Structural Basis of Protein Protein Recognition

A specific interaction between a signaling and membrane receptor molecule is critical to obtain a well-directed signaling event. The molecular recognition process that underlies a specific interaction is provided by the complementarity of the physicochemical and geometrical properties of the two protein surfaces to obtain an energetically favorable complex. This is determined by the hydrophobic effect, close packing with favorable van der Waals interactions, and the formation of hydrogen and ionic bonds. Computational analyses of atomic structures of protein-protein complexes have identified the structural and physiochemical properties of these interfaces [3-5] (see Kleanthous [6] for reviews). Structures of various extracellular molecular signaling complexes have been elucidated so far: the extracellular domains of receptors complexed with hormones and cytokines, the major his-tocompatibility complex with diverse peptides (pMHC) in association with the T-cell receptor (TCR), and antibody (Fab fragments)-antigen complexes [6]. For the majority of these complexes, the individual components form homo- or hetero-oligomers by itself or upon ligand binding. Table I summarizes the receptor-ligand complexes and their oligomeric disposition [7].

In general, protein-protein interfaces exhibit a mixture of apolar and polar interactions scattered over the binding surface with the polar residues providing fine specificity. The interfaces of non-obligate complexes (i.e., between molecules that also exist on their own), such as extracellular signalling and enzyme-inhibitor complexes are generally more polar than homodimers (Fig. 2), because of the solubility

Table I Current Receptor-Protein Signaling Complexes in the Protein Data Bank

Proteinl

Protein2

pdbcode(s) (resolution in Â) Oligomeric State1

Receptor Activity

Ref.

Growth hormone receptor

Prolactin receptor

Erythropoietin receptor

Interleukin-1 receptor

Interleukin-1 receptor

Interleukin-4 receptor a chain

Granulocyte colony-stimulating factor receptor

Interleukin-6 receptor GP130 chain

Trka receptor Bone morphogenetic protein receptor 1a Interferon-gamma receptor a Interleukin-10

receptor Fibroblast growth factor receptor 1 Fibroblast growth factor receptor 1 Fibroblast growth factor receptor 2 Fibroblast growth factor receptor 2

Death receptor 5

Tumor necrosis factor receptor

Somatotropin

(growth hormone) Gh antagonist g130r

Somatotropin

(growth hormone) Placental lactogen Erythropoietin

Interleukin-1

receptor antagonist

Interleukin-1 beta

Interleukin-4

G-csf

Interleukin-6

Nerve growth factor Bone morphogenetic protein-2

Interferon-y

Interleukin-10

Fibroblast growth factor-1 Fibroblast growth factor-2 Fibroblast growth factor-1 Fibroblast growth factor-2 Fibroblast growth factor-2 apert syndrome variant TRAIL

Tumor necrosis factor ß

1hwh (2.9)3, 1a22 (2.6) remodeled interface: 1axi (2.1)3

4:42

Receptor dimerization upon ligand binding

Homo-and heterodimerization, interdomain ligand binding; heparin involved

Non-protein kinase, associated jak kinases

Non-protein kinase, associated JAK kinases

Non-protein kinase, associated JAK kinases Non-protein kinase, associated accessory protein

Non-protein kinase, associated accessory protein Non-protein kinase, associates with common Y chain, associated JAK kinase Non-protein kinase, associated JAK kinases

Non-protein kinase, common ß chain (e.g., Gp130), associated JAK kinase Tyrosine kinase Serine-threonine kinase

Non-protein kinase, associated JAK kinases Non-protein kinase, associated JAK kinases Tyrosine kinase dimerizes without ligand;

trimerizes upon ligand binding

Non-protein kinase, associated TRAF

Non-protein kinase, associated TRAF

18 19

2G 21

1There are differences in the literature about the nomenclature used to describe receptor-ligand complexes. Here, we use the stoichiometry of the complex (i.e., the number of protomer chains of the receptor and ligand, respectively, involved in the complex). We identify two types of 2:2 complexes: the 2:2-I type, where each receptor chain contacts both monomeric ligands, and the 2:2-II type, where each receptor chain contacts both protomers of the dimeric ligand.

