News & Views | Published:

Structural biology

Snapshot of a signalling complex

Nature volume 477, pages 540541 (29 September 2011) | Download Citation

G-protein-coupled receptors initiate signalling pathways by forming complexes with agonist molecules and G proteins. The first crystal structure of such a complex is both reassuring and provocative. See Article p.549 & Letter p.611

The most widely used signal-transduction mechanism in nature involves G-protein-coupled receptors (GPCRs) embedded in cell membranes. Understanding precisely how this process occurs remains one of the most important and challenging questions in biology, not least because GPCRs act as sensors for a multitude of signals — from hormones and nutrients to pheromones and even light. And they're not just of interest to academics: GPCRs are the largest class of drug targets, with literally hundreds of discrete receptors amenable to pharmacological manipulation.

The binding of an agonist — a receptor activator — at the extracellular face of a GPCR causes the exchange of one nucleotide (GDP) for another (GTP) at a G protein bound to the receptor on the other side of the plasma membrane. The GTP-occupied α-subunit of the G protein and/or the βγ-subunit then activates intracellular effector molecules. In a paper that will become an instant classic (page 549), Rasmussen et al.1 present the long-awaited crystal structure of a basic transmembrane signalling complex: the agonist-bound β2-adrenergic receptor (β2AR) coupled to its G-protein partner, Gs. And in an accompanying paper on page 611 from the same group, Chung et al.2 probe the protein dynamics of this signalling complex using a technique known as hydrogen–deuterium-exchange mass spectrometry (DXMS).

During the past two decades, a series of X-ray structures has provided great insight into the structural biology surrounding GPCR signalling. Of particular note are the recently reported structures showing the receptor part of the active signalling complex bound to small fragments of the G protein's α-subunit, or to antibodies that mimic the G protein3,4,5. In addition, essential dynamic properties of the complex have been uncovered by spectroscopic and biophysical studies, many in the transduction system associated with vision6,7. Importantly, however, Rasmussen et al.1 are the first to have crystallized the complete complex of an active receptor and its G-protein partner.

There is a direct line connecting the fundamental, 30-year-old concept of 'allosteric' coupling between receptor and G protein — and the corresponding effect of GDP on agonist binding8,9 — to Rasmussen and colleagues' crystallographic snapshot1 of the GPCR-mediated signal-transduction process. Making the connection involved hard work rather than good luck — much of it old-fashioned, cold-room biochemistry based on more than 20 years' worth of meticulously compiled tricks, skills and tools, not to mention considerable understanding of the β2AR system.

A complete review of the biochemical methods used to prepare the material that eventually yielded crystals for Rasmussen and co-workers' study would occupy an entire semester-long course in graduate school (and maybe it should). Selected highlights would include: the generation of a stable, nucleotide-free receptor–G-protein complex using an enzyme to remove GDP from the analogous GDP-containing complex; the development of antibodies from llamas that stabilize the active form of the G protein's α-subunit; and the use of single-particle electron microscopy to identify the most stable signalling complexes. These and similar approaches will certainly be useful in future work directed at improving the overall resolution of the structure of the agonist-bound β2AR–Gs complex — especially the receptor portion of the complex, which is currently visualized at relatively low resolution1 — and for trapping other important forms of the complex, such as those involved in receptor deactivation.

The structure of Rasmussen and colleagues' complex (Fig. 1) and its mechanistic implications are striking, and would not all have been predicted. Surprisingly, the β- and γ-subunits of Gs barely make any direct contact with the receptor, but do facilitate interaction between the helical amino-terminal segment of the α-subunit of Gs and an intracellular loop (known as intracellular loop 2) of the receptor. The carboxy-terminal tail of the α-subunit engages the receptor in a cleft generated by the large outward movement of TM6, one of the receptor's seven transmembrane segments. The role of the α-subunit's C terminus was, in fact, clear from other work, such as the previously reported structures4 of rhodopsin (a GPCR involved in vision) in complex with a fragment of its corresponding G protein. However, the orientation and binding mode of this end of the α-subunit are different in the old4 and new1 structures, showing that major conformational changes occur between the first encounter of the receptor with the GDP-occupied G protein and the formation of the nucleotide-free complex depicted by Rasmussen and colleagues' structure.

Figure 1: Activated receptor–G-protein complex.
Figure 1

Transmembrane G-protein-coupled receptors (GPCRs) form complexes with agonist molecules and G proteins to initiate a wide range of signalling pathways in cells. Rasmussen et al.1 report the first crystal structure of a complete, activated GPCR signalling complex: the agonist-bound β2-adrenergic receptor (β2AR) in complex with its G protein, Gs. In this schematic of the complex, α, β and γ are the three subunits of Gs. Also pictured are the protein T4-lysozyme (T4L) and a single-chain antibody (SCA), which were incorporated into the complex to facilitate its crystallization. The detergent micelle acts in lieu of a cell membrane and maintains the structure of the complex. The zigzag lines on α and γ are endogenous lipid modifications that anchor the G protein to the micelle. Broken arrows within the α-subunit indicate structural changes that lead to the separation of its two domains (solid double-headed arrow), which allows the G protein to exchange one nucleotide (GDP) for another (GTP; nucleotides not shown). The broken, double-headed arrow in β2AR indicates the allosteric mechanism that connects G-protein binding to high-affinity agonist binding. (Graphic designed by Karina Åberg.)

