge bending is evidence of the paucity of strong interactions between domains within the protomer, in contrast to the extensive interactions between protomers in the dimer. These observations are consistent with domain swapping in the dimer, such that the interdomain interactions of the monomer are replaced by equivalent interprotomer interactions in the dimer and offer a route for protein engineering of NK1 variants which may act as receptor antagonists.

Selected figure(s)

Figure 1.
Figure 1. Comparison of the 1gp9, 1nk1 and 1bht crystal structures. (a) The expression constructs used for the 1bht[19], 1nk1 [20] and 1gp9 structures. The sequence of human HGF/SF is shown at the top, leader sequences and/or sequences required for construct assembly are shown in red, arrows below the sequence show the cleavage sites for mammalian, bacterial or yeast peptidases. The sequences underlined in black are disordered in the three structures while the sequence underlined in blue corresponds to the first secondary structure element (the b1 strand) of the N domain. The bacterial construct employed by Ultsch et al.,[19] introduced a bacterial stII leader and a ten amino acid residue antibody tag before residue Q[32] of HGF/SF. The yeast construct employed by Chirgadze et al.,[20] introduced an a-pheromone leader and an extra YV sequence before E[30] of HGF/SF. In this construct (NK1[28-210]), the yeast leader was processed completely first by the KEX2 gene product (LEKR|EAEA) and subsequently by the STE13 gene product (EAEA|YVEG), leading to removal of the EAEA repeats. The lowest line shows the sequence of the construct that yielded the NK1 structure reported here (1gp9). (b) The two NK1 dimers present in the asymmetric cell of the 1gp9 structure. In each dimer one protomer (A or C) is represented as a ribbon diagram (N and K domains in pale green, linkers in red) while the second protomer (B or D) is shown as a space filling model (N and K domains in dark green, linkers in brown). (c) Superposition of the protomers from the 1bht, 1nk1 and 1gp9 structures. Values show rmsd of C^a atoms (Å).

Figure 2.
Figure 2. Hinge movement and mechanism of dimerization of NK1. (a) Orthogonal views of the K domains of protomer C from the 1gp9 structure and protomer A from the 1nk1 structure superposed in order to highlight the tilt and rotation (21.7°) of the N versus the K domain. In both panels the line shows the axis across the hinge. (b) Schematic representation of the hinge movement in the NK1 dimer. The dimer on the left corresponds to the one seen in the 1bht and 1nk1 structures and in the A-B dimer of the 1gp9 structure; the dimer on the right corresponds to the C-D dimer of the 1gp9 structure. (c) Proposed mechanism of NK1 dimerization via domain swapping. The drawings on the left and in the centre represent the conformation of the putative `closed' and `open' NK1 monomers in solution. The drawing on the right represents the NK1 dimer observed in the available crystal structures.

The above figures are reprinted by permission from Elsevier: J Mol Biol (2002, 319, 283-288) copyright 2002.