2The 2:2 complex is thought to form an intermediate receptor-ligand complex, whereas the 4:4 is the active receptor-ligand complex. Structural parameters (Fig. 2) have been computed for the former.

3These entries have not been included in the computational analysis shown in Fig. 2.

Figure 2 Correlation between the contact area of the interface and the percentage of the contact area that involves polar atoms for extracellular signaling complexes, diverse other nonobligate complexes, and homodimers. Structures used in this analysis are those studied previously [3,5] and more recent solved structures. Parameters have been calculated as described in Jones et al. [27]. When one or both of the proteins involved in the complex are multimers, the contact area is summed over all receptor-lig-and interfaces to give a total for the complete assembly and the percentage polarity has been averaged respectively.

Figure 2 Correlation between the contact area of the interface and the percentage of the contact area that involves polar atoms for extracellular signaling complexes, diverse other nonobligate complexes, and homodimers. Structures used in this analysis are those studied previously [3,5] and more recent solved structures. Parameters have been calculated as described in Jones et al. [27]. When one or both of the proteins involved in the complex are multimers, the contact area is summed over all receptor-lig-and interfaces to give a total for the complete assembly and the percentage polarity has been averaged respectively.

requirements of the individual molecules. Whereas the percentage of polar atoms in the interface is variable in the receptor-ligand and pMHC-TCR complexes, the antibody-antigen complexes consistently have more than 40% polar atoms in the interface and have a relatively small contact area (i.e., interface smaller than 1500 A2). The surface area buried in the specific non-obligate protein-protein associations is also highly variable. Large contact areas up to 5000 A2 are found for homodimers and various nonobligate complexes, such as multimeric receptor-ligand and large enzyme-inhibitor complexes (Fig. 2).

Structural rearrangements upon protein-protein association have been identified for many complexes, such as enzyme-inhibitor (e.g., thrombin-hirudin), intracellular signalling (e.g., Ga-Gpy protein) and receptor-ligand (e.g., receptor-human growth factor) complexes. Remarkably, protein-protein complexes that undergo structural rearrangements usually have large interfaces (i.e., >1500 A2). They involve disorder-to-order transitions, small changes in side-chain conformations (i.e., translational, rotational, and side-chain degrees of freedom), or gross conformational changes such as loop or domain movements. The monomelic human growth factor, for example, shows large helix movements upon binding into a cleft formed by the two subunits of the homodimeric receptor. Conformational changes are expected to play a major role in the transmembrane signaling process. Upon receptor-ligand complexation, side-chain flexibility may facilitate finding the complementary fit [8], whereas larger conformational changes may reveal hydropho-bic surfaces or propagate a long-range structural rearrangement of the monomeric or oligomeric transmembrane receptor required for signal transduction. Residue spacing and molecular flexibility have been demonstrated to be important for protein-carbohydrate recognition in signalling, as well [9].

The structural basis of conformational flexibility is difficult to assess experimentally, as the current available experimental methods in structural determination (i.e., X-ray crystallography, nuclear magnetic resonance spectroscopy, electron microscopy) require a stable structure. Capturing the structures under different conditions (e.g., free and bound) or having thermodynamic and mutagenesis data can help to identify these changes or relate them to signal trans-duction activity. The current structural data in the Protein Data Bank (PDB) and a computational analysis of these protein-protein complexes in the PDB (Fig. 2) are probably biased toward structures that form stable structures. The atomic structures of many transmembrane domains or proteins have yet to be determined, which leaves the molecular mechanisms responsible for the signal transduction across the membrane still largely unknown.

Diabetes Sustenance

Diabetes Sustenance

Get All The Support And Guidance You Need To Be A Success At Dealing With Diabetes The Healthy Way. This Book Is One Of The Most Valuable Resources In The World When It Comes To Learning How Nutritional Supplements Can Control Sugar Levels.

Get My Free Ebook


Post a comment