Authors' Summary

Research Articles

Authors' Summary:
GPCR Engineering Yields High-Resolution Structural Insights into β2-Adrenergic Receptor Function

Daniel M. Rosenbaum et al.

About This Summary

In selected Research Articles, Science is experimenting with author-written summaries designed to provide an accessible statement of the article's findings and significance. Do you find these summaries useful? Let us know your thoughts by filling out a brief online survey.

Heterotrimeric guanine nucleotide–binding protein (G protein) coupled receptors (GPCRs) are a remarkably versatile family of transmembrane signaling molecules that present extraordinary therapeutic potential as drug targets for a wide spectrum of diseases (1). A better understanding of GPCR structure will help to elucidate the mechanism by which they transmit signals across the plasma membrane and will facilitate the development of more effective and selective drugs. Rhodopsin, the light-sensitive photoreceptor found in the mammalian eye, is also a GPCR, and its structure has been determined. Obtaining high-resolution structures of GPCRs other than rhodopsin, however, has been challenging because of their low natural abundance and inherent structural flexibility and instability (2). As a result, the application of structure-based drug discovery to GPCRs has largely been limited to the use of rhodopsin-based homology models. We have applied targeted protein engineering to make other GPCRs better candidates for crystallographic structure analysis, using the human β2-adrenergic receptor as an initial candidate. The β2-adrenergic receptor binds the hormones adrenaline and noradrenaline to regulate cardiovascular and pulmonary function. Our results show that targeted protein engineering can be used to obtain high-resolution structures of GPCRs.

Efforts to crystallize the wild-type (WT) β2-adrenergic receptor have been unsuccessful. We identified the third intracellular loop (ICL3) as a possible contributor to the difficulties in crystallization. This region interacts with signaling molecules such as G proteins. ICL3 is poorly structured and contributes to the relatively unrestricted movement of the transmembrane helices, which leads to the receptor's conformational heterogeneity and crystallization problems. Truncation of ICL3 could reduce the movement of the helices, but this would also reduce the polar surface area that is important for forming crystal-lattice contacts. Thus, we replaced ICL3 with T4 lysozyme (T4L), a well-folded protein that adds a polar surface and simultaneously restricts the movement of transmembrane helices linked by the inserted sequence (see the figure). The resulting fusion protein could be efficiently expressed in insect cells and purified. It also retained ligand binding affinities near those of the WT receptor and could undergo conformational changes upon agonist binding.

Resize Image

Replacement of an intracellular loop of the β2-adrenergic receptor with lysozyme stabilized two flexible helices (5 and 6), allowing crystallization of the fusion protein and determination of the structure of this medically important membrane receptor.

The engineered protein formed crystals in two different lipid environments: bicelles and lipidic cubic phase. Crystals grown in lipidic cubic phase yielded a 2.4 Å x-ray diffraction data set from which we determined the structure by molecular replacement, using T4 lysozyme and a polyalanine model of the transmembrane regions of rhodopsin as search models [see (3)].

One must be concerned that the structural modifications that facilitated crystal formation could produce a high-resolution structure that does not reflect the WT β2-adrenergic receptor. The fusion protein exhibited slightly elevated agonist binding affinities; however, antagonist binding affinities were normal, and only minor differences were observed when the β2-adrenergic receptor–T4L structure was compared to a 3.4 Å structure of the WT β2-adrenergic receptor crystallized in complex with an antibody fragment (4). Therefore, with the exception of ICL3, the structure of β2-adrenergic receptor–T4L probably reflects that of the native receptor.

Although we know the structure of the GPCR rhodopsin, our structure of the β2-adrenergic receptor now provides a high-resolution view of a GPCR that binds diffusible hormones and neurotransmitters and should facilitate structure-based drug development. Moreover, it can help to reveal how structural changes are propagated from the agonist binding site to the G protein–coupling domains. The methods that we used to obtain these crystals will probably be applicable to other GPCRs, allowing the determination of their structures and increasing our understanding of transmembrane signal transduction and drug discovery for this important class of molecules.

Nevertheless, a full understanding of the structural basis of GPCR activation will require a high-resolution structure of a complex between a receptor with an agonist bound and its G protein, as well as methods to assess the dynamics of their interaction that cannot be captured by the static snapshots provided by x-ray crystallography. Future nuclear magnetic resonance (NMR) experiments, which capture some of the dynamics, could identify key GPCR residues: those critical for binding different classes of ligands, propagating conformational changes, and forming allosteric interaction sites for signaling partners such as G proteins and arrestins. When interpreted in the context of high-resolution crystal structures, NMR and other dynamics experiments will lead to a better understanding of GPCR signaling and how such signaling can be manipulated with small molecules to therapeutic effect.

Summary References

  1. K. L. Pierce, R. T. Premont, R. J. Lefkowitz, Nat. Rev. Mol. Cell Biol.3, 639 (2002).
  2. B. K. Kobilka, X. Deupi, Trends Pharmacol. Sci.28, 397 (2007).
  3. V. Cherezov et al., Science318, 1258, (2007); published online 25 October 2007 (10.1126/science.1150577).
  4. S. G. F. Rasmussen et al., Nature; published online 21 October 2007 (10.1038/nature06325).

Read the Full-Text Research Article