In contrast to Rasmussen et al., who had to make several structural changes to the β2AR–Gs complex to facilitate its crystallization, Chung et al.2 could perform their DXMS analysis on a mostly unmodified complex. Importantly, Chung and colleagues' study confirms most of the changes of conformation and interactions predicted from the X-ray structure1. However, their results suggest that the RLLL peptide motif in Gs (a motif found in all G-protein α-subunits) undergoes considerable changes to its structure and/or stability during formation of the β2AR–Gs complex. This is at odds with the X-ray structure, in which the motif seems to be firmly locked in the middle of a β-sheet. Because the motif forms part of a protein strand (a β-strand) that directly connects intracellular loop 2 of the agonist-bound receptor with the 'P-loop' of Gs that binds to the β-phosphate in GDP, Chung and colleagues' data2 provide a novel connection between receptor binding and nucleotide exchange that is not obvious from the X-ray structure1.

The authors foresee that simultaneously, or possibly in rapid succession, the N-terminal helix of the α-subunit of Gs destabilizes the adjacent β-strand, and that the α-5 helix connected to the bound C-terminal tail becomes disordered. Both of these changes are communicated to the GDP-binding pocket of Gs, where two large domains of the protein's α-subunit open like a clamshell. This confirms a prediction made 18 years ago10, on the basis of the first X-ray structures of a G-protein α-subunit, that an activated receptor must open up the cleft in the α-subunit to allow nucleotide exchange. However, it seems that the 'clamshell' in Rasmussen and colleagues' structure may have been caught in an extreme, open form in their crystals, as a recent spectroscopic analysis7 of the rhodopsin–G-protein complex demonstrated that a more subtle, but still clear, opening occurs during nucleotide exchange.

For the first time, we have a detailed three-dimensional structure of the large (1,300 Å2) interface between an activated GPCR and its G-protein partner. We can now breathe a sigh of relief. But there will be no rest for weary workers in the GPCR field, because although the structure of the β2AR–Gs complex is reassuring, it is also provocative. Structural connections in the β2AR–Gs interface seem to be responsible for the allosteric mechanism that links receptor coupling to nucleotide exchange on the G protein. However, the structural basis for the increase in agonist affinity that occurs on G-protein binding is still elusive. Moreover, we still have no clear picture of what determines G-protein selectivity — the last five amino acids of a G-protein α-subunit determine its receptor binding partner11, but Rasmussen and colleagues' β2AR–Gs structure1 does not tell us how this happens.

Structures of additional receptor–G-protein complexes are therefore required, as are complementary approaches of investigation, such as systems and molecular-dynamics simulations, and spectroscopic and other biophysical approaches6,7. And we have not even mentioned GPCR kinases and β-arrestins, the proteins that are central to receptor desensitization and signalling through other intracellular pathways12. But, for now, it's time to celebrate and enjoy a brief respite. Then it's back to the cold room.


  1. 1.

    et al. Nature 477, 549–555 (2011).

  2. 2.

    et al. Nature 477, 611–615 (2011).

  3. 3.

    et al. Nature 469, 175–180 (2011).

  4. 4.

    et al. Nature 471, 651–655 (2011).

  5. 5.

    et al. Nature 471, 656–660 (2011).

  6. 6.

    & Trends Pharmacol. Sci. 32, 410–419 (2011).

  7. 7.

    et al. Proc. Natl Acad. Sci. USA 108, 9420–9424 (2011).

  8. 8.

    Nature 284, 17–22 (1980).

  9. 9.

    , & J. Biol. Chem. 255, 7108–7117 (1980).

  10. 10.

    , & Nature 366, 654–663 (1993).

  11. 11.

    , , , & Nature 363, 274–276 (1993).

  12. 12.

    & Science 308, 512–517 (2005).

Download references

Author information


  1. Thue W. Schwartz is in the Laboratory for Molecular Pharmacology, Department of Neuroscience and Pharmacology, and at the NNF Center for Basic Metabolic Research, Faculty of Health Sciences, University of Copenhagen, DK-2200 Copenhagen, Denmark.

    • Thue W. Schwartz
  2. Thomas P. Sakmar is in the Laboratory of Molecular Biology and Biochemistry, The Rockefeller University, New York 10065, USA.

    • Thomas P. Sakmar


  1. Search for Thue W. Schwartz in:

  2. Search for Thomas P. Sakmar in:

Corresponding authors

Correspondence to Thue W. Schwartz or Thomas P. Sakmar.

About this article

Publication history



Